To see the other types of publications on this topic, follow the link: Layer-wise Relevance Propagation (LRP).
Journal articles on the topic 'Layer-wise Relevance Propagation (LRP)'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 journal articles for your research on the topic 'Layer-wise Relevance Propagation (LRP).'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse journal articles on a wide variety of disciplines and organise your bibliography correctly.
1
Ullah, Ihsan, Andre Rios, Vaibhav Gala, and Susan Mckeever. "Explaining Deep Learning Models for Tabular Data Using Layer-Wise Relevance Propagation." Applied Sciences 12, no. 1 (2021): 136. http://dx.doi.org/10.3390/app12010136.
Full textAbstract:
Trust and credibility in machine learning models are bolstered by the ability of a model to explain its decisions. While explainability of deep learning models is a well-known challenge, a further challenge is clarity of the explanation itself for relevant stakeholders of the model. Layer-wise Relevance Propagation (LRP), an established explainability technique developed for deep models in computer vision, provides intuitive human-readable heat maps of input images. We present the novel application of LRP with tabular datasets containing mixed data (categorical and numerical) using a deep neural network (1D-CNN), for Credit Card Fraud detection and Telecom Customer Churn prediction use cases. We show how LRP is more effective than traditional explainability concepts of Local Interpretable Model-agnostic Explanations (LIME) and Shapley Additive Explanations (SHAP) for explainability. This effectiveness is both local to a sample level and holistic over the whole testing set. We also discuss the significant computational time advantage of LRP (1–2 s) over LIME (22 s) and SHAP (108 s) on the same laptop, and thus its potential for real time application scenarios. In addition, our validation of LRP has highlighted features for enhancing model performance, thus opening up a new area of research of using XAI as an approach for feature subset selection.
2
HSU, SHENG-YI, MAU-HSIANG SHIH, WU-HSIUNG WU, HAO-REN YAO, and FENG-SHENG TSAI. "Gene reduction for cancer detection using layer-wise relevance propagation." Journal of Decision Making and Healthcare 1, no. 1 (2024): 30–44. http://dx.doi.org/10.69829/jdmh-024-0101-ta03.
Full textAbstract:
Precise detection of cancer types and normal tissues is crucial for cancer diagnosis. Specifically, cancer classification using gene expression data is key to identify genes whose expression patterns are tumor-specific. Here we aim to search for a minimal set of genes that may reduce the expression complexity and retain a qualified classification accuracy accordingly. We applied neural network models with layer-wise relevance propagation (LRP) to find genes that significantly contribute to classification. Two algorithms for the LRP-candidate gene selection and the cycle of gene reduction were proposed. By implementing the two algorithms for gene reduction, our model retained 95.32% validation accuracy to make classification of six cancer types and normal with a minimal set of seven genes. Furthermore, a cross-evaluation process was performed on the minimal set of seven genes, indicating that the selected marker genes in five out of six cancer types are biologically relevant to cancer annotated by the COSMIC Cancer Gene Census.
3
Li, Y. Y., S. Y. Huang, S. B. Xu, et al. "Selection of the Main Control Parameters for the Dst Index Prediction Model Based on a Layer-wise Relevance Propagation Method." Astrophysical Journal Supplement Series 260, no. 1 (2022): 6. http://dx.doi.org/10.3847/1538-4365/ac616c.
Full textAbstract:
Abstract The prediction of the Dst index is an important subject in space weather. It has significant progress with the prevalent applications of neural networks. The selection of input parameters is critical for the prediction model of the Dst index or other space-weather models. In this study, we perform a layer-wise relevance propagation (LRP) method to select the main parameters for the prediction of the Dst index and understand the physical interpretability of neural networks for the first time. Taking an hourly Dst index and 10 types of solar wind parameters as the inputs, we utilize a long short-term memory network to predict the Dst index and present the LRP method to analyze the dependence of the Dst index on these parameters. LRP defines the relevance score for each input, and a higher relevance score indicates that the corresponding input parameter contributes more to the output. The results show that Dst, E y , B z , and V are the main control parameters for Dst index prediction. In order to verify the LRP method, we design two more supplementary experiments for further confirmation. These results confirm that the LRP method can reduce the initial dimension of neural network input at the cost of minimum information loss and contribute to the understanding of physical processes in space weather.
4
Neha Ahlawat. "Multimodal Deep Belief Network with Layer-Wise Relevance Propagation: A Solution for Heterogeneous Image Challenges in Big Data." Journal of Information Systems Engineering and Management 10, no. 22s (2025): 736–41. https://doi.org/10.52783/jisem.v10i22s.3616.
Full textAbstract:
Cancer is a complex and heterogeneous disease, with diverse molecular profiles and clinical outcomes. Accurate cancer classification is crucial for personalized treatment strategies and improved patient survival. The advent of high-throughput technologies has generated vast amounts of multi-dimensional data, including genomic, proteomic, and clinical information. Analyzing this "big data" requires sophisticated computational methods. This paper presents an improvised approach for Layer-wise Relevance Propagation (LRP) in Multimodal Deep Belief Networks (MDBNs) for cancer classification. By integrating Clipped Activation and Contrastive Divergence (CD), we enhance model interpretability and performance, addressing challenges like vanishing gradients and slow convergence. Our approach improves the efficiency of LRP while ensuring stable training and faster model convergence. Experiments on multimodal medical data, including brain, breast, and bone scans, demonstrate significant gains in classification accuracy and interpretability compared to traditional methods, offering a scalable solution for deep learning in healthcare.
5
Ado, Abubakar, Olalekan J. Awujoola, Sabiu Danlami Abdullahi, and Sulaiman Hashim Ibrahim. "INTEGRATION OF LAYER-WISE RELEVANCE PROPAGATION, RECURSIVE DATA PRUNING, AND CONVOLUTIONAL NEURAL NETWORKS FOR IMPROVED TEXT CLASSIFICATION." FUDMA JOURNAL OF SCIENCES 9, no. 2 (2025): 35–41. https://doi.org/10.33003/fjs-2025-0902-3058.
Full textAbstract:
This research presents a significant advancement in text classification by integrating Layer-wise Relevance Propagation (LRP), recursive data pruning, and Convolutional Neural Networks (CNNs) with cross-validation. The study addresses the critical limitations of existing text classification methods, particularly issues of information loss and overfitting, which often hinder the efficiency and interpretability of models in natural language processing (NLP). To overcome these challenges, the proposed model employs LRP to enhance the interpretability of the classification process, allowing for precise identification of relevant features that contribute to decision-making. Additionally, the implementation of recursive data pruning optimizes model efficiency by dynamically eliminating irrelevant or redundant data, thereby reducing computational complexity without compromising performance. The effectiveness of the approach is further bolstered by utilizing cross-validation techniques to ensure robust evaluation across diverse datasets. The empirical evaluation of the integrated model revealed remarkable improvements in classification performance, achieving an accuracy of 94%, surpassing the benchmark of 92.88% established by the ReDP-CNN model proposed by Li et al. (2020). The comprehensive assessment included detailed metrics such as precision, recall, and F1-score, confirming the model's robust capability in accurately classifying text data across various categories.
6
Lee, Jae-Eung, and Ji-Hyeong Han. "Layer-wise Relevance Propagation (LRP) Based Technical and Macroeconomic Indicator Impact Analysis for an Explainable Deep Learning Model to Predict an Increase and Decrease in KOSPI." Journal of KIISE 48, no. 12 (2021): 1289–97. http://dx.doi.org/10.5626/jok.2021.48.12.1289.
Full text
7
Du, Meng, Daping Bi, Mingyang Du, Xinsong Xu, and Zilong Wu. "ULAN: A Universal Local Adversarial Network for SAR Target Recognition Based on Layer-Wise Relevance Propagation." Remote Sensing 15, no. 1 (2022): 21. http://dx.doi.org/10.3390/rs15010021.
Full textAbstract:
Recent studies have proven that synthetic aperture radar (SAR) automatic target recognition (ATR) models based on deep neural networks (DNN) are vulnerable to adversarial examples. However, existing attacks easily fail in the case where adversarial perturbations cannot be fully fed to victim models. We call this situation perturbation offset. Moreover, since background clutter takes up most of the area in SAR images and has low relevance to recognition results, fooling models with global perturbations is quite inefficient. This paper proposes a semi-white-box attack network called Universal Local Adversarial Network (ULAN) to generate universal adversarial perturbations (UAP) for the target regions of SAR images. In the proposed method, we calculate the model’s attention heatmaps through layer-wise relevance propagation (LRP), which is used to locate the target regions of SAR images that have high relevance to recognition results. In particular, we utilize a generator based on U-Net to learn the mapping from noise to UAPs and craft adversarial examples by adding the generated local perturbations to target regions. Experiments indicate that the proposed method effectively prevents perturbation offset and achieves comparable attack performance to conventional global UAPs by perturbing only a quarter or less of SAR image areas.
