• Development and validation of...

*Result*: Development and validation of a deep learning-based automatic classification algorithm for the medial temporal lobe atrophy score using a multimodality cascade transformer.

Title:
Development and validation of a deep learning-based automatic classification algorithm for the medial temporal lobe atrophy score using a multimodality cascade transformer.
Authors:
Lee SJ; Department of Radiology, Dongguk University Ilsan Hospital, Goyang, Republic of Korea., Lee D; VUNO Inc., Seoul, Republic of Korea., Suh CH; Department of Radiology and Research Institute of Radiology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea. Electronic address: chonghyunsuh@amc.seoul.kr., Jeong SY; Department of Radiology and Research Institute of Radiology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea; Department of Radiology, Seoul National University Bundang Hospital, Seongnam, Republic of Korea., Shin HM; VUNO Inc., Seoul, Republic of Korea., Jung W; VUNO Inc., Seoul, Republic of Korea., Kim J; VUNO Inc., Seoul, Republic of Korea., Lim JS; Department of Neurology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea., Kim HS; Department of Radiology and Research Institute of Radiology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea., Kim SJ; Department of Radiology and Research Institute of Radiology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea., Lee JH; Department of Neurology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea. Electronic address: jhlee@amc.seoul.kr.
Source:
Clinical radiology [Clin Radiol] 2025 Sep; Vol. 88, pp. 106993. Date of Electronic Publication: 2025 Jun 17.
Publication Type:
Journal Article; Observational Study; Validation Study
Language:
English
Journal Info:
Publisher: Blackwell Scientific Publications Ltd Country of Publication: England NLM ID: 1306016 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1365-229X (Electronic) Linking ISSN: 00099260 NLM ISO Abbreviation: Clin Radiol Subsets: MEDLINE
Imprint Name(s):
Publication: Oxford : Blackwell Scientific Publications Ltd
Original Publication: Edinburgh, Livingstone.
MeSH Terms:
Temporal Lobe*/pathology , Temporal Lobe*/diagnostic imaging , Cognitive Dysfunction*/diagnostic imaging , Cognitive Dysfunction*/pathology , Magnetic Resonance Imaging*/methods , Deep Learning*, Atrophy/diagnostic imaging ; Humans ; Retrospective Studies ; Female ; Male ; Algorithms ; Middle Aged ; Aged ; Reproducibility of Results ; Classification Algorithms
Entry Date(s):
Date Created: 20250717 Date Completed: 20250820 Latest Revision: 20250822
Update Code:
20260130
DOI:
10.1016/j.crad.2025.106993
PMID:
40675115
Database:
MEDLINE

*Further Information*

*Aim: The aim of this study was to develop and validate a deep learning-based automatic classification algorithm for the medial temporal lobe atrophy (MTA) score in patients with cognitive impairment.
Materials and Methods: This retrospective, observational study included consecutive patients with cognitive impairment from a tertiary hospital between March 2017 and June 2021. We developed a deep learning-based model and a machine learning-based model to automate MTA classification. We reorganised the MTA scores into 3 classes (0/1), (2), and (3/4) then classified the right and left MTA scores separately. The internal testing and external testing datasets were applied and compared to validate the performance of the MTA prediction model.
Results: A total of 1694 patients were evaluated for the training dataset, and 297 patients evaluated for the internal testing dataset. 400 patients were evaluated for the external testing dataset. In the internal testing dataset, the accuracy was 0.82 and 0.87 for the left and right MTA classifications, respectively. In the external testing dataset, the accuracy was 0.82 and 0.85 for the left and right MTA classifications, respectively. When comparing the performance between a deep learning-based model and a machine learning-based model, the results were similar.
Conclusion: The deep learning- and machine learning-based automatic classification algorithms for the MTA score accurately classified the MTA score in patients with cognitive impairment.
(Copyright © 2025 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.)*

*Conflict of interest The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Chong Hyun Suh reports financial support was provided by National Research Foundation of Korea. Woo Hyun Shim reports financial support was provided by Ministry of Science and ICT. Dongsoo Lee reports a relationship with VUNO Inc. that includes employment. Hye Min Shin reports a relationship with VUNO Inc. that includes employment. Wooseok Jung reports a relationship with VUNO Inc. that includes employment. Jinyoung Kim reports a relationship with VUNO Inc. that includes employment. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.*