Develop a ChatGPT-style system for CT scan analysis

Principal Investigator

Name
Neet Patel

Degrees
B. Tech

Institution
University of California San Diego

Position Title
Student

Email
neetp868@gmail.com

About this CDAS Project

Study
NLST (Learn more about this study)

Project ID
NLST-1073

Initial CDAS Request Approval
May 30, 2023

Title
Develop a ChatGPT-style system for CT scan analysis

Summary
The proposed system will leverage the capabilities of ChatGPT to provide an interactive and user-friendly interface for medical professionals and patients who need to analyze and understand CT scan images.
Receive CT scan images as input and ask various questions like generate natural language descriptions of the findings, such as abnormalities, tumor, etc.
Answer questions about the CT scan images, such as the location, size, shape, and severity of the findings, the possible causes and diagnoses etc.
Learn from feedback and corrections from the users and improve its accuracy and reliability over time.

Aims

The specific aims of this project are:
To collect and preprocess a large dataset of CT scan images and their corresponding natural language descriptions.
To evaluate the performance of the system in terms of accuracy, consistency, comprehensibility, informativeness, and user satisfaction.

Collaborators

Prof. Pengtao Xie - University of California, San Diego