8
Nazari, Mahmood, Andreas Kluge, Ivayla Apostolova, et al. "Explainable AI to improve acceptance of convolutional neural networks for automatic classification of dopamine transporter SPECT in the diagnosis of clinically uncertain parkinsonian syndromes." European Journal of Nuclear Medicine and Molecular Imaging 49, no. 4 (2021): 1176–86. http://dx.doi.org/10.1007/s00259-021-05569-9.
Full textAbstract:
Abstract Purpose Deep convolutional neural networks (CNN) provide high accuracy for automatic classification of dopamine transporter (DAT) SPECT images. However, CNN are inherently black-box in nature lacking any kind of explanation for their decisions. This limits their acceptance for clinical use. This study tested layer-wise relevance propagation (LRP) to explain CNN-based classification of DAT-SPECT in patients with clinically uncertain parkinsonian syndromes. Methods The study retrospectively included 1296 clinical DAT-SPECT with visual binary interpretation as “normal” or “reduced” by two experienced readers as standard-of-truth. A custom-made CNN was trained with 1008 randomly selected DAT-SPECT. The remaining 288 DAT-SPECT were used to assess classification performance of the CNN and to test LRP for explanation of the CNN-based classification. Results Overall accuracy, sensitivity, and specificity of the CNN were 95.8%, 92.8%, and 98.7%, respectively. LRP provided relevance maps that were easy to interpret in each individual DAT-SPECT. In particular, the putamen in the hemisphere most affected by nigrostriatal degeneration was the most relevant brain region for CNN-based classification in all reduced DAT-SPECT. Some misclassified DAT-SPECT showed an “inconsistent” relevance map more typical for the true class label. Conclusion LRP is useful to provide explanation of CNN-based decisions in individual DAT-SPECT and, therefore, can be recommended to support CNN-based classification of DAT-SPECT in clinical routine. Total computation time of 3 s is compatible with busy clinical workflow. The utility of “inconsistent” relevance maps to identify misclassified cases requires further investigation.
9
Zang, Bo, Linlin Ding, Zhenpeng Feng, et al. "CNN-LRP: Understanding Convolutional Neural Networks Performance for Target Recognition in SAR Images." Sensors 21, no. 13 (2021): 4536. http://dx.doi.org/10.3390/s21134536.
Full textAbstract:
Target recognition is one of the most challenging tasks in synthetic aperture radar (SAR) image processing since it is highly affected by a series of pre-processing techniques which usually require sophisticated manipulation for different data and consume huge calculation resources. To alleviate this limitation, numerous deep-learning based target recognition methods are proposed, particularly combined with convolutional neural network (CNN) due to its strong capability of data abstraction and end-to-end structure. In this case, although complex pre-processing can be avoided, the inner mechanism of CNN is still unclear. Such a “black box” only tells a result but not what CNN learned from the input data, thus it is difficult for researchers to further analyze the causes of errors. Layer-wise relevance propagation (LRP) is a prevalent pixel-level rearrangement algorithm to visualize neural networks’ inner mechanism. LRP is usually applied in sparse auto-encoder with only fully-connected layers rather than CNN, but such network structure usually obtains much lower recognition accuracy than CNN. In this paper, we propose a novel LRP algorithm particularly designed for understanding CNN’s performance on SAR image target recognition. We provide a concise form of the correlation between output of a layer and weights of the next layer in CNNs. The proposed method can provide positive and negative contributions in input SAR images for CNN’s classification, viewed as a clear visual understanding of CNN’s recognition mechanism. Numerous experimental results demonstrate the proposed method outperforms common LRP.
10
Wang, He-Sheng, Dah-Jing Jwo, and Zhi-Hang Gao. "Towards Explainable Artificial Intelligence for GNSS Multipath LSTM Training Models." Sensors 25, no. 3 (2025): 978. https://doi.org/10.3390/s25030978.
Full textAbstract:
This paper addresses the critical challenge of understanding and interpreting deep learning models in Global Navigation Satellite System (GNSS) applications, specifically focusing on multipath effect detection and analysis. As GNSS systems become increasingly reliant on deep learning for signal processing, the lack of model interpretability poses significant risks for safety-critical applications. We propose a novel approach combining Recurrent Neural Networks (RNNs) with Long Short-Term Memory (LSTM) cells with Layer-wise Relevance Propagation (LRP) to create an explainable framework for multipath detection. Our key contributions include: (1) the development of an interpretable LSTM architecture for processing GNSS observables, including multipath variables, carrier-to-noise ratios, and satellite elevation angles; (2) the adaptation of the LRP technique for GNSS signal analysis, enabling attribution of model decisions to specific input features; and (3) the discovery of a correlation between LRP relevance scores and signal anomalies, leading to a new method for anomaly detection. Through systematic experimental validation, we demonstrate that our LSTM model achieves high prediction accuracy across all GNSS parameters while maintaining interpretability. A significant finding emerges from our controlled experiments: LRP relevance scores consistently increase during anomalous signal conditions, with growth rates varying from 7.34% to 32.48% depending on the feature type. In our validation experiments, we systematically introduced signal anomalies in specific time segments of the data sequence and observed corresponding increases in LRP scores: multipath parameters showed increases of 7.34–8.81%, carrier-to-noise ratios exhibited changes of 12.50–32.48%, and elevation angle parameters increased by 16.10%. These results demonstrate the potential of LRP-based analysis for enhancing GNSS signal quality monitoring and integrity assessment. Our approach not only improves the interpretability of deep learning models in GNSS applications but also provides a practical framework for detecting and analyzing signal anomalies, contributing to the development of more reliable and trustworthy navigation systems.
11
Qiu, Changqing, Fusheng Jin, and Yining Zhang. "Empowering CAM-Based Methods with Capability to Generate Fine-Grained and High-Faithfulness Explanations." Proceedings of the AAAI Conference on Artificial Intelligence 38, no. 5 (2024): 4587–95. http://dx.doi.org/10.1609/aaai.v38i5.28258.
Full textAbstract:
Recently, the explanation of neural network models has garnered considerable research attention. In computer vision, CAM (Class Activation Map)-based methods and LRP (Layer-wise Relevance Propagation) method are two common explanation methods. However, since most CAM-based methods can only generate global weights, they can only generate coarse-grained explanations at a deep layer. LRP and its variants, on the other hand, can generate fine-grained explanations. But the faithfulness of the explanations is too low. To address these challenges, in this paper, we propose FG-CAM (Fine-Grained CAM), which extends CAM-based methods to enable generating fine-grained and high-faithfulness explanations. FG-CAM uses the relationship between two adjacent layers of feature maps with resolution differences to gradually increase the explanation resolution, while finding the contributing pixels and filtering out the pixels that do not contribute. Our method not only solves the shortcoming of CAM-based methods without changing their characteristics, but also generates fine-grained explanations that have higher faithfulness than LRP and its variants. We also present FG-CAM with denoising, which is a variant of FG-CAM and is able to generate less noisy explanations with almost no change in explanation faithfulness. Experimental results show that the performance of FG-CAM is almost unaffected by the explanation resolution. FG-CAM outperforms existing CAM-based methods significantly in both shallow and intermediate layers, and outperforms LRP and its variants significantly in the input layer. Our code is available at https://github.com/dongmo-qcq/FG-CAM.
12
Aeles, Jeroen, Fabian Horst, Sebastian Lapuschkin, Lilian Lacourpaille, and François Hug. "Revealing the unique features of each individual's muscle activation signatures." Journal of The Royal Society Interface 18, no. 174 (2021): 20200770. http://dx.doi.org/10.1098/rsif.2020.0770.
