김경수 (서울대학교) | 한빛사논문 > 한빛사

한빛사논문

조회229

서울대학교

CV File 1 MB

Med. Image Anal. , Apr 01 2025, 103 S1361-8415(25)00106-9 | https://doi.org/10.1016/j.media.2025.103559 Improved unsupervised 3D lung lesion detection and localization by fusing global and local features: Validation in 3D low-dose computed tomography

Ju Hwan Lee ^a,b,1, Seong Je Oh ^c,1, Kyungsu Kim ^d,e,2, Chae Yeon Lim ^a,f, Seung Hong Choi ^g,h, Myung Jin Chung ^a,i,j,3

^aMedical AI Research Center, Research Institute for Future Medicine, Samsung Medical Center, Seoul, Republic of Korea
^bDepartment of Health Sciences and Technology, SAIHST, Sungkyunkwan University, Seoul, Republic of Korea
^cInstitute of Radiation Medicine, Seoul National University Medical Research Center, Seoul, Republic of Korea
^dSchool of Transdisciplinary Innovations, Artificial Intelligence Institute, Interdisciplinary Program in Bioengineering, and Interdisciplinary Program in Artificial Intelligence, Seoul, Republic of Korea
^eDepartment of Biomedical Science, Medical Research Center, SNUH Institute of Convergence Medicine with Innovative Technology, SNUH Institute of Healthcare AI Research, Seoul, Republic of Korea
^fDepartment of Medical Device Management and Research, SAIHST, Sungkyunkwan University, Seoul, Republic of Korea
^gSchool of Transdisciplinary Innovations, Interdisciplinary Program in Bioengineering, and Interdisciplinary Program in Artificial Intelligence, Seoul National University, Seoul, Republic of Korea
^hDepartment of Radiology, Department of Biomedical Science and Medical Research Center, Seoul National University, Seoul, Republic of Korea
ⁱDepartment of Data Convergence and Future Medicine, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea
^jDepartment of Radiology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea

¹Ju Hwan Lee and Seong Je Oh contributed equally to this work as co-first authors.
²Kyungsu Kim contributed equally to this work as the co-first author and as the co-corresponding author.
³Myung Jin Chung contributed equally to this work as the co-corresponding author.
Corresponding authors: Kyungsu Kim, Myung Jin Chung

Abstract

Unsupervised anomaly detection (UAD) is crucial in low-dose computed tomography (LDCT). Recent AI technologies, leveraging global features, have enabled effective UAD with minimal training data of normal patients. However, this approach, devoid of utilizing local features, exhibits vulnerability in detecting deep lesions within the lungs. In other words, while the conventional use of global features can achieve high specificity, it often comes with limited sensitivity. Developing a UAD AI model with high sensitivity is essential to prevent false negatives, especially in screening patients with diseases demonstrating high mortality rates. We have successfully pioneered a new LDCT UAD AI model that leverages local features, achieving a previously unattainable increase in sensitivity compared to global methods (17.5% improvement). Furthermore, by integrating this approach with conventional global-based techniques, we have successfully consolidated the advantages of each model - high sensitivity from the local model and high specificity from the global model - into a single, unified, trained model (17.6% and 33.5% improvement, respectively). Without the need for additional training, we anticipate achieving significant diagnostic efficacy in various LDCT applications, where both high sensitivity and specificity are essential, using our fixed model. Code is available at https://github.com/kskim-phd/Fusion-UADL.

논문정보

형식Research article
게재일2025년 04월 (BRIC 등록일 2025-04-14)
연구진국내 연구진
분야 바이오·의료융합 > 다학제 융합/복합

댓글 작성 로그인

CV 열람하기

연락처 보기