Nearest-Neighbor Dual-Path Contrastive Learning for Lumbar Disc Herniation MRI Image Classification

Lumbar disc herniation (LDH) is a common spinal condition that profoundly affects patients’ quality of life. Timely and precise diagnosis is essential for efficient therapy and enhancing patient outcomes. This paper introduces an innovative LDH classification framework utilizing nearest-n...

Full description

Saved in:
Bibliographic Details
Main Authors: Dan Pan, Yu-Xiang Pan, Hui Wang, Qi-Jing Liu, Chang-Mao Qiu
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11006703/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849395857359634432
author Dan Pan
Yu-Xiang Pan
Hui Wang
Qi-Jing Liu
Chang-Mao Qiu
author_facet Dan Pan
Yu-Xiang Pan
Hui Wang
Qi-Jing Liu
Chang-Mao Qiu
author_sort Dan Pan
collection DOAJ
description Lumbar disc herniation (LDH) is a common spinal condition that profoundly affects patients’ quality of life. Timely and precise diagnosis is essential for efficient therapy and enhancing patient outcomes. This paper introduces an innovative LDH classification framework utilizing nearest-neighbor dual-path contrastive learning, integrating global and local feature learning to improve lumbar MRI image classification efficacy. The global path delineates semantic linkages among samples by nearest-neighbor contrastive learning, enhancing global representations, whereas the local path employs clipped regions and data augmentation to highlight essential details, thus improving fine-grained feature modeling. The novel nearest-neighbor-based positive sample construction enhances feature consistency and classification accuracy by reducing the impact of irrelevant examples. Our method achieves state-of-the-art accuracy and robustness in complicated classification tasks, as shown by experimental results on the lumbar MRI dataset. This discovery enhances automated LDH diagnosis and offers a viable avenue for accurate and efficient automated diagnosis in intricate medical imaging contexts.
format Article
id doaj-art-8dfcb10dd1f74775a3ea03e927c1b923
institution Kabale University
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-8dfcb10dd1f74775a3ea03e927c1b9232025-08-20T03:39:29ZengIEEEIEEE Access2169-35362025-01-0113929069292010.1109/ACCESS.2025.357101411006703Nearest-Neighbor Dual-Path Contrastive Learning for Lumbar Disc Herniation MRI Image ClassificationDan Pan0https://orcid.org/0009-0005-4698-6324Yu-Xiang Pan1Hui Wang2Qi-Jing Liu3Chang-Mao Qiu4Department of Rehabilitation Medicine, The First Affiliated Hospital of Gannan Medical University, Ganzhou, ChinaDepartment of Rehabilitation Medicine, Gannan Healthcare Vocational College, Ganzhou, ChinaDepartment of Pharmacy, Ganzhou People’s Hospital, Ganzhou, ChinaDepartment of Rehabilitation Sciences, The Hong Kong Polytechnic University, Hung Hom, Hong Kong, ChinaMedical Imaging Center, The First Affiliated Hospital of Gannan Medical University, Ganzhou, ChinaLumbar disc herniation (LDH) is a common spinal condition that profoundly affects patients’ quality of life. Timely and precise diagnosis is essential for efficient therapy and enhancing patient outcomes. This paper introduces an innovative LDH classification framework utilizing nearest-neighbor dual-path contrastive learning, integrating global and local feature learning to improve lumbar MRI image classification efficacy. The global path delineates semantic linkages among samples by nearest-neighbor contrastive learning, enhancing global representations, whereas the local path employs clipped regions and data augmentation to highlight essential details, thus improving fine-grained feature modeling. The novel nearest-neighbor-based positive sample construction enhances feature consistency and classification accuracy by reducing the impact of irrelevant examples. Our method achieves state-of-the-art accuracy and robustness in complicated classification tasks, as shown by experimental results on the lumbar MRI dataset. This discovery enhances automated LDH diagnosis and offers a viable avenue for accurate and efficient automated diagnosis in intricate medical imaging contexts.https://ieeexplore.ieee.org/document/11006703/Lumbar spine disc classificationlumbar spinecontrastive learningdeep learning
spellingShingle Dan Pan
Yu-Xiang Pan
Hui Wang
Qi-Jing Liu
Chang-Mao Qiu
Nearest-Neighbor Dual-Path Contrastive Learning for Lumbar Disc Herniation MRI Image Classification
IEEE Access
Lumbar spine disc classification
lumbar spine
contrastive learning
deep learning
title Nearest-Neighbor Dual-Path Contrastive Learning for Lumbar Disc Herniation MRI Image Classification
title_full Nearest-Neighbor Dual-Path Contrastive Learning for Lumbar Disc Herniation MRI Image Classification
title_fullStr Nearest-Neighbor Dual-Path Contrastive Learning for Lumbar Disc Herniation MRI Image Classification
title_full_unstemmed Nearest-Neighbor Dual-Path Contrastive Learning for Lumbar Disc Herniation MRI Image Classification
title_short Nearest-Neighbor Dual-Path Contrastive Learning for Lumbar Disc Herniation MRI Image Classification
title_sort nearest neighbor dual path contrastive learning for lumbar disc herniation mri image classification
topic Lumbar spine disc classification
lumbar spine
contrastive learning
deep learning
url https://ieeexplore.ieee.org/document/11006703/
work_keys_str_mv AT danpan nearestneighbordualpathcontrastivelearningforlumbardischerniationmriimageclassification
AT yuxiangpan nearestneighbordualpathcontrastivelearningforlumbardischerniationmriimageclassification
AT huiwang nearestneighbordualpathcontrastivelearningforlumbardischerniationmriimageclassification
AT qijingliu nearestneighbordualpathcontrastivelearningforlumbardischerniationmriimageclassification
AT changmaoqiu nearestneighbordualpathcontrastivelearningforlumbardischerniationmriimageclassification