Optimizing CNN Architectures for Advanced Thoracic Disease Classification
This work addresses medical image analysis for thoracic disease detection, but it is incremental as it builds on existing CNN methods without major breakthroughs.
The study tackled thoracic disease classification from chest X-rays by evaluating CNN architectures and introducing preprocessing and a class-weighted loss function to address dataset imbalance and image quality issues, but did not report concrete performance numbers.
Machine learning, particularly convolutional neural networks (CNNs), has shown promise in medical image analysis, especially for thoracic disease detection using chest X-ray images. In this study, we evaluate various CNN architectures, including binary classification, multi-label classification, and ResNet50 models, to address challenges like dataset imbalance, variations in image quality, and hidden biases. We introduce advanced preprocessing techniques such as principal component analysis (PCA) for image compression and propose a novel class-weighted loss function to mitigate imbalance issues. Our results highlight the potential of CNNs in medical imaging but emphasize that issues like unbalanced datasets and variations in image acquisition methods must be addressed for optimal model performance.