Full textAbstract:
There is growing evidence that each individual has unique movement patterns, or signatures. The exact origin of these movement signatures, however, remains unknown. We developed an approach that can identify individual muscle activation signatures during two locomotor tasks (walking and pedalling). A linear support vector machine was used to classify 78 participants based on their electromyographic (EMG) patterns measured on eight lower limb muscles. To provide insight into decision-making by the machine learning classification model, a layer-wise relevance propagation (LRP) approach was implemented. This enabled the model predictions to be decomposed into relevance scores for each individual input value. In other words, it provided information regarding which features of the time-varying EMG profiles were unique to each individual. Through extensive testing, we have shown that the LRP results, and by extent the activation signatures, are highly consistent between conditions and across days. In addition, they are minimally influenced by the dataset used to train the model. Additionally, we proposed a method for visualizing each individual's muscle activation signature, which has several potential clinical and scientific applications. This is the first study to provide conclusive evidence of the existence of individual muscle activation signatures.
13
Colin, Jovito, and Nico Surantha. "Interpretable Deep Learning for Pneumonia Detection Using Chest X-Ray Images." Information 16, no. 1 (2025): 53. https://doi.org/10.3390/info16010053.
Full textAbstract:
Pneumonia remains a global health issue, creating the need for accurate detection methods for effective treatment. Deep learning models like ResNet50 show promise in detecting pneumonia from chest X-rays; however, their black-box nature limits the transparency, which fails to meet that needed for clinical trust. This study aims to improve model interpretability by comparing four interpretability techniques, which are Layer-wise Relevance Propagation (LRP), Adversarial Training, Class Activation Maps (CAMs), and the Spatial Attention Mechanism, and determining which fits best the model, enhancing its transparency with minimal impact on its performance. Each technique was evaluated for its impact on the accuracy, sensitivity, specificity, AUC-ROC, Mean Relevance Score (MRS), and a calculated trade-off score that balances interpretability and performance. The results indicate that LRP was the most effective in enhancing interpretability, achieving high scores across all metrics without sacrificing diagnostic accuracy. The model achieved 0.91 accuracy and 0.85 interpretability (MRS), demonstrating its potential for clinical integration. In contrast, Adversarial Training, CAMs, and the Spatial Attention Mechanism showed trade-offs between interpretability and performance, each highlighting unique image features but with some impact on specificity and accuracy.
14
Sandeep Sharma. "Graph LRP and Domain Adversarial Neural Networks: An Approach to evaluate the performance and Design an Iterative method for Lumpy skin disease prediction." Journal of Information Systems Engineering and Management 10, no. 5s (2025): 558–71. https://doi.org/10.52783/jisem.v10i5s.684.
Full textAbstract:
The rising incidence of Lumpy Skin Disease in livestock is a threat to animal health and agricultural economies worldwide. Traditional approaches towards the prediction of this disease often fall short of expectations in explain ability, cross-domain adaptability, and real-time responsiveness for effective intervention. The authors mentioned the limitations by introducing a comprehensive machine learning framework LSD prediction. Specifically, graph neural networks combined with layer-wise relevance propagation, Graph LRP, to make the model more interpretable and transparent about the model's decision process. Graph LRP assigns a relevance score for input features. DANN generalizes it better and reduces labeled data requirements. Afterwards, to ensure adaptability in real-time, TinyML-based approach is followed by using the lightweight MobileNetV3 with GCNs for edge device deployment. This will enable low latency and efficient predictions to be made using continuous IoT sensor data streams and satellite images. Our proposed model improves the aspect of interpretability by 85% in relevance precision and cross-domain accuracy up to 95%, real-time inference performance at a 70-80% reduction in inference time. These make for a robust, scalable solution for the early detection and intervention of diseases, hence conferring benefits of magnitude in LSD management and control across different environments.
15
Seibold, Clemens, Anna Hilsmann, and Peter Eisert. "Feature Focus: Towards Explainable and Transparent Deep Face Morphing Attack Detectors." Computers 10, no. 9 (2021): 117. http://dx.doi.org/10.3390/computers10090117.
Full textAbstract:
Detecting morphed face images has become an important task to maintain the trust in automated verification systems based on facial images, e.g., at automated border control gates. Deep Neural Network (DNN)-based detectors have shown remarkable results, but without further investigations their decision-making process is not transparent. In contrast to approaches based on hand-crafted features, DNNs have to be analyzed in complex experiments to know which characteristics or structures are generally used to distinguish between morphed and genuine face images or considered for an individual morphed face image. In this paper, we present Feature Focus, a new transparent face morphing detector based on a modified VGG-A architecture and an additional feature shaping loss function, as well as Focused Layer-wise Relevance Propagation (FLRP), an extension of LRP. FLRP in combination with the Feature Focus detector forms a reliable and accurate explainability component. We study the advantages of the new detector compared to other DNN-based approaches and evaluate LRP and FLRP regarding their suitability for highlighting traces of image manipulation from face morphing. To this end, we use partial morphs which contain morphing artifacts in predefined areas only and analyze how much of the overall relevance each method assigns to these areas.
16
Slijepcevic, Djordje, Fabian Horst, Sebastian Lapuschkin, et al. "Explaining Machine Learning Models for Clinical Gait Analysis." ACM Transactions on Computing for Healthcare 3, no. 2 (2022): 1–27. http://dx.doi.org/10.1145/3474121.
Full textAbstract:
Machine Learning (ML) is increasingly used to support decision-making in the healthcare sector. While ML approaches provide promising results with regard to their classification performance, most share a central limitation, their black-box character. This article investigates the usefulness of Explainable Artificial Intelligence (XAI) methods to increase transparency in automated clinical gait classification based on time series. For this purpose, predictions of state-of-the-art classification methods are explained with a XAI method called Layer-wise Relevance Propagation (LRP). Our main contribution is an approach that explains class-specific characteristics learned by ML models that are trained for gait classification. We investigate several gait classification tasks and employ different classification methods, i.e., Convolutional Neural Network, Support Vector Machine, and Multi-layer Perceptron. We propose to evaluate the obtained explanations with two complementary approaches: a statistical analysis of the underlying data using Statistical Parametric Mapping and a qualitative evaluation by two clinical experts. A gait dataset comprising ground reaction force measurements from 132 patients with different lower-body gait disorders and 62 healthy controls is utilized. Our experiments show that explanations obtained by LRP exhibit promising statistical properties concerning inter-class discriminativity and are also in line with clinically relevant biomechanical gait characteristics.
17
Wan, Xuanshen, Wei Liu, Chaoyang Niu, and Wanjie Lu. "Attention Heat Map-Based Black-Box Local Adversarial Attack for Synthetic Aperture Radar Target Recognition." Photogrammetric Engineering & Remote Sensing 90, no. 10 (2024): 601–9. http://dx.doi.org/10.14358/pers.24-00015r2.
Full textAbstract:
Synthetic aperture radar (SAR) automatic target recognition (ATR) models based on deep neural networks (DNNs) are susceptible to adversarial attacks. In this study, we proposed an SAR black-box local adversarial attack algorithm named attention heat map- based black-box local adversarial attack (AH-BLAA). First, we designed an attention heat map extraction module combined with the layer-wise relevance propagation (LRP) algorithm to obtain the high concerning areas of the SAR-ATR models. Then, to gener- ate SAR adversarial attack examples, we designed a perturbation generator module, introducing the structural dissimilarity (DSSIM) metric in the loss function to limit image distortion and the dif- ferential evolution (DE) algorithm to search for optimal perturba- tions. Experimental results on the MSTAR and FUSAR-Ship datasets showed that compared with existing adversarial attack algorithms, the attack success rate of the AH-BLAA algorithm increased by 0.63% to 33.59% and 1.05% to 17.65%, respectively. Moreover, the low- est perturbation ratios reached 0.23% and 0.13%, respectively.
18
Gupta, Siddharth, Arun K. Dubey, Rajesh Singh, et al. "Four Transformer-Based Deep Learning Classifiers Embedded with an Attention U-Net-Based Lung Segmenter and Layer-Wise Relevance Propagation-Based Heatmaps for COVID-19 X-ray Scans." Diagnostics 14, no. 14 (2024): 1534. http://dx.doi.org/10.3390/diagnostics14141534.
