An automatic scoring system for medial temporal lobe atrophy based on YOLO deep learning

By using an improved YOLOv8 model and an adaptive loss function, the problems of slice selection and scoring accuracy in the assessment of medial temporal lobe atrophy were solved, realizing an efficient and accurate automated scoring system suitable for the early diagnosis of Alzheimer's disease.

CN120107158BActive Publication Date: 2026-05-26CHONGQING UNIV
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
CHONGQING UNIV
Filing Date
2025-01-17
Publication Date
2026-05-26

AI Technical Summary

Technical Problem

Current technologies for the early diagnosis of Alzheimer's disease suffer from problems such as inconsistent slice selection, insufficient robustness of scoring algorithms, and lack of efficient clinical integration systems, resulting in low scoring accuracy and difficulty in meeting the needs of high-throughput image analysis.

Method used

An improved YOLOv8 model was used, combined with the EfficientViT structure and the AdaptiveSlide Loss function, to construct a coronal slice screening module, an optimal coronal slice automatic selection module, and a medial temporal lobe atrophy detection and automatic scoring module, thereby achieving automated MRI data processing and scoring.

Benefits of technology

It improves the accuracy and efficiency of medial temporal lobe atrophy (MTA) scoring, reduces subjective error, and achieves rapid and accurate MTA scoring. It is suitable for high-throughput clinical image analysis and has high consistency and convenient clinical application value.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120107158B_ABST
    Figure CN120107158B_ABST
Patent Text Reader

Abstract

An automated scoring system for medial temporal lobe atrophy (MTA) based on YOLO deep learning includes: a coronal slice selection module, an improved YOLOv8 model construction module, an optimal coronal slice automatic selection module, and a medial temporal lobe atrophy detection and automatic scoring module. This invention provides an automated MTA scoring technique that balances accuracy, versatility, and ease of use. By using a lightweight deep learning YOLO object detection framework and combining it with an appropriate loss function, computational overhead is reduced while maintaining algorithm performance. Furthermore, a user-friendly front-end interface improves the efficiency and accuracy of clinical MTA scoring. This invention provides an efficient, accurate, and consistent automated MTA scoring system with significant clinical application value and promising prospects, and is expected to play an important role in the early diagnosis and monitoring of neurodegenerative diseases such as Alzheimer's disease.
Need to check novelty before this filing date? Find Prior Art