Skip to Main Content

An official website of the United States government

Principal Investigator
Name
Yin-Hung Lin
Degrees
B.S., M.S., Ph.D
Institution
Taiwan AI Labs
Position Title
Project Manager
Email
About this CDAS Project
Study
NLST (Learn more about this study)
Project ID
NLST-441
Initial CDAS Request Approval
Sep 4, 2018
Title
Artificial Intelligence Based Detection and Radiomics Analysis of Lung Nodules from Lung Screening CT
Summary
The objective of our study is to apply Artificial Intelligence (AI) technology to the detection and analysis of lung nodules from CT scans. Recent advances in AI has led to breakthrough applications in various domains of image analysis, including object detection, semantic segmentation and image classification. Such advancements have also shown great promise in the realm of medical image analysis. From a clinical and operations standpoint, AI applications in object detection can significantly augment a clinician's ability to accurately detect and manage pathology, a task of particular significance in the screening of lung cancer and clinical medicine in general. From a research standpoint, AI applications in classification have also demonstrated potential in radiomics, which characterizes pathology using quantitative image features for diagnosis and prognosis. To actualize the clinical and research potential of AI applications in the screening of lung cancer, we plan to build an end-to-end AI pipeline that detects clinically relevant lung nodules, and performs quantitative imaging analysis for the discovery of diagnostic and prognostic radiographic biomarkers for lung cancer.
Aims

1. Use the NLST CT dataset to validate and improve our existing lung nodule detection algorithm
2. Develop an AI radiomics algorithm that uses quantitative image features to uncover diseases characteristics of lung nodule images that could be useful radiographic biomarkers in diagnosis and prognosis.
3. Develop an open-source end-to-end AI software pipeline for the advancement of lung cancer screening research

Collaborators

Ethan Tu, Taiwan AI Labs
Stefan Hong, Taiwan AI Labs