Full textAbstract:
Background: Diagnosing lung diseases accurately is crucial for proper treatment. Convolutional neural networks (CNNs) have advanced medical image processing, but challenges remain in their accurate explainability and reliability. This study combines U-Net with attention and Vision Transformers (ViTs) to enhance lung disease segmentation and classification. We hypothesize that Attention U-Net will enhance segmentation accuracy and that ViTs will improve classification performance. The explainability methodologies will shed light on model decision-making processes, aiding in clinical acceptance. Methodology: A comparative approach was used to evaluate deep learning models for segmenting and classifying lung illnesses using chest X-rays. The Attention U-Net model is used for segmentation, and architectures consisting of four CNNs and four ViTs were investigated for classification. Methods like Gradient-weighted Class Activation Mapping plus plus (Grad-CAM++) and Layer-wise Relevance Propagation (LRP) provide explainability by identifying crucial areas influencing model decisions. Results: The results support the conclusion that ViTs are outstanding in identifying lung disorders. Attention U-Net obtained a Dice Coefficient of 98.54% and a Jaccard Index of 97.12%. ViTs outperformed CNNs in classification tasks by 9.26%, reaching an accuracy of 98.52% with MobileViT. An 8.3% increase in accuracy was seen while moving from raw data classification to segmented image classification. Techniques like Grad-CAM++ and LRP provided insights into the decision-making processes of the models. Conclusions: This study highlights the benefits of integrating Attention U-Net and ViTs for analyzing lung diseases, demonstrating their importance in clinical settings. Emphasizing explainability clarifies deep learning processes, enhancing confidence in AI solutions and perhaps enhancing clinical acceptance for improved healthcare results.
19
Dong, Sunghee, Yan Jin, SuJin Bak, Bumchul Yoon, and Jichai Jeong. "Explainable Convolutional Neural Network to Investigate Age-Related Changes in Multi-Order Functional Connectivity." Electronics 10, no. 23 (2021): 3020. http://dx.doi.org/10.3390/electronics10233020.
Full textAbstract:
Functional connectivity (FC) is a potential candidate that can increase the performance of brain-computer interfaces (BCIs) in the elderly because of its compensatory role in neural circuits. However, it is difficult to decode FC by the current machine learning techniques because of a lack of physiological understanding. To investigate the suitability of FC in BCIs for the elderly, we propose the decoding of lower- and higher-order FC using a convolutional neural network (CNN) in six cognitive-motor tasks. The layer-wise relevance propagation (LRP) method describes how age-related changes in FCs impact BCI applications for the elderly compared to younger adults. A total of 17 young adults 24.5±2.7 years and 12 older 72.5±3.2 years adults were recruited to perform tasks related to hand-force control with or without mental calculation. The CNN yielded a six-class classification accuracy of 75.3% in the elderly, exceeding the 70.7% accuracy for the younger adults. In the elderly, the proposed method increased the classification accuracy by 88.3% compared to the filter-bank common spatial pattern. The LRP results revealed that both lower- and higher-order FCs were dominantly overactivated in the prefrontal lobe, depending on the task type. These findings suggest a promising application of multi-order FC with deep learning on BCI systems for the elderly.
20
N., N., M. Balakrishnan, S. R. Indurekaa, and A. B. Arockia Christopher. "Explainable AI for Automated Feature Extraction in Medical Image Segmentation." International Journal of BIM and Engineering Science 9, no. 2 (2024): 10–18. http://dx.doi.org/10.54216/ijbes.090202.
Full textAbstract:
Automated feature extraction and segmentation of medical images are essential for accurate diagnostics, enabling the identification of relevant structures with minimal human intervention. This study introduces an Explainable AI (XAI) framework for automated feature extraction in medical image segmentation, aiming to enhance transparency in deep learning models used in medical imaging. The proposed framework uses a Convolutional Neural Network (CNN) with integrated attention mechanisms and layer-wise relevance propagation (LRP) to identify critical features while segmenting regions of interest. Testing on datasets of MRI brain scans and CT liver scans, the model achieved an accuracy of 94%, a Dice similarity coefficient (DSC) of 0.88, and an Intersection over Union (IoU) score of 0.83. These results outperform conventional CNN-based segmentation techniques by 10% on average, highlighting the framework's precision in identifying and segmenting intricate structures, including lesions and abnormalities. Additionally, the XAI components provide visual explanations of the segmentation process, enabling clinicians to understand which features influenced the model's decisions. This enhanced transparency is crucial for building trust in AI-driven medical solutions, ultimately facilitating their integration into clinical workflows.
21
Weitz, Katharina, Teena Hassan, Ute Schmid, and Jens-Uwe Garbas. "Deep-learned faces of pain and emotions: Elucidating the differences of facial expressions with the help of explainable AI methods." tm - Technisches Messen 86, no. 7-8 (2019): 404–12. http://dx.doi.org/10.1515/teme-2019-0024.
Full textAbstract:
AbstractDeep neural networks are successfully used for object and face recognition in images and videos. In order to be able to apply such networks in practice, for example in hospitals as a pain recognition tool, the current procedures are only suitable to a limited extent. The advantage of deep neural methods is that they can learn complex non-linear relationships between raw data and target classes without limiting themselves to a set of hand-crafted features provided by humans. However, the disadvantage is that due to the complexity of these networks, it is not possible to interpret the knowledge that is stored inside the network. It is a black-box learning procedure. Explainable Artificial Intelligence (AI) approaches mitigate this problem by extracting explanations for decisions and representing them in a human-interpretable form. The aim of this paper is to investigate the explainable AI methods Layer-wise Relevance Propagation (LRP) and Local Interpretable Model-agnostic Explanations (LIME). These approaches are applied to explain how a deep neural network distinguishes facial expressions of pain from facial expressions of emotions such as happiness and disgust.
22
Cho, Kyung-Chul, Si-Woo Park, Injun Lee, and Jaesool Shim. "Process Prediction and Feature Visualization of Meltblown Nonwoven Fabrics Using Scanning Electron Microscopic (SEM) Image-Based Deep Neural Network Algorithms." Processes 11, no. 12 (2023): 3388. http://dx.doi.org/10.3390/pr11123388.
Full textAbstract:
Meltblown nonwoven fabrics are used in various products, such as masks, protective clothing, industrial filters, and sanitary products. As the range of products incorporating meltblown nonwoven fabrics has recently expanded, numerous studies have been conducted to explore the correlation between production process conditions and the performance of meltblown nonwoven fabrics. Deep neural network algorithms, including convolutional neural networks (CNNs), have been widely applied in numerous industries for tasks such as object detection, recognition, classification, and fault detection. In this study, the correlation between the meltblown nonwoven fabric production process and performance was analyzed using deep neural network algorithms for classifying SEM images. The SEM images of meltblown nonwovens produced under various process conditions were trained using well-known convolutional neural network models (VGG16, VGG19, ResNet50, and DenseNet121), and each model showed high accuracy ranging from 95% to 99%. In addition, LRP (Layer-wise Relevance Propagation) and Gradient-weighted Class Activation Mapping (Grad-CAM) models were applied to visualize and analyze the characteristics and correlation of the SEM images to predict the meltblown nonwoven fabric production process.
23
Taiyeb Khosroshahi, Mahdieh, Soroush Morsali, Sohrab Gharakhanlou, et al. "Explainable Artificial Intelligence in Neuroimaging of Alzheimer’s Disease." Diagnostics 15, no. 5 (2025): 612. https://doi.org/10.3390/diagnostics15050612.
Full textAbstract:
Alzheimer’s disease (AD) remains a significant global health challenge, affecting millions worldwide and imposing substantial burdens on healthcare systems. Advances in artificial intelligence (AI), particularly in deep learning and machine learning, have revolutionized neuroimaging-based AD diagnosis. However, the complexity and lack of interpretability of these models limit their clinical applicability. Explainable Artificial Intelligence (XAI) addresses this challenge by providing insights into model decision-making, enhancing transparency, and fostering trust in AI-driven diagnostics. This review explores the role of XAI in AD neuroimaging, highlighting key techniques such as SHAP, LIME, Grad-CAM, and Layer-wise Relevance Propagation (LRP). We examine their applications in identifying critical biomarkers, tracking disease progression, and distinguishing AD stages using various imaging modalities, including MRI and PET. Additionally, we discuss current challenges, including dataset limitations, regulatory concerns, and standardization issues, and propose future research directions to improve XAI’s integration into clinical practice. By bridging the gap between AI and clinical interpretability, XAI holds the potential to refine AD diagnostics, personalize treatment strategies, and advance neuroimaging-based research.
24
Golla, Alena-K., Christian Tönnes, Tom Russ, et al. "Automated Screening for Abdominal Aortic Aneurysm in CT Scans under Clinical Conditions Using Deep Learning." Diagnostics 11, no. 11 (2021): 2131. http://dx.doi.org/10.3390/diagnostics11112131.
