Automated lumbar intervertebral disc identification and herniation detection in MR images using cascade CNN architecture

Objective: Identifying herniated discs in MRI scans is inherently challenging due to the small size, irregular shape, and complex appearance of the affected regions. Conventional approaches typically rely on semi-automated region-of-interest (ROI) selection and single-model classification using eith...

Full description

Saved in:
Bibliographic Details
Main Authors: Md Abu Sayed, Ashiqur Rahman, Sadman Mohammad Nasif, Sudipto Halder, Akram Hossain, Hasan Ahmed, Muhammad Abdul Kadir
Format: Article
Language:English
Published: Elsevier 2025-01-01
Series:Informatics in Medicine Unlocked
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S235291482500036X
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850033130549805056
author Md Abu Sayed
Ashiqur Rahman
Sadman Mohammad Nasif
Sudipto Halder
Akram Hossain
Hasan Ahmed
Muhammad Abdul Kadir
author_facet Md Abu Sayed
Ashiqur Rahman
Sadman Mohammad Nasif
Sudipto Halder
Akram Hossain
Hasan Ahmed
Muhammad Abdul Kadir
author_sort Md Abu Sayed
collection DOAJ
description Objective: Identifying herniated discs in MRI scans is inherently challenging due to the small size, irregular shape, and complex appearance of the affected regions. Conventional approaches typically rely on semi-automated region-of-interest (ROI) selection and single-model classification using either axial or sagittal views, limiting diagnostic performance. This study aims to develop an automated, accurate, and efficient system for the detection and classification of lumbar intervertebral disc herniation using deep learning models applied to axial and sagittal MR images. Methods: A YOLO-based framework was developed to automatically identify lumbar intervertebral discs (IVD1-5) and extract ROIs from MR images. Attention-enhanced and fine-tuned VGG19 and ResNet50 models were employed to analyze axial and sagittal images for herniation detection. A decision fusion strategy was used to combine the classification probabilities from both models to further enhance accuracy. The dataset underwent extensive preprocessing and augmentation to improve model robustness and generalization. Results: The proposed approach demonstrated exceptional performance in detection and classification tasks. For detection, the model achieved mAP50 scores of 95.18 % (axial IVD1-5), 99.50 % (lumbar regions), and 94.87 % (sagittal IVD1-5). Classification accuracy reached 97.05 % for axial images and 97.45 % for sagittal images, increasing to 98.09 % with decision fusion. Conclusion: Designed to assist physicians, especially during high-demand periods such as pandemics, this approach has the potential to improve diagnostic efficiency and reduce clinical workload.
format Article
id doaj-art-a7ad7e360e4c40daa7b4a767acb269e1
institution DOAJ
issn 2352-9148
language English
publishDate 2025-01-01
publisher Elsevier
record_format Article
series Informatics in Medicine Unlocked
spelling doaj-art-a7ad7e360e4c40daa7b4a767acb269e12025-08-20T02:58:19ZengElsevierInformatics in Medicine Unlocked2352-91482025-01-015510164810.1016/j.imu.2025.101648Automated lumbar intervertebral disc identification and herniation detection in MR images using cascade CNN architectureMd Abu Sayed0Ashiqur Rahman1Sadman Mohammad Nasif2Sudipto Halder3Akram Hossain4Hasan Ahmed5Muhammad Abdul Kadir6Department of Biomedical Physics and Technology, University of Dhaka, Dhaka, 1000, BangladeshDepartment of Biomedical Physics and Technology, University of Dhaka, Dhaka, 1000, BangladeshDepartment of Computer Science Engineering, Khulna University of Engineering & Technology, Khulna, 9203, BangladeshDepartment of Mechatronics Engineering, Khulna University of Engineering & Technology, Khulna, 9203, BangladeshDepartment of Biomedical Physics and Technology, University of Dhaka, Dhaka, 1000, BangladeshTechnical Technical Expert-Environment, UNOPS South Asia Multi-Country Office, Dhaka, 1212, BangladeshDepartment of Biomedical Physics and Technology, University