Academic literature on the topic 'Chest X-ray classification'

The thesis develops a model (that includes a conceptual framework and an implementation) for analysing and classifying traditional X-ray images (MACXI) according to the severity of diseases as a Computer-Aided-Diagnosis tool with three initial objectives. � The first objective was to interpret X-ray images by transferring expert knowledge into a knowledge base (CXKB): to help medical staff to concentrate only on the interest areas of the images. � The second objective was to analyse and classify X-ray images according to the severity of diseases through the knowledge base equipped with an image processor (CXIP). � The third objective was to demonstrate the effectiveness and limitations of several image-processing techniques for analysing traditional chest X-ray images. A database was formed based on collection of expert diagnosis details for lung images. Five important features from lung images, as well as diagnosis rules were identified and simplified. The expert knowledge was transformed into a Knowledge base (KB) for analysing and classifying traditional X-ray images according to the severity of diseases (CXKB). Finally, an image processor named CXIP was developed to extract the features of lung images features and image classification. CXKB contains 63 distinct lung diseases with detailed descriptions. Some 80-chest X-ray images with diagnosis details were collected for the database from different sources, including online medical resources. A total of 61 images were used to determine the important features; 19 chest X-ray images were not used because of low visibility or the difficulty of diagnosis. Finally, only 12 images were selected after examining the diagnosis details, image clarity, image completeness, and image orientation. The most important features of lung diseases are a pattern of lesions with different levels of intensity or brightness. The other major anatomical structures of the chest are the hilum area, the rib area, the trachea area, and the heart area. Seven different severity levels of diseases were determined. Development and simplification of rules based on the image library were analysed, developed, and tested against the 12 images. A level of severity was labelled for each image based on a personal understanding of all the image and diagnosis details. Then, MACXI processed the selected 12 images to determine the level of severity. These 12 images were fed into the CXIP for recognition of the features and classification of the images to an accurate level of severity. Currently, the processor has the ability to identify diseased lung areas with approximately 80% success rate. A step by step demonstration of several image processing techniques that were used to build the processor is given to highlight the effectiveness and limitations of the techniques for analysing traditional chest X-ray images is also presented.

Les maladies pulmonaires représentent une cause majeure de décès dans le monde, et le diagnostic précoce est crucial pour améliorer les chances de rétablissement. Les technologies d’Intelligence Artificielle ont ouvert des voies prometteuses dans le domaine biomédical. Ainsi dans cette thèse,des modèles d’IA sont utilisés pour améliorer la performance de classification des maladies pulmonaires à partir des images de radiographie thoracique. De nouvelles approches de prétraitement basées sur CycleGAN sont développées pour réduire l’effet du bruit causé par les artefacts tels que des dispositifs médicaux dans les radiographies thoraciques, ainsi que pour générer des masques incluant les zones pathologiques dans les régions d’intérêt. Ensuite, une nouvelle approche de sélection de caractéristiques est développée pour identifier a priori les caractéristiques statistiquement les plus significatives avant la classification. Au-delà de l’analyse des images, les données cliniques associées sont également examinées pour affiner le modèle de classification selon le profil du patient, ce qui améliore l’efficacité diagnostique. Les avancées proposées présentent des résultats prometteurs améliorant la performance de la classification binaire et multi-label des maladies pulmonaires<br>Lung diseases are a major cause of death worldwide, and early diagnosis is crucial to improve the chance of recovery. Artificial Intelligence technologies have opened promising avenues in the biomedical field. Thus, in this thesis, AI models are used to improve the classification performance of lung diseases from chest X-ray images. New preprocessing approaches based on CycleGAN are developed to reduce the noise effect caused by artifacts such as medical devices in chest Xrays, as well as to generate masks that include pathological areas within the regions of interest. Additionally, a new feature selection approach is developed to identify the statistically most significant features a priori before classification. Beyond image analysis, the associated clinical data are also examined to refine the classification model according to the patient’s profile, enhancing diagnostic effectiveness. The proposed advancements show promising results in improving the performance of both binary and multi-label classification of lung diseases