Full textAbstract:
Abdominal aortic aneurysms (AAA) may remain clinically silent until they enlarge and patients present with a potentially lethal rupture. This necessitates early detection and elective treatment. The goal of this study was to develop an easy-to-train algorithm which is capable of automated AAA screening in CT scans and can be applied to an intra-hospital environment. Three deep convolutional neural networks (ResNet, VGG-16 and AlexNet) were adapted for 3D classification and applied to a dataset consisting of 187 heterogenous CT scans. The 3D ResNet outperformed both other networks. Across the five folds of the first training dataset it achieved an accuracy of 0.856 and an area under the curve (AUC) of 0.926. Subsequently, the algorithms performance was verified on a second data set containing 106 scans, where it ran fully automated and resulted in an accuracy of 0.953 and an AUC of 0.971. A layer-wise relevance propagation (LRP) made the decision process interpretable and showed that the network correctly focused on the aortic lumen. In conclusion, the deep learning-based screening proved to be robust and showed high performance even on a heterogeneous multi-center data set. Integration into hospital workflow and its effect on aneurysm management would be an exciting topic of future research.
25
Masri, Sari, Ahmad Hasasneh, Mohammad Tami, and Chakib Tadj. "Exploring the Impact of Image-Based Audio Representations in Classification Tasks Using Vision Transformers and Explainable AI Techniques." Information 15, no. 12 (2024): 751. http://dx.doi.org/10.3390/info15120751.
Full textAbstract:
An important hurdle in medical diagnostics is the high-quality and interpretable classification of audio signals. In this study, we present an image-based representation of infant crying audio files to predict abnormal infant cries using a vision transformer and also show significant improvements in the performance and interpretability of this computer-aided tool. The use of advanced feature extraction techniques such as Gammatone Frequency Cepstral Coefficients (GFCCs) resulted in a classification accuracy of 96.33%. For other features (spectrogram and mel-spectrogram), the performance was very similar, with an accuracy of 93.17% for the spectrogram and 94.83% accuracy for the mel-spectrogram. We used our vision transformer (ViT) model, which is less complex but more effective than the proposed audio spectrogram transformer (AST). We incorporated explainable AI (XAI) techniques such as Layer-wise Relevance Propagation (LRP), Local Interpretable Model-agnostic Explanations (LIME), and attention mechanisms to ensure transparency and reliability in decision-making, which helped us understand the why of model predictions. The accuracy of detection was higher than previously reported and the results were easy to interpret, demonstrating that this work can potentially serve as a new benchmark for audio classification tasks, especially in medical diagnostics, and providing better prospects for an imminent future of trustworthy AI-based healthcare solutions.
26
Seong, Jihyeon, Jungmin Kim, and Jaesik Choi. "Towards Diverse Perspective Learning with Selection over Multiple Temporal Poolings." Proceedings of the AAAI Conference on Artificial Intelligence 38, no. 8 (2024): 8948–56. http://dx.doi.org/10.1609/aaai.v38i8.28743.
Full textAbstract:
In Time Series Classification (TSC), temporal pooling methods that consider sequential information have been proposed. However, we found that each temporal pooling has a distinct mechanism, and can perform better or worse depending on time series data. We term this fixed pooling mechanism a single perspective of temporal poolings. In this paper, we propose a novel temporal pooling method with diverse perspective learning: Selection over Multiple Temporal Poolings (SoM-TP). SoM-TP dynamically selects the optimal temporal pooling among multiple methods for each data by attention. The dynamic pooling selection is motivated by the ensemble concept of Multiple Choice Learning (MCL), which selects the best among multiple outputs. The pooling selection by SoM-TP's attention enables a non-iterative pooling ensemble within a single classifier. Additionally, we define a perspective loss and Diverse Perspective Learning Network (DPLN). The loss works as a regularizer to reflect all the pooling perspectives from DPLN. Our perspective analysis using Layer-wise Relevance Propagation (LRP) reveals the limitation of a single perspective and ultimately demonstrates diverse perspective learning of SoM-TP. We also show that SoM-TP outperforms CNN models based on other temporal poolings and state-of-the-art models in TSC with extensive UCR/UEA repositories.
27
Alam, Mahbub Ul, Jaakko Hollmén, Jón Rúnar Baldvinsson, and Rahim Rahmani. "SHAMSUL: Systematic Holistic Analysis to investigate Medical Significance Utilizing Local interpretability methods in deep learning for chest radiography pathology prediction." Nordic Machine Intelligence 3, no. 1 (2023): 27–47. http://dx.doi.org/10.5617/nmi.10471.
Full textAbstract:
The interpretability of deep neural networks has become a subject of great interest within the medical and healthcare domain. This attention stems from concerns regarding transparency, legal and ethical considerations, and the medical significance of predictions generated by these deep neural networks in clinical decision support systems. To address this matter, our study delves into the application of four well-established interpretability methods: Local Interpretable Model-agnostic Explanations (LIME), Shapley Additive exPlanations (SHAP), Gradient-weighted Class Activation Mapping (Grad-CAM), and Layer-wise Relevance Propagation (LRP). Leveraging the approach of transfer learning with a multi-label-multi-class chest radiography dataset, we aim to interpret predictions pertaining to specific pathology classes. Our analysis encompasses both single-label and multi-label predictions, providing a comprehensive and unbiased assessment through quantitative and qualitative investigations, which are compared against human expert annotation. Notably, Grad-CAM demonstrates the most favorable performance in quantitative evaluation, while the LIME heatmap score segmentation visualization exhibits the highest level of medical significance. Our research underscores both the outcomes and the challenges faced in the holistic approach adopted for assessing these interpretability methods and suggests that a multimodal-based approach, incorporating diverse sources of information beyond chest radiography images, could offer additional insights for enhancing interpretability in the medical domain.
28
Moon, Jucheol, Yong-Min Shin, Jin-Duk Park, Nelson Hebert Minaya, Won-Yong Shin, and Sang-Il Choi. "Explainable gait recognition with prototyping encoder–decoder." PLOS ONE 17, no. 3 (2022): e0264783. http://dx.doi.org/10.1371/journal.pone.0264783.
Full textAbstract:
Human gait is a unique behavioral characteristic that can be used to recognize individuals. Collecting gait information widely by the means of wearable devices and recognizing people by the data has become a topic of research. While most prior studies collected gait information using inertial measurement units, we gather the data from 40 people using insoles, including pressure sensors, and precisely identify the gait phases from the long time series using the pressure data. In terms of recognizing people, there have been a few recent studies on neural network-based approaches for solving the open set gait recognition problem using wearable devices. Typically, these approaches determine decision boundaries in the latent space with a limited number of samples. Motivated by the fact that such methods are sensitive to the values of hyper-parameters, as our first contribution, we propose a new network model that is less sensitive to changes in the values using a new prototyping encoder–decoder network architecture. As our second contribution, to overcome the inherent limitations due to the lack of transparency and interpretability of neural networks, we propose a new module that enables us to analyze which part of the input is relevant to the overall recognition performance using explainable tools such as sensitivity analysis (SA) and layer-wise relevance propagation (LRP).
29
Radke, Tim, Susanne Fuchs, Christian Wilms, Iuliia Polkova, and Marc Rautenhaus. "Explaining neural networks for detection of tropical cyclones and atmospheric rivers in gridded atmospheric simulation data." Geoscientific Model Development 18, no. 4 (2025): 1017–39. https://doi.org/10.5194/gmd-18-1017-2025.
Full textAbstract:
Abstract. Detection of atmospheric features in gridded datasets from numerical simulation models is typically done by means of rule-based algorithms. Recently, the feasibility of learning feature detection tasks using supervised learning with convolutional neural networks (CNNs) has been demonstrated. This approach corresponds to semantic segmentation tasks widely investigated in computer vision. However, while in recent studies the performance of CNNs was shown to be comparable to human experts, CNNs are largely treated as a “black box”, and it remains unclear whether they learn the features for physically plausible reasons. Here we build on the recently published “ClimateNet” dataset that contains features of tropical cyclones (TCs) and atmospheric rivers (ARs) as detected by human experts. We adapt the explainable artificial intelligence technique “Layer-wise Relevance Propagation” (LRP) to the semantic segmentation task and investigate which input information CNNs with the Context-Guided Network (CGNet) and U-Net architectures use for feature detection. We find that both CNNs indeed consider plausible patterns in the input fields of atmospheric variables. For instance, relevant patterns include point-shaped extrema in vertically integrated precipitable water (TMQ) and circular wind motion for TCs. For ARs, relevant patterns include elongated bands of high TMQ and eastward winds. Such results help to build trust in the CNN approach. We also demonstrate application of the approach for finding the most relevant input variables (TMQ is found to be most relevant, while surface pressure is rather irrelevant) and evaluating detection robustness when changing the input domain (a CNN trained on global data can also be used for a regional domain, but only partially contained features will likely not be detected). However, LRP in its current form cannot explain shape information used by the CNNs, although our findings suggest that the CNNs make use of both input values and the shape of patterns in the input fields. Also, care needs to be taken regarding the normalization of input values, as LRP cannot explain the contribution of bias neurons, accounting for inputs close to zero. These shortcomings need to be addressed by future work to obtain a more complete explanation of CNNs for geoscientific feature detection.