of Dhaka, Dhaka, 1000, Bangladesh; Corresponding author.Objective: Identifying herniated discs in MRI scans is inherently challenging due to the small size, irregular shape, and complex appearance of the affected regions. Conventional approaches typically rely on semi-automated region-of-interest (ROI) selection and single-model classification using either axial or sagittal views, limiting diagnostic performance. This study aims to develop an automated, accurate, and efficient system for the detection and classification of lumbar intervertebral disc herniation using deep learning models applied to axial and sagittal MR images. Methods: A YOLO-based framework was developed to automatically identify lumbar intervertebral discs (IVD1-5) and extract ROIs from MR images. Attention-enhanced and fine-tuned VGG19 and ResNet50 models were employed to analyze axial and sagittal images for herniation detection. A decision fusion strategy was used to combine the classification probabilities from both models to further enhance accuracy. The dataset underwent extensive preprocessing and augmentation to improve model robustness and generalization. Results: The proposed approach demonstrated exceptional performance in detection and classification tasks. For detection, the model achieved mAP50 scores of 95.18 % (axial IVD1-5), 99.50 % (lumbar regions), and 94.87 % (sagittal IVD1-5). Classification accuracy reached 97.05 % for axial images and 97.45 % for sagittal images, increasing to 98.09 % with decision fusion. Conclusion: Designed to assist physicians, especially during high-demand periods such as pandemics, this approach has the potential to improve diagnostic efficiency and reduce clinical workload.http://www.sciencedirect.com/science/article/pii/S235291482500036XDisc herniationLumbar intervertebral discIVDYOLOv11ROI detectionAttention module
spellingShingle Md Abu Sayed
Ashiqur Rahman
Sadman Mohammad Nasif
Sudipto Halder
Akram Hossain
Hasan Ahmed
Muhammad Abdul Kadir
Automated lumbar intervertebral disc identification and herniation detection in MR images using cascade CNN architecture
Informatics in Medicine Unlocked
Disc herniation
Lumbar intervertebral disc
IVD
YOLOv11
ROI detection
Attention module
title Automated lumbar intervertebral disc identification and herniation detection in MR images using cascade CNN architecture
title_full Automated lumbar intervertebral disc identification and herniation detection in MR images using cascade CNN architecture
title_fullStr Automated lumbar intervertebral disc identification and herniation detection in MR images using cascade CNN architecture
title_full_unstemmed Automated lumbar intervertebral disc identification and herniation detection in MR images using cascade CNN architecture
title_short Automated lumbar intervertebral disc identification and herniation detection in MR images using cascade CNN architecture
title_sort automated lumbar intervertebral disc identification and herniation detection in mr images using cascade cnn architecture
topic Disc herniation
Lumbar intervertebral disc
IVD
YOLOv11
ROI detection
Attention module
url http://www.sciencedirect.com/science/article/pii/S235291482500036X
work_keys_str_mv AT mdabusayed automatedlumbarintervertebraldiscidentificationandherniationdetectioninmrimagesusingcascadecnnarchitecture
AT ashiqurrahman automatedlumbarintervertebraldiscidentificationandherniationdetectioninmrimagesusingcascadecnnarchitecture
AT sadmanmohammadnasif automatedlumbarintervertebraldiscidentificationandherniationdetectioninmrimagesusingcascadecnnarchitecture
AT sudiptohalder automatedlumbarintervertebraldiscidentificationandherniationdetectioninmrimagesusingcascadecnnarchitecture
AT akramhossain automatedlumbarintervertebraldiscidentificationandherniationdetectioninmrimagesusingcascadecnnarchitecture
AT hasanahmed automatedlumbarintervertebraldiscidentificationandherniationdetectioninmrimagesusingcascadecnnarchitecture
AT muhammadabdulkadir automatedlumbarintervertebraldiscidentificationandherniationdetectioninmrimagesusingcascadecnnarchitecture