Comprehensive comparative analysis of explainable deep learning model for differentiation of brucellar spondylitis and tuberculous spondylitis through MRI sequences
Open Access
- 24 December 2025
- journal article
- research article
- Published by Springer Nature in European Journal of Medical Research
- Vol. 31 (1)
Abstract
Background: The differentiation of brucellar spondylitis (BS) from tuberculous spondylitis (TS) on magnetic resonance imaging (MRI) is a critical clinical challenge. While deep learning holds promise, the optimal architectural strategy for integrating information from multi-sequence MRI remains unclear. This study systematically compared distinct deep learning architectures to identify a valid and effective integration strategy for this diagnostic problem. Methods: In this retrospective, single-center diagnostic study, we included 235 patients with surgically and pathologically confirmed BS ( n = 82) or TS ( n = 153) from January 2014 to December 2024. We systematically evaluated four distinct architectural strategies for processing sagittal T1-weighted, T2-weighted, and fat-suppressed MRI sequences: (1) baseline models trained on single sequences; (2) a single-branch model that fused sequences as input channels; (3) a heterogeneous multi-branch model using different backbones for each sequence; and (4) a homogeneous multi-branch model using identical backbones. Models were developed on patient-level data splits for training (70%), validation (15%), and internal testing (15%). The primary performance metric was the area under the receiver operating characteristic curve (AUC) on the test set. Statistical significance of performance differences between models was assessed using the DeLong test, with P values adjusted for multiple comparisons using the Benjamini–Hochberg procedure. Results: The single-branch fusion model, which treated the three sequences as channels in a single input, failed to learn, yielding performance equivalent to random chance (test AUC range: 0.474–0.538). In stark contrast, both the single-sequence and multi-branch architectures proved to be effective. The best single-sequence model achieved a test AUC of 0.765 (95% CI 0.759–0.771). The optimal multi-branch model, which successfully integrated all three sequences, achieved a comparable test AUC of 0.764 (95% CI 0.757–0.770). Conclusions: The choice of architecture for integrating multi-sequence MRI data is a critical determinant of model viability. Our findings demonstrate that naive channel wise fusion is an invalid strategy for this task. In contrast, both processing a single MRI sequence and utilizing a multi-branch parallel-processing architecture are valid and effective strategies, achieving comparable diagnostic performance. This study clarifies the architectural principles required for successfully applying deep learning to this multi-modal diagnostic challenge.Keywords
Funding Information
- Xinjiang Uygur Autonomous Region "Tianshan talents" medical and health high-level talent training plan (TSYC202301B087)
- Natural Science Foundation General Project of the Science and Technology Department of Xinjiang Uygur Autonomous Region (2021D01A161)
This publication has 31 references indexed in Scilit:
- MRI‐Based Multiple Instance Convolutional Neural Network for Increased Accuracy in the Differentiation of Borderline and Malignant Epithelial Ovarian TumorsJournal of Magnetic Resonance Imaging, 2021
- Deep Learning for Classification of Bone Lesions on Routine MRIEBioMedicine, 2021
- The Angiopoietin-like protein 4: a promising biomarker to distinguish brucella spondylitis from tuberculous spondylitisClinical Rheumatology, 2021
- Management of Pyogenic Spinal Infection, review of literatureJournal of Orthopaedics, 2019
- Semi-automatic classification of prostate cancer on multi-parametric MR imaging using a multi-channel 3D convolutional neural networkEuropean Radiology, 2019
- Multi-branch convolutional neural network for multiple sclerosis lesion segmentationNeuroImage, 2019
- The Efficacy of Computed Tomography-Guided Percutaneous Spine Biopsies in Determining a Causative Organism in Cases of Suspected Infection: A Systematic ReviewCanadian Association of Radiologists Journal, 2019
- Differentiation Between Brucellar and Tuberculous Spondylodiscitis in the Acute and Subacute Stages by MRIAcademic Radiology, 2018
- Comparison of brucellar and tuberculous spondylodiscitis patients: results of the multicenter “Backbone-1 Study”The Spine Journal, 2015
- Efficacy of CT-guided biopsies of the spine in patients with spondylitis – an analysis of 164 proceduresEuropean Journal of Radiology, 2012