Jie Sheng
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Jie Sheng has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
- Xinghua Song
Department of Spine Surgery, The Sixth Affiliated Hospital of Xinjiang Medical University, Urumqi, People’s Republic of China
2 shared publications Abudouresuli Tuersun
Xinjiang Key Laboratory of Artificial Intelligence Assisted Imaging Diagnosis, Department of Radiology, The First People’s Hospital of Kashi Prefecture, Kashi, People’s Republic of China
1 shared publicationAnuar Ashir
Mangistau Regional Multifunctional Hospital, Aktau City, Republic of Kazakhstan
1 shared publicationHongwei Ye
Department of Pathology, the Sixth People’s Hospital of Xinjiang Uygur Autonomous Region, Urumqi, People’s Republic of China
1 shared publicationParhat Yasin
Department of Spine Surgery, The Sixth Affiliated Hospital of Xinjiang Medical University, Urumqi, People’s Republic of China
1 shared publication- Parhat Yasin
Department of Spine Surgery, The Sixth Affiliated Hospital of Xinjiang Medical University, Urumqi, People’s Republic of China
1 shared publication
Latest Publications
- Open Access
- 2 November 2026
- Published by Springer Nature in BMC Musculoskeletal Disorders
A novel minimally invasive metabolomic strategy for differential diagnosis of spinal infections: blood-derived metabolic signatures as candidate biomarkers distinguishing tuberculous spondylitis from brucellar spondylitis
Background: Tuberculous spondylitis (TS) and brucellar spondylitis (BS) can present with overlapping manifestations. This study assessed plasma metabolic differences between the two diagnostic groups. Methods: We conducted a retrospective case-control study of 53 patients (36 with TS and 17 with BS) enrolled between January 2021 and December 2024. Diagnoses were established using a composite reference standard integrating compatible clinical and radiological findings with microbiological and/or histopathological evidence. Untargeted ultra-high-performance liquid chromatography-tandem mass spectrometry (UHPLC-MS/MS), multivariate analyses and receiver operating characteristic (ROC) analyses were used. Candidate selection and ROC assessment used the same participants. Results: The processed dataset contained 1060 features, of which 190 met the nominal differential-screening criteria (56 with higher and 134 with lower relative peak areas in TS). Lipids and lipid-like molecules accounted for 390 of 1058 classified features (36.86%); 2 features were unclassified. No pathway reached nominal P < 0.05 among 27 tested pathways; linoleic acid metabolism had the lowest P value (0.0767). The highest individual apparent area under the ROC curve (AUC) was 0.922 for the feature annotated as 12-hydroxyjasmonic acid. An unweighted sum of the top nine relative peak areas had an apparent AUC of 0.904 (95% CI 0.825-0.983), apparent sensitivity 0.806 and apparent specificity 0.882. Conclusion: The results identify candidate features for distinguishing the sampled TS and BS groups. They do not establish independent associations, specificity against other pathogens, validated diagnostic accuracy or clinical utility.
- 2 November 2026
- Published by Springer Nature in BMC Musculoskeletal Disorders
Open AccessA novel minimally invasive metabolomic strategy for differential diagnosis of spinal infections: blood-derived metabolic signatures as candidate biomarkers distinguishing tuberculous spondylitis from brucellar spondylitis
Background: Tuberculous spondylitis (TS) and brucellar spondylitis (BS) can present with overlapping manifestations. This study assessed plasma metabolic differences between the two diagnostic groups. Methods: We conducted a retrospective case-control study of 53 patients (36 with TS and 17 with BS) enrolled between January 2021 and December 2024. Diagnoses were established using a composite reference standard integrating compatible clinical and radiological findings with microbiological and/or histopathological evidence. Untargeted ultra-high-performance liquid chromatography-tandem mass spectrometry (UHPLC-MS/MS), multivariate analyses and receiver operating characteristic (ROC) analyses were used. Candidate selection and ROC assessment used the same participants. Results: The processed dataset contained 1060 features, of which 190 met the nominal differential-screening criteria (56 with higher and 134 with lower relative peak areas in TS). Lipids and lipid-like molecules accounted for 390 of 1058 classified features (36.86%); 2 features were unclassified. No pathway reached nominal P < 0.05 among 27 tested pathways; linoleic acid metabolism had the lowest P value (0.0767). The highest individual apparent area under the ROC curve (AUC) was 0.922 for the feature annotated as 12-hydroxyjasmonic acid. An unweighted sum of the top nine relative peak areas had an apparent AUC of 0.904 (95% CI 0.825-0.983), apparent sensitivity 0.806 and apparent specificity 0.882. Conclusion: The results identify candidate features for distinguishing the sampled TS and BS groups. They do not establish independent associations, specificity against other pathogens, validated diagnostic accuracy or clinical utility.
- Open Access
- 24 December 2025
- Published by Springer Nature in European Journal of Medical Research
- Vol. 31 (1)
Comprehensive comparative analysis of explainable deep learning model for differentiation of brucellar spondylitis and tuberculous spondylitis through MRI sequences
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.
- 24 December 2025
- Published by Springer Nature in European Journal of Medical Research
- Vol. 31 (1)
Open AccessComprehensive comparative analysis of explainable deep learning model for differentiation of brucellar spondylitis and tuberculous spondylitis through MRI sequences
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.