30
Yi, *Eun-Gyoung, Miseon Shim, Hyeon-Ho Hwang, Sunhae Jeon, Han-Jeong Hwang, and Seung-Hwan Lee. "DEVELOPMENT OF AN XAI-BASED COMPUTER-AIDED DIAGNOSTIC SYSTEM FOR DRUG-NAÏ VE MALE MDD PATIENTS." International Journal of Neuropsychopharmacology 28, Supplement_1 (2025): i331. https://doi.org/10.1093/ijnp/pyae059.591.
Full textAbstract:
Abstract Background Recently, deep learning-based computer-aided diagnostic (CAD) systems have been actively developed to assist in the accurate diagnosis of patients with major depressive disorder (MDD). While deep learning- based CAD systems utilizing electroencephalography (EEG) data have emerged as promising diagnostic tools, challenges such as data transparency and neurophysiological interpretability persist. Aims & Objectives The present study aims to facilitate precise diagnosis and investigation of the neurophysiological characteristics inherent to patients with Major Depressive Disorder (MDD) by utilizing explainable artificial intelligence (XAI) technology. Also, we integrate XAI into EEG-based CAD systems to increasing its practicality in terms of the number of EEG channels. Method To achieve this objective, resting-state EEG data were collected from 40 male MDD patients and 41 sex- matched healthy controls (HCs). The EEG data were band-pass filtered from 1 to 55 Hz. Subsequently, independent component analysis and common average reference techniques were applied to remove various external artifacts, such as eye blinks and electrocardiogram. After that, the data were downsampled to 200 Hz to reduce the computational cost and divided into approximately 3-minute segments to ensure uniform length across all subjects. A shallow ConvNet model developed specifically for EEG data analysis, based on the convolution neural network (CNN) algorithm, was used to classify between MDD patients and HCs [1]. To prevent overfitting and improve the generalization of the diagnostic system, a leave-one-out cross-validation approach was employed. Furthermore, the relevance scores computed using the layer-wise relevance propagation (LRP) method guided the channel selection process [2]. Results The diagnostic performance of the proposed CAD system was evaluated based on the number of selected EEG channels. The results showed a classification accuracy of 100.00% when distinguishing between MDD patients and HCs using all 62 channels. In addition, for MDD patients, higher relevance scores were observed in the prefrontal and occipital lobes of the right hemisphere. In contrast, HCs showed higher relevance scores in the prefrontal and occipital lobes of the left hemisphere, which differed from MDD patients. Notably, when employing the XAI-based channel selection algorithm, a substantial accuracy of 92.59% was achieved with only 5 channels, primarily located in the occipital lobe. The selected channels notably aligned with areas that exhibited the most discernible power spectral density (PSD) features between MDD patients and HCs. Discussion & Conclusion In the present study, we developed a deep-learning-based CAD system to assist in accurately diagnosing patients with Major Depressive Disorder (MDD). We achieved a high diagnostic performance of 100.00% without the hand-crafted feature extraction. It was found from the results that only 5 channels were sufficient to diagnose MDD patients with a high diagnostic accuracy of 92.59%. Furthermore, the proposed CAD system enabled the investigation of unique neurophysiological characteristics of MDD patients. In summary, our proposed deep learning-based CAD system not only provides high diagnostic accuracy but also offers improved practicality in terms of the number of channels used. References [1]Schirrmeister, R. T. et al, 2017, ‘Deep learning with convolutional neural networks for EEG decoding and visualization’, Human Brain Mapping, vol. 38, pp. 5391-5420 [2]Binder, A. et al, 2016, ‘Layer-wise relevance propagation for neural networks with local renormalization layers’, Artificial Neural Networks and Machine Learning, vol. 9887, pp. 63-71
31
Sar, Shuvam, Soumya Mitra, Parthasarathi Panda, et al. "In Silico Modeling and Structural Analysis of Soluble Epoxide Hydrolase Inhibitors for Enhanced Therapeutic Design." Molecules 28, no. 17 (2023): 6379. http://dx.doi.org/10.3390/molecules28176379.
Full textAbstract:
Human soluble epoxide hydrolase (sEH), a dual-functioning homodimeric enzyme with hydrolase and phosphatase activities, is known for its pivotal role in the hydrolysis of epoxyeicosatrienoic acids. Inhibitors targeting sEH have shown promising potential in the treatment of various life-threatening diseases. In this study, we employed a range of in silico modeling approaches to investigate a diverse dataset of structurally distinct sEH inhibitors. Our primary aim was to develop predictive and validated models while gaining insights into the structural requirements necessary for achieving higher inhibitory potential. To accomplish this, we initially calculated molecular descriptors using nine different descriptor-calculating tools, coupled with stochastic and non-stochastic feature selection strategies, to identify the most statistically significant linear 2D-QSAR model. The resulting model highlighted the critical roles played by topological characteristics, 2D pharmacophore features, and specific physicochemical properties in enhancing inhibitory potential. In addition to conventional 2D-QSAR modeling, we implemented the Transformer-CNN methodology to develop QSAR models, enabling us to obtain structural interpretations based on the Layer-wise Relevance Propagation (LRP) algorithm. Moreover, a comprehensive 3D-QSAR analysis provided additional insights into the structural requirements of these compounds as potent sEH inhibitors. To validate the findings from the QSAR modeling studies, we performed molecular dynamics (MD) simulations using selected compounds from the dataset. The simulation results offered crucial insights into receptor–ligand interactions, supporting the predictions obtained from the QSAR models. Collectively, our work serves as an essential guideline for the rational design of novel sEH inhibitors with enhanced therapeutic potential. Importantly, all the in silico studies were performed using open-access tools to ensure reproducibility and accessibility.
32
Croce, Danilo, Daniele Rossini, and Roberto Basili. "Neural embeddings: accurate and readable inferences based on semantic kernels." Natural Language Engineering 25, no. 4 (2019): 519–41. http://dx.doi.org/10.1017/s1351324919000238.
Full textAbstract:
AbstractSentence embeddings are the suitable input vectors for the neural learning of a number of inferences about content and meaning. Similarity estimation, classification, emotional characterization of sentences as well as pragmatic tasks, such as question answering or dialogue, have largely demonstrated the effectiveness of vector embeddings to model semantics. Unfortunately, most of the above decisions are epistemologically opaque as for the limited interpretability of the acquired neural models based on the involved embeddings. We think that any effective approach to meaning representation should be at least epistemologically coherent. In this paper, we concentrate on the readability of neural models, as a core property of any embedding technique consistent and effective in representing sentence meaning. In this perspective, this paper discusses a novel embedding technique (the Nyström methodology) that corresponds to the reconstruction of a sentence in a kernel space, inspired by rich semantic similarity metrics (a semantic kernel) rather than by a language model. In addition to being based on a kernel that captures grammatical and lexical semantic information, the proposed embedding can be used as the input vector of an effective neural learning architecture, called Kernel-based deep architectures (KDA). Finally, it also characterizes by design the KDA explanatory capability, as the proposed embedding is derived from examples that are both human readable and labeled. This property is obtained by the integration of KDAs with an explanation methodology, called layer-wise relevance propagation (LRP), already proposed in image processing. The Nyström embeddings support here the automatic compilation of argumentations in favor or against a KDA inference, in form of an explanation: each decision can in fact be linked through LRP back to the real examples, that is, the landmarks linguistically related to the input instance. The KDA network output is explained via the analogy with the activated landmarks. Quantitative evaluation of the explanations shows that richer explanations based on semantic and syntagmatic structures characterize convincing arguments, as they effectively help the user in assessing whether or not to trust the machine decisions in different tasks, for example, Question Classification or Semantic Role Labeling. This confirms the epistemological benefit that Nyström embeddings may bring, as linguistically rich and meaningful representations for a variety of inference tasks.
