CS Student
CS student
RadianceAI
Our project aims to revolutionize the healthcare industry by developing an advanced deep learning-based system for automated diagnosis and reporting of chest X-ray images. The primary objective is to leverage the power of computer vision and natural language processing (NLP) to enhance the efficiency, accuracy, and accessibility of medical diagnostics. We start by utilizing a large-scale dataset of chest X-ray images, such as the widely recognized CheXpert dataset, which contains a diverse range of pathological conditions. These images serve as the foundation for training and fine-tuning deep learning models. Our approach involves exploring and comparing different state-of-the-art convolutional neural network (CNN) architectures, including Resnet50, Resnet101, and VGG16, to determine the most suitable model for our diagnostic system. Additionally, we explore the integration of NLP techniques to further enhance the diagnostic capabilities of our system. We develop a parallel LSTM decoder that takes as input the feature vector extracted from the last layer of the CNN model and the corresponding radiology report. This decoder employs a softmax activation function to generate word probabilities, allowing for the production of accurate and contextually relevant diagnostic descriptions. Our ultimate goal is to provide healthcare professionals with a powerful tool that enhances their diagnostic capabilities, reduces workload, and improves patient outcomes. By automating the analysis and interpretation of chest X-ray images, our system has the potential to significantly impact the efficiency and accuracy of medical diagnostics, leading to more timely interventions and better patient care.