Cheung, J.P.Y., Kuang, X., Lai, M.K.L. et al. Learning-based fully automated prediction of lumbar disc degeneration progression with specified clinical parameters and preliminary validation. Eur Spine J 31, 1960–1968 (2022). https://doi.org/10.1007/s00586-021-07020-x
Learning-based fully automated prediction of lumbar disc degeneration progression with specified clinical parameters and preliminary validation
|Author:||Cheung, Jason Pui Yin1; Kuang, Xihe1; Lai, Marcus Kin Long1;|
1Department of Orthopaedics and Traumatology, Queen Mary Hospital, The University of Hong Kong, 5/F Professorial Block, Pokfulam, Hong Kong
2Medical Research Center Oulu, Oulu University Hospital and University of Oulu, Oulu, Finland
3Finnish Institute of Occupational Health, Oulu, Finland
4Department of Orthopedic Surgery, Rush University Medical Center, Chicago, Il, USA
5International Spine Research and Innovation Initiative, RUSH University Medical Center, Chicago, IL, USA
6Institute of Health Informatics, University College London, London, UK
7Department of Orthopaedics, Faculty of Surgery, Zhejiang University Affiliated Sir Run Run Shaw Hospital, Zhejiang University, Hangzhou, China
8Department of Spine Surgery, The First Affiliated Hospital, Sun Yat Sen University, Guangzhou, China
|Online Access:||PDF Full Text (PDF, 0.5 MB)|
|Persistent link:|| http://urn.fi/urn:nbn:fi-fe20231011139739
|Publish Date:|| 2023-10-11
Background: Lumbar disc degeneration (LDD) may be related to aging, biomechanical and genetic factors. Despite the extensive work on understanding its etiology, there is currently no automated tool for accurate prediction of its progression.
Purpose: We aim to establish a novel deep learning-based pipeline to predict the progression of LDD-related findings using lumbar MRIs.
Materials and methods: We utilized our dataset with MRIs acquired from 1,343 individual participants (taken at the baseline and the 5-year follow-up timepoint), and progression assessments (the Schneiderman score, disc bulging, and Pfirrmann grading) that were labelled by spine specialists with over ten years clinical experience. Our new pipeline was realized by integrating the MRI-SegFlow and the Visual Geometry Group-Medium (VGG-M) for automated disc region detection and LDD progression prediction correspondingly. The LDD progression was quantified by comparing the Schneiderman score, disc bulging and Pfirrmann grading at the baseline and at follow-up. A fivefold cross-validation was conducted to assess the predictive performance of the new pipeline.
Results: Our pipeline achieved very good performances on the LDD progression prediction, with high progression prediction accuracy of the Schneiderman score (Accuracy: 90.2 ± 0.9%), disc bulging (Accuracy: 90.4% ± 1.1%), and Pfirrmann grading (Accuracy: 89.9% ± 2.1%).
Conclusions: This is the first attempt of using deep learning to predict LDD progression on a large dataset with 5-year follow-up. Requiring no human interference, our pipeline can potentially achieve similar predictive performances in new settings with minimal efforts.
European spine journal
|Pages:||1960 - 1968|
|Type of Publication:||
A1 Journal article – refereed
|Field of Science:||
3126 Surgery, anesthesiology, intensive care, radiology
We would like to thank the Hong Kong Theme-Based Research Scheme (T12-708/12N) for supporting the establishment of the MRI dataset. We would like to thank the Innovation and Technology Commission Seed Fund (ITS/404/18) for supporting the equipment used in this project.