33
Chintha Vishnu Vardhana Reddy. "Lrfe: A Novel Local Response Feature Elimination Process for Identification of Lung Cancer Cells." Journal of Electrical Systems 20, no. 3 (2024): 1375–96. http://dx.doi.org/10.52783/jes.3546.
Full textAbstract:
One of the main causes of cancer-related mortality across the globe is lung cancer. Early-stage lung cancer frequently exhibits no symptoms, which delays diagnosis until the illness has progressed. Before symptoms manifest, screening and early detection techniques can aid in the early diagnosis and treatment of lung cancer. Many olden research papers have implemented image processing and few latest papers have implemented computer vision techniques to detect lung cancer. Particularly when dealing with minor or subtle anomalies, image processing algorithms may not be able to detect lung cancer lesions with sufficient sensitivity and specificity. It is still difficult to increase the detection algorithms' accuracy and dependability, especially when dealing with early-stage lesions or situations where attributes overlap. It takes a lot of processing power, such as high-performance GPUs and enormous memory capacities, to train deep learning models, particularly large-scale convolutional neural networks (CNNs). In this proposed research, the model pre-processes the images using the ostu and sober filter mechanisms because Otsu's approach adjusts to the features of the input image, including noise, contrast, and lighting fluctuations. It is capable of handling images with varying dynamic ranges and intensity distributions without depending on pre-established threshold settings. When it comes to image noise, the Sobel filter is more resilient than other edge detection methods. It produces clearer edge maps and fewer false detections by determining the gradient magnitude, which amplifies edge information while suppressing noise. The features are extracted using the tuned AlexNet pre-trained model, in AlexNet there is a layer known as “Layer-wise Relevance Propagation”. By giving each pixel or feature in the input image a relevance score, the LRP layer offers fine-grained feature attribution. This makes it possible to analyze in great depth which particular elements or areas of the input image are most important for the network to forecast, which helps to clarify the underlying patterns that the network has learned. At last, the extracted features are further reduced using the enhanced feature elimination method. By iteratively selecting subsets of features based on their importance, RFE helps to identify the most relevant features for the classification task. Integrating RFE with SVM classifiers can lead to improved model performance by focusing on the subset of features that are most discriminative and informative for the classification problem.
34
Chintha Vishnu Vardhana Reddy. "LRFE: A NOVEL LOCAL RESPONSE FEATURE ELIMINATION PROCESS FOR IDENTIFICATION OF LUNG CANCER CELLS." Journal of Electrical Systems 20, no. 5s (2024): 2879–97. http://dx.doi.org/10.52783/jes.3202.
Full textAbstract:
One of the main causes of cancer-related mortality across the globe is lung cancer. Early-stage lung cancer frequently exhibits no symptoms, which delays diagnosis until the illness has progressed. Before symptoms manifest, screening and early detection techniques can aid in the early diagnosis and treatment of lung cancer. Many olden research papers have implemented image processing and few latest papers have implemented computer vision techniques to detect lung cancer. Particularly when dealing with minor or subtle anomalies, image processing algorithms may not be able to detect lung cancer lesions with sufficient sensitivity and specificity. It is still difficult to increase the detection algorithms' accuracy and dependability, especially when dealing with early-stage lesions or situations where attributes overlap. It takes a lot of processing power, such as high-performance GPUs and enormous memory capacities, to train deep learning models, particularly large-scale convolutional neural networks (CNNs). In this proposed research, the model pre-processes the images using the ostu and sober filter mechanisms because Otsu's approach adjusts to the features of the input image, including noise, contrast, and lighting fluctuations. It is capable of handling images with varying dynamic ranges and intensity distributions without depending on pre-established threshold settings. When it comes to image noise, the Sobel filter is more resilient than other edge detection methods. It produces clearer edge maps and fewer false detections by determining the gradient magnitude, which amplifies edge information while suppressing noise. The features are extracted using the tuned AlexNet pre-trained model, in AlexNet there is a layer known as “Layer-wise Relevance Propagation”. By giving each pixel or feature in the input image a relevance score, the LRP layer offers fine-grained feature attribution. This makes it possible to analyze in great depth which particular elements or areas of the input image are most important for the network to forecast, which helps to clarify the underlying patterns that the network has learned. At last, the extracted features are further reduced using the enhanced feature elimination method. By iteratively selecting subsets of features based on their importance, RFE helps to identify the most relevant features for the classification task. Integrating RFE with SVM classifiers can lead to improved model performance by focusing on the subset of features that are most discriminative and informative for the classification problem.
35
Gutiérrez-Mondragón, Mario A., Alfredo Vellido, and Caroline König. "A Study on the Robustness and Stability of Explainable Deep Learning in an Imbalanced Setting: The Exploration of the Conformational Space of G Protein-Coupled Receptors." International Journal of Molecular Sciences 25, no. 12 (2024): 6572. http://dx.doi.org/10.3390/ijms25126572.
Full textAbstract:
G-protein coupled receptors (GPCRs) are transmembrane proteins that transmit signals from the extracellular environment to the inside of the cells. Their ability to adopt various conformational states, which influence their function, makes them crucial in pharmacoproteomic studies. While many drugs target specific GPCR states to exert their effects—thereby regulating the protein’s activity—unraveling the activation pathway remains challenging due to the multitude of intermediate transformations occurring throughout this process, and intrinsically influencing the dynamics of the receptors. In this context, computational modeling, particularly molecular dynamics (MD) simulations, may offer valuable insights into the dynamics and energetics of GPCR transformations, especially when combined with machine learning (ML) methods and techniques for achieving model interpretability for knowledge generation. The current study builds upon previous work in which the layer relevance propagation (LRP) technique was employed to interpret the predictions in a multi-class classification problem concerning the conformational states of the β2-adrenergic (β2AR) receptor from MD simulations. Here, we address the challenges posed by class imbalance and extend previous analyses by evaluating the robustness and stability of deep learning (DL)-based predictions under different imbalance mitigation techniques. By meticulously evaluating explainability and imbalance strategies, we aim to produce reliable and robust insights.
36
Ranjit M. Gawande. "Machine Learning Approaches for Fault Detection and Diagnosis in Electrical Machines: A Comparative Study of Deep Learning and Classical Methods." Panamerican Mathematical Journal 34, no. 2 (2024): 121–37. http://dx.doi.org/10.52783/pmj.v34.i2.930.
Full textAbstract:
Fault detection and diagnosis in electrical machines are crucial for ensuring their safe and reliable operation. In recent years, machine learning techniques have emerged as powerful tools for addressing this challenge, offering the potential for more accurate and efficient fault detection and diagnosis compared to traditional methods. Among these techniques, deep learning has gained significant attention due to its ability to automatically learn relevant features from raw data. However, the performance of deep learning models in this domain has not been extensively compared to classical methods. This paper presents a comparative study of deep learning and classical methods for fault detection and diagnosis in electrical machines. The study evaluates the performance of various machine learning algorithms, including deep neural networks, support vector machines, decision trees, and ensemble methods, in detecting and diagnosing faults such as stator winding faults, rotor faults, and bearing faults. The experimental evaluation is conducted using real-world datasets obtained from electrical machines in industrial settings. Performance metrics such as accuracy, precision, recall, and F1-score are used to assess the effectiveness of each approach in detecting and diagnosing faults accurately and efficiently. The results of the study indicate that deep learning approaches, particularly convolutional neural networks (CNNs) and recurrent neural networks (RNNs), outperform classical methods in terms of fault detection and diagnosis accuracy. These deep learning models demonstrate the ability to automatically extract informative features from raw sensor data, enabling them to effectively identify subtle patterns indicative of faults. The study investigates the interpretability of deep learning models compared to classical methods, examining the extent to which the models can provide insights into the underlying causes of faults. While deep learning models typically operate as black boxes, techniques such as layer-wise relevance propagation (LRP) are employed to enhance their interpretability and facilitate the identification of relevant features contributing to fault detection and diagnosis. This comparative study provides valuable insights into the strengths and limitations of deep learning and classical methods for fault detection and diagnosis in electrical machines, offering guidance for practitioners and researchers in selecting appropriate approaches for their specific applications.
37
R, Jain. "Transparency in AI Decision Making: A Survey of Explainable AI Methods and Applications." Advances in Robotic Technology 2, no. 1 (2024): 1–10. http://dx.doi.org/10.23880/art-16000110.
