Automated Detection of Acute Appendicitis from Contrast-enhanced CT Scans using Generative Adversarial Networks for Augmented Classification in Acute Abdomen Cases

The research introduces an innovative method to detect acute appendicitis by uniting symptom-based forecasting with state-of-the-art image analysis techniques. The system employs an FNN for syndrome evaluation which interacts with a 3D ResNet model for analyzing CT scan contrast images with the objective to enhance diagnostic speed while increasing accuracy. Users supply patient symptoms to an FNN model that calculates appendicitis risk probabilities. User needs to upload CT scan files when the system detects probabilities exceeding the designated threshold. The system uses 3D ResNet to examine the uploaded images. Users benefit from the MERN stack-based website because it delivers an intuitive experience for both symptom entry and CT scan file upload capabilities. The machine learning models that run for real-time predictions operate on FastAPI infrastructure. This two-level diagnostic process quickly delivers precise appendicitis diagnosis that assists health care professionals in their medical choices. A forthcoming enhancement plan includes the addition of components for patient management as well as appointment scheduling while adding extra predictive models to enhance system strength.