Arun KumarKaur, Manjeet2022-11-052022-11-052022-08https://krishikosh.egranth.ac.in/handle/1/5810189618Currently, lung diseases are extremely common throughout the globe and few of which include chronic obstructive pulmonary disease, pneumonia, pneumothorax, tuberculosis etc. Air in the pleural cavity is thought to be as “pneumothorax". This is a state when the lung surface or chest wall is breached, allowing air to enter the pleural space and causing the lung to collapse, it is a serious situation and might be life-threatening. Chest radiographs are the foremost significant and extensively used diagnostic techniques to detect thoracic or lung diseases among various medical imaging technologies. Pneumothorax can be diagnosed by chest X-rays, CT scans or ultrasound techniques but chest X-ray is the most often used accessible radiological method for screening and diagnosing thoracic diseases. The diagnosis from the chest X-Ray can be of great efficiency and time saving if done through automated systems instead of manual image reading. Tremendous research works are being conducted to develop reliable automatic diagnostic systems for detecting diseases within the chest radiographs. Artificial Intelligence (AI) tools have proven to be effective in optimizing the medical industry. Several framework-supported computing techniques have been proposed for the automated identification of pneumothorax from chest radiographs. Numerous models for pneumothorax automated diagnosis are certainly available. In this study, we have taken chest X-ray images and then used the Deep Transfer Learning technique along with Machine Learning to detect the presence or absence of pneumothorax in chest X-ray images. We have aggregated the unique and computational attributes of Deep Learning and Machine Learning. We have used the Deep Transfer Learning technique which is a pre-trained model that is a Residual Network for image feature extraction and the Support Vector Machine (SVM) algorithm which creates the best decision boundary known as “Hyperplane” to classify data points, is a Machine Learning model used for pneumothorax binary classification. To achieve effective outcomes, a balancing data technique with augmentation is used to create a balance between the training and validation dataset, as well as an automatic adjusting learning rate technique called "ReduceLROnPlateau" to monitor validation loss and obtain optimal learning rates. As an implementation tool, “Google Colab” is used. In our proposed method model, it outperforms in terms of metrics that include accuracy, recall (Sensitivity), f-1 score, precision, loss, and AUC(Area Under The Curve). The performance of the framework is evaluated on a dataset that is available on “Kaggle” which is chest X-ray images. This research work has set a new record with a good performance by achieving state-of-the-art results as 0.8831 in terms of AUC and 0.4375 in terms of loss. The Precision, Recall, and f-1 score is obtained as 0.78, 0.81, and 0.794 respectively.EnglishThesis