University of Technology Sydney. Faculty of Engineering and Information Technology.<br>Computer-aided diagnosis (CAD) systems have been successfully helped to clinical diagnosis. This dissertation considers one essential task in CAD, the chest X-ray (CXR) image classification problem, with the deep learning technologies from the following three aspects. First, considering most diseases existing in CXRs usually happen in small, localized areas, we propose to localize the local discriminative regions and integrate the global and local cues into an attention guided convolution neural network (AG-CNN) to identify thorax diseases. AG-CNN consists of three branches (global, local, and fusion branches). The global branch learns the global features for classification. The local branch localizes the discriminative regions, which avoids noise and improves misalignment in the global branch. AG-CNN fuses the global and local features for diagnosis in a fusion branch. Second, due to the common and complex relationships of multiple diseases in CXRs, it is worth exploiting their correlations to help the diagnosis. This thesis will present a category-wise residual attention learning method to concentrate on learning the correlations of multiple diseases. It is expected to suppress the obstacles of irrelevant categories and strengthen the relevant features at the same time. Last, a robust and stable CXR image analysis system should be able to: 1) automatically focus on the disease-critical regions, which usually are of small sizes; 2) adaptively capture the intrinsic relationships among different disease features and utilize them to boost the multi-label disease recognition rates jointly. We introduce a discriminative feature learning framework, ConsultNet, to achieve those two purposes simultaneously. ConsultNet consists of a variational selective information bottleneck branch and a spatial-and-channel encoding branch. These two branches learn discriminative features collaboratively. In addition, each of the proposed methods is comprehensively verified and analysed by conducting various experiments.

碩士<br>國立交通大學<br>統計學研究所<br>107<br>Deep learning nowadays has attracted attention, especially in medical images classification because of its effectiveness and good performance that can compete with the medical images expert. Despite these successes there are the strong belief among experts that deep learning only efficient for the big datasets and for small datasets deep learning would produce a bad performance. For this study, it is aimed to build a deep learning model for image classification that can achieve high accuracy using chest x-ray images with relatively small dataset. We classify all normal chest x-ray images and all abnormalities in chest x-ray images into a binary classifier. We built and tested our model using the public dataset of Shenzen Hospital dataset. We use different type of input images based on different preprocessing and different type of learning technique so that the model can perform accurate classification for this particular dataset. Based on the result, pre-trained CheXNet with new trained fully connected network on cropped dataset can achieve the accuracy 0.8761, the sensitivity 0.8909, and the specificity 0.8621. The performance of the model also influenced by the certain area in the images, like other region outside the lung and black region outside the body.

碩士<br>國立交通大學<br>工業工程與管理系所<br>107<br>In recent years, rapid development of artificial intelligence in all fields, such as Drones, Smart home or autonomous cars, brought significant convenience to people's lives. However, advances in medical imaging techniques have completely changed the diagnosis method of medical images. Through the development of this technology, all traditional medical images gradually evolved from manual interpretation to digital assisting-interpretation. Therefore, the high development of artificial intelligence in medical field can not only assist physicians in disease diagnosis, but also boost the future development and innovation in medical field. Therefore, this study formed up a chest X-ray image classification model based on convolutional neural network. Actual chest X-ray images provided by Kaggle were used for modeling, testing and verifying the feasibility and effectiveness of the model constructed by this study. After verifying results in this study, the model in this study can provide an effective diagnostic technique for physicians in medical imaging diagnosis and expecting for more development of imaging technology in the medical field based on artificial intelligence.

碩士<br>國立臺中科技大學<br>資訊管理系碩士班<br>105<br>Digital image processing has been applied in medical domain widely, but the multitude still needs to manual processing. Automatic image segmentation and features analysis can assist doctor treatment and diagnose diseases more accurately, reduce the time of diagnosing and improve efficiency. Automatic medical image segmentation is difficult in that the image quality varied by equipment and dosage. In this thesis, the automatic method employed image multiscale intensity texture analysis and segmentation to surmount this problem. The proposed method automatically recognize and classify abnormal region without manual segmentation. Generally, automatic identification is based on the difference of the texture and organ shape, or any pathological changes of lung area. Therefore, the important features could be retained to identify abnormal areas. In this thesis, the chest x-ray images for finding whether lung region is healthy or not. The first proposed identifying common pneumothorax is based on SVM to classification method. Features are extracted from the lung image by the local binary pattern. Then, classification of pneumothorax lung is determined by support vector machines. The second proposed automatic pneumothorax detection is based on multiscale intensity texture segmentation. Remove the background and noises in the chest images for segmenting the lung of abnormal region. The segmenting the abnormal region. is used texture transforms from computing multiple overlapping blocks. Because the ribs boundaries are affected easily, the rib boundaries are identified by using Sobel edge detection. Finally, in order to obtain a complete disease region, the rib boundary is filled up in the rib boundary located between the abnormal regions.

