Featured Application: The method presented in this paper can be applied in medical computer systems for supporting medical diagnosis.Abstract: Thoracic radiography (chest X-ray) is an inexpensive but effective and widely used medical imaging procedure. However, a lack of qualified radiologists severely limits the applicability of the technique. Even current Deep Learning-based approaches often require strong supervision, e.g., annotated bounding boxes, to train such systems, which is impossible to harvest on a large scale. In this work, we proposed the classification and prediction of lung pathologies of frontal thoracic X-rays using a modified model MobileNet V2. We considered using transfer learning with metadata leverage. We used the NIH Chest-Xray-14 database, and we did a comparison of performance of our approach to other state-of-the-art methods for pathology classification. The main comparison was by Area under the Receiver Operating Characteristic Curve (AUC) statistics and analyzed the differences between classifiers. Overall, we notice a considerable spread in the achieved result with an average AUC of 0.811 and an accuracy above 90%. We conclude that resampling the dataset gives a huge improvement to the model performance. In this work, we intended to create a model that is capable of being trained, and modified devices with low computing power because they can be implemented into smaller IoT devices.
30 November 2023 in Multimedia Tools and Applications
Pulmonary pathologies are a significant global health concern, often leading to fatal outcomes if not diagnosed and treated promptly. Chest radiography serves as a primary diagnostic tool, but the availability of experienced radiologists remains limited. Advances in Artificial Intelligence (AI) and machine learning, particularly in computer vision, offer promising solutions to address this challenge. This research evaluates a deep learning model designed to detect lung cancer, specifically pulmonary nodules, along with eight other lung pathologies, using chest radiographs. The study leverages diverse datasets comprising over 135,120 frontal chest radiographs to train a Convolutional Neural Network (CNN). A two-stage classification system, utilizing ensemble methods and transfer learning, is employed to first triage images into Normal or Abnormal categories and then identify specific pathologies, including lung nodules. The deep learning model achieves notable results in nodule classification, with a top-performing accuracy of 77%, a sensitivity of 0.713, a specificity of 0.776 during external validation, and an AUC score of 0.888. Despite these successes, some misclassifications were observed, primarily false negatives. In conclusion, the model demonstrates robust potential for generalization across diverse patient populations, attributed to the geographic diversity of the training dataset. Future work could focus on integrating ETL data distribution strategies and expanding the dataset with additional nodule-type samples to further enhance diagnostic accuracy.