How does AI fail at diagnosing Chest Scans?
CWSF · 2026 Disease & Illness
Overview
My project explores the reason Artifical Intelligence makes errors when diagnosing chest x-ray scans. To do this I developed an AI model and ran several experiments to find the root cause of its inaccuracies and how it can be improved. These experiments involved having the model diagnose images of varying levels of clarity, comparing models trained on varying datasets, and analyzing patterns the AI was making when it had trained. The results showed that the AI was heavily limited to its dataset. Specifically it was making patterns only applicable to its training images and so made mistakes when diagnosing outside of its dataset. AI is likely going to contribute more in the medical space, and more understanding of its failures are needed to integrate AI safely.
Awards (1)
- Selected for CWSF 2026
Competition history
- CWSF 2026
Related projects
ISEF · 2025
Convolutional Clarity: Accurate, Reliable, Explainable CNNs by Evaluating the Impact of Data Augmentation on Chest X-Ray Pneumonia Classification
CWSF · 2026
An Efficient AI-Driven Tool for Pediatric Pneumonia Detection
ISEF · 2023
Detection of Malignant Lung Tumors From CT Scans Through Deep Learning-Based Artificial Intelligence Algorithms
ISEF · 2022
Detection of Benign and Malignant Lung Nodules in 3D Volumes Generated From Thoracic Computed Tomography Scans Leveraging Artificial Intelligence, Year 2
Closest projects by meaning, across every fair and year in the corpus.