Full textAbstract:
Artificial Intelligence (AI) systems have become pervasive in numerous facets of modern life, wielding considerable influence in critical decision-making realms such as healthcare, finance, criminal justice, and beyond. Yet, the inherent opacity of many AI models presents significant hurdles concerning trust, accountability, and fairness. To address these challenges, Explainable AI (XAI) has emerged as a pivotal area of research, striving to augment the transparency and interpretability of AI systems. This survey paper serves as a comprehensive exploration of the state-of-the-art in XAI methods and their practical applications. We delve into a spectrum of techniques, spanning from model-agnostic approaches to interpretable machine learning models, meticulously scrutinizing their respective strengths, limitations, and real-world implications. The landscape of XAI is rich and varied, with diverse methodologies tailored to address different facets of interpretability. Model-agnostic approaches offer versatility by providing insights into model behavior across various AI architectures. In contrast, interpretable machine learning models prioritize transparency by design, offering inherent understandability at the expense of some predictive performance. Layer-wise Relevance Propagation (LRP) and attention mechanisms delve into the inner workings of neural networks, shedding light on feature importance and decision processes. Additionally, counterfactual explanations open avenues for exploring what-if scenarios, elucidating the causal relationships between input features and model outcomes. In tandem with methodological exploration, this survey scrutinizes the deployment and impact of XAI across multifarious domains. Successful case studies showcase the practical utility of transparent AI in healthcare diagnostics, financial risk assessment, criminal justice systems, and more. By elucidating these use cases, we illuminate the transformative potential of XAI in enhancing decision-making processes while fostering accountability and fairness. Nevertheless, the journey towards fully transparent AI systems is fraught with challenges and opportunities. As we traverse the current landscape of XAI, we identify pressing areas for further research and development. These include refining interpretability metrics, addressing the scalability of XAI techniques to complex models, and navigating the ethical dimensions of transparency in AI decision-making.Through this survey, we endeavor to cultivate a deeper understanding of transparency in AI decision-making, empowering stakeholders to navigate the intricate interplay between accuracy, interpretability, and ethical considerations. By fostering interdisciplinary dialogue and inspiring collaborative innovation, we aspire to catalyze future advancements in Explainable AI, ultimately paving the way towards more accountable and trustworthy AI systems.
38
Huang, Xinyi, Suphanut Jamonnak, Ye Zhao, Tsung Heng Wu, and Wei Xu. "A Visual Designer of Layer‐wise Relevance Propagation Models." Computer Graphics Forum 40, no. 3 (2021): 227–38. http://dx.doi.org/10.1111/cgf.14302.
Full text
39
Jung, Yeon-Jee, Seung-Ho Han, and Ho-Jin Choi. "Explaining CNN and RNN Using Selective Layer-Wise Relevance Propagation." IEEE Access 9 (2021): 18670–81. http://dx.doi.org/10.1109/access.2021.3051171.
Full text
40
Jung, Yeon‐Jee, Seung‐Ho Han, and Ho‐Jin Choi. "SLRP: Improved heatmap generation via selective layer‐wise relevance propagation." Electronics Letters 57, no. 10 (2021): 393–96. http://dx.doi.org/10.1049/ell2.12061.
Full text
41
Bach, Sebastian, Alexander Binder, Grégoire Montavon, Frederick Klauschen, Klaus-Robert Müller, and Wojciech Samek. "On Pixel-Wise Explanations for Non-Linear Classifier Decisions by Layer-Wise Relevance Propagation." PLOS ONE 10, no. 7 (2015): e0130140. http://dx.doi.org/10.1371/journal.pone.0130140.
Full text
42
Kim, Juhwan, Geun Ho Gu, Juhwan Noh, et al. "Predicting potentially hazardous chemical reactions using an explainable neural network." Chemical Science 12, no. 33 (2021): 11028–37. http://dx.doi.org/10.1039/d1sc01049b.
Full textAbstract:
An explainable neural network model is developed to predict the formation of hazardous products for chemical reactions. An input attribution method, layer-wise relevance propagation, is used to explain the decision-making process.
43
Berrone, Stefano, Francesco Della Santa, Antonio Mastropietro, Sandra Pieraccini, and Francesco Vaccarino. "Layer-wise relevance propagation for backbone identification in discrete fracture networks." Journal of Computational Science 55 (October 2021): 101458. http://dx.doi.org/10.1016/j.jocs.2021.101458.
Full text
44
Xu, Jincheng, and Qingfeng Du. "Adversarial attacks on text classification models using layer‐wise relevance propagation." International Journal of Intelligent Systems 35, no. 9 (2020): 1397–415. http://dx.doi.org/10.1002/int.22260.
Full text
45
A. Ahmed, Awadelrahman M., and Leen A. M. Ali. "Explainable Medical Image Segmentation via Generative Adversarial Networks and Layer-wise Relevance Propagation." Nordic Machine Intelligence 1, no. 1 (2021): 20–22. http://dx.doi.org/10.5617/nmi.9126.
Full textAbstract:
This paper contributes in automating medical image segmentation by proposing generative adversarial network based models to segment both polyps and instruments in endoscopy images. A main contribution of this paper is providing explanations for the predictions using layer-wise relevance propagation approach, showing which pixels in the input image are more relevant to the predictions. The models achieved 0.46 and 0.70, on Jaccard index and 0.84 and 0.96 accuracy, on the polyp segmentation and the instrument segmentation, respectively.
46
Grezmak, John, Jianjing Zhang, Peng Wang, Kenneth A. Loparo, and Robert X. Gao. "Interpretable Convolutional Neural Network Through Layer-wise Relevance Propagation for Machine Fault Diagnosis." IEEE Sensors Journal 20, no. 6 (2020): 3172–81. http://dx.doi.org/10.1109/jsen.2019.2958787.
Full text
47
Seetharam, Akshay. "U-Net Color Bias for Image Segmentation Demonstrated by Layer-Wise Relevance Propagation." International Journal of High School Research 5, no. 2 (2023): 1–4. http://dx.doi.org/10.36838/v5i2.1.
Full text
48
Li, Heyi, Yunke Tian, Klaus Mueller, and Xin Chen. "Beyond saliency: Understanding convolutional neural networks from saliency prediction on layer-wise relevance propagation." Image and Vision Computing 83-84 (March 2019): 70–86. http://dx.doi.org/10.1016/j.imavis.2019.02.005.
Full text
49
Taghian, Mehran, Shotaro Miwa, Yoshihiro Mitsuka, Johannes Günther, Shadan Golestan, and Osmar Zaiane. "Explainability of deep reinforcement learning algorithms in robotic domains by using Layer-wise Relevance Propagation." Engineering Applications of Artificial Intelligence 137 (November 2024): 109131. http://dx.doi.org/10.1016/j.engappai.2024.109131.
Full text
50
Lennartz, Rebecca, Arash Khassetarash, Sandro R. Nigg, Bjoern M. Eskofier, and Benno M. Nigg. "Neural network and layer-wise relevance propagation reveal how ice hockey protective equipment restricts players’ motion." PLOS ONE 19, no. 10 (2024): e0312268. http://dx.doi.org/10.1371/journal.pone.0312268.
Full textAbstract:
Understanding the athlete’s movements and the restrictions incurred by protective equipment is crucial for improving the equipment and subsequently, the athlete’s performance. The task of equipment improvement is especially challenging in sports including advanced manoeuvres such as ice hockey and requires a holistic approach guiding the researcher’s attention toward the right variables. The purposes of this study were (a) to quantify the effects of protective equipment in ice hockey on player’s performance and (b) to identify the restrictions incurred by it. Twenty male hockey players performed four different drills with and without protective equipment while their performance was quantified. A neural network accompanied by layer-wise relevance propagation was applied to the 3D kinematic data to identify variables and time points that were most relevant for the neural network to distinguish between the equipment and no equipment condition, and therefore presumable result from restrictions incurred by the protective equipment. The study indicated that wearing the protective equipment, significantly reduced performance. Further, using the 3D kinematics, an artificial neural network could accurately distinguish between the movements performed with and without the equipment. The variables contributing the most to distinguishing between the equipment conditions were related to the upper extremities and movements in the sagittal plane. The presented methodology consisting of artificial neural networks and layer-wise relevance propagation contributed to insights without prior knowledge of how and to which extent joint angles are affected in complex maneuvers in ice hockey in the presence of protective equipment. It was shown that changes to the equipment should support the flexion movements of the knee and hip and should allow players to keep their upper extremities closer to the torso.