Tese de mestrado em Bioinformática e Biologia Computacional, Universidade de Lisboa, Faculdade de Ciências, 2021<br>A Tuberculose (TB) continua a ser um dos principais problemas de saúde global na actualidade, com especial incidência em países de terceiro mundo pertencentes a África e Sudoeste Asiático, que somam 84% dos 1.5 milhões de óbitos derivados de TB durante o ano de 2017. A interpretação de Raio-X é um indicador forte no diagnóstico de TB que, quando combinado com outros indicadores como tosse, febre ou outros sintomas suspeitos, pode levar a um diagnóstico bastante preciso. A interpretação de uma imagem de Raio-X requer a competência de um Médico Radiologista experiente, um requisito limitado especialmente considerando a incidência de TB em países de terceiro mundo. Esta interpretação pode ser facilitada através do uso de Redes Neuronais Convolucionais (CNN) que, quando treinadas correctamente, conseguem ultrapassar o desempenho de profissionais de saúde. No entanto, o correcto treino de CNN requer largas quantidades de imagens classificadas, um recurso inexistente no domínio público para TB. O uso de Aprendizagem por Transferência, de fácil implementação para CNNs, é uma solução bastante popular na implementação de CNNs para a interpretação de imagens médicas, contornando os largos requisitos de imagens. Contudo, a sua comum implementação tende a não usar uma abordagem eficaz, e poucos trabalhos exploram as vantagens do uso de Aprendizagem por Transferência. Este trabalho procura explorar o uso de Aprendizagem por Transferência para a optimização do treino de CNNs em conjuntos de dados de TB bastante limitados. A exploração passa pelo uso de Bases de Referência Aleatórias e treinadas no grande conjunto de dados ImageNet, de modo a explorar as vantagens do uso de Aprendizagem por Transferência. Além destes, cinco Bases de Referência adicionais são treinadas em dois conjuntos de Raio-X de larga escala, o ChestX-ray8 e o CheXpert, na tentativa de optimizar a transferência de conhecimento para a classificação de TB. O treino de modelos em TB faz uso do conjunto de dados \Shenzhen Hospital X-ray Set", no qual os modelos são treinados, validados e testados. O conjunto de dados \Montgomery Hospital X-ray Set" é usado apenas para teste. O resultado deste trabalho são 155 classificadores de TB, para os quais os melhores resultados são atingidos usando uma Base de Referência treinada no conjunto completo de CheXpert, atingindo um valor mediano de 0.65 de WAF, e 0.77 de AUROC, no conjunto de teste externo. Adicionalmente, este trabalho verifica resultados mais optimistas pelas medidas de AUROC. Esta diferença resulta do limite usado para sumarizar o output das redes, para o qual este trabalho sugere uma estimativa alternativa usando um número limitado de dados de teste que acaba por melhorar os resultados de WAF, aproximando-os das medidas de AUROC.<br>Tuberculosis (TB) continues to be one of the main sources of global health concern, with increased incidence in third world countries in Africa and Southwest Asia, which account for 84% of the 1.5 million deaths due to TB during the year of 2017. The interpretation of X-ray is a strong indicator in the diagnosis of TB which, when combined with other indicators such as cough, fever or other suspicious symptoms, can lead to a very accurate diagnosis. The interpretation of an X-ray image requires the expertise of an experienced Radiologist, a limited resource emphasized by the incidence of TB in third world countries. This interpretation can be assisted through the use of Convolutional Neural Networks (CNN) which, when properly trained, can surpass the performance of health professionals. However, the correct training of CNN requires large amounts of classified images, a resource that does not exist in the public domain for TB. The use of Transfer Learning is a very popular solution when implementing CNNs for the interpretation of medical images, bypassing the wide requirements of images. However, its common implementation tends to not use an effective approach, and few studies explore the advantages of using Transfer Learning. This work seeks to explore the use of Transfer Learning for the optimization of CNN training in very limited TB datasets. Exploration involves the use of Random Baselines and Baselines trained on the large dataset ImageNet, exploring the advantages of Transfer Learning. In addition to these, five additional Baselines are trained on two large-scale X-ray sets, the ChestX-ray8 and the CheXpert, in an attempt to optimize the transfer of knowledge for the classification of TB. The training of models for TB uses the \Shenzhen Hospital X-ray Set" dataset for training, validation and testing. The \Montgomery Hospital X-ray Set" dataset is used for testing purposes only. The result of this work is 155 TB classifiers, for which the best results are achieved using a Baseline trained in the complete set of CheXpert, reaching a median value of 0.65 WAF, and 0.77 of AUROC, on the external test set. Additionally, this work verifies more optimistic results for AUROC measures. This difference results from the threshold used to summarize the output of the networks, for which this work suggests an alternative estimate using a limited number of test data that ends up improving the results of WAF, bringing them closer to the AUROC measures.