The effect of artificial intelligence-driven, computed tomography-based assessment of body composition parameters on 28-day mortality of patients with severe fever with thrombocytopenia syndrome: a retrospective cohort study.
4 affiliations
- 1Department of Radiology, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, 430022, China.
- 2Department of Infectious Diseases, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, 1277 Jiefang Avenue, Wuhan, 430022, China.
- 3DAMO Academy, Alibaba Group, Hangzhou, 310000, China.
- 4Department of Infectious Diseases, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, 1277 Jiefang Avenue, Wuhan, 430022, China. [email protected].
Abstract
Background
Severe fever with thrombocytopenia syndrome (SFTS) is a life-threatening infectious disease with high mortality. Although sarcopenia and myosteatosis are associated with adverse outcomes in multiple acute illnesses, their prognostic impact on SFTS remains unknown. This study aimed to evaluate the association between body composition parameters and 28-day mortality in SFTS patients, and to identify independent prognostic factors.Methods
In this retrospective observational single-center study, 234 SFTS patients with complete clinical data and available computed tomography (CT) images were enrolled. Body composition indices, including skeletal muscle index (SMI), subcutaneous fat index (SFI), visceral fat index (VFI), and intermuscular fat index (IMFI), were measured at the 12th thoracic (T12) and 3rd lumbar (L3) vertebral levels using a validated artificial intelligence-driven CT segmentation model. Univariable logistic regression and least absolute shrinkage and selection operator (LASSO) regression were applied to screen candidate predictors of 28-day mortality, and multivariable logistic regression was used to identify independent risk factors and construct a prognostic model. Model performance was evaluated using receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA). Kaplan-Meier curves with the log-rank test were used to compare survival across different muscle phenotypes.Results
The overall 28-day mortality rate was 13.7% (32/234). T12-IMFI was identified as an independent risk factor for 28-day mortality (adjusted odds ratio [aOR] = 1.30, 95% confidence interval [CI] 1.04-1.63, P = 0.021). A prediction model comprising age, log₁₀-transformed viral load, procalcitonin (PCT), and T12-IMFI was constructed, which exhibited satisfactory discrimination performance with an area under the ROC curve (AUC) of 0.877 (95% CI 0.805-0.950) and good calibration (Hosmer-Lemeshow P = 0.278). DCA demonstrated that the model provided superior net benefit across the clinically relevant threshold range of 0.05 to 0.68. The addition of T12-IMFI significantly improved individual risk stratification compared with the clinical-only model (net reclassification improvement [NRI] = 0.470, 95% CI 0.078- 0.862, P = 0.019). Patients with myosteatosis alone showed the highest 28-day mortality rate (26.1%), and significant differences in survival were observed among the four muscle phenotypes (log-rank P = 0.012).Conclusions
CT-defined myosteatosis, rather than sarcopenia, is independently associated with 28-day mortality in SFTS patients. T12-IMFI is a valuable prognostic biomarker, and the model incorporating T12-IMFI provides a practical tool for early risk stratification in SFTS patients at admission.Clinical trial number
Not applicable.
Related
- Determination of urinary stone composition using computed tomography using artificial intelligence.
- Agreement Between Artificial Intelligence-Based and Expert Consensus Implant Planning Using Cone-Beam Computed Tomography: A Retrospective Observational Study.
- Artificial Intelligence‐Assisted Automated Optical Coherence Tomography Analysis for Percutaneous Coronary Intervention Planning: From Plaque Analysis to Precision Lesion Preparation and Stent Optimization
- The integration of artificial intelligence into computed tomography: from segmentation to the standardisation of clinical decisions
- Association of sarcopenia with treatment tolerance in older adults with acute myeloid leukemia using artificial intelligence computed tomography–segmentation and a geriatric assessment
BibTeX
@article{yang2026effect,
title = {{The effect of artificial intelligence-driven, computed tomography-based assessment of body composition parameters on 28-day mortality of patients with severe fever with thrombocytopenia syndrome: a retrospective cohort study.}},
author = {Fan Yang and Shiyu Zhang and Shuting Liu and Jianfeng Zhang and Jinrong Yang and Dilihumaer Zhayier and Ruihan Gao and Fan Pu and Minfeng Xu and Bin Zhu and Xin Zheng and Baoju Wang},
journal = {BMC infectious diseases},
year = {2026},
month = jul,
doi = {10.1186/s12879-026-14025-8},
url = {https://europepmc.org/article/MED/42902163},
note = {CC BY-NC-ND},
}
RIS
TY - JOUR
TI - The effect of artificial intelligence-driven, computed tomography-based assessment of body composition parameters on 28-day mortality of patients with severe fever with thrombocytopenia syndrome: a retrospective cohort study.
AU - Fan Yang
AU - Shiyu Zhang
AU - Shuting Liu
AU - Jianfeng Zhang
AU - Jinrong Yang
AU - Dilihumaer Zhayier
AU - Ruihan Gao
AU - Fan Pu
AU - Minfeng Xu
AU - Bin Zhu
AU - Xin Zheng
AU - Baoju Wang
PY - 2026
DA - 2026/07/01
JO - BMC infectious diseases
DO - 10.1186/s12879-026-14025-8
UR - https://europepmc.org/article/MED/42902163
AB - Background Severe fever with thrombocytopenia syndrome (SFTS) is a life-threatening infectious disease with high mortality. Although sarcopenia and myosteatosis are associated with adverse outcomes in multiple acute illnesses, their prognostic impact on SFTS remains unknown. This study aimed to evaluate the association between body composition parameters and 28-day mortality in SFTS patients, and to identify independent prognostic factors.Methods In this retrospective observational single-center study, 234 SFTS patients with complete clinical data and available computed tomography (CT) images were enrolled. Body composition indices, including skeletal muscle index (SMI), subcutaneous fat index (SFI), visceral fat index (VFI), and intermuscular fat index (IMFI), were measured at the 12th thoracic (T12) and 3rd lumbar (L3) vertebral levels using a validated artificial intelligence-driven CT segmentation model. Univariable logistic regression and least absolute shrinkage and selection operator (LASSO) regression were applied to screen candidate predictors of 28-day mortality, and multivariable logistic regression was used to identify independent risk factors and construct a prognostic model. Model performance was evaluated using receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA). Kaplan-Meier curves with the log-rank test were used to compare survival across different muscle phenotypes.Results The overall 28-day mortality rate was 13.7% (32/234). T12-IMFI was identified as an independent risk factor for 28-day mortality (adjusted odds ratio [aOR] = 1.30, 95% confidence interval [CI] 1.04-1.63, P = 0.021). A prediction model comprising age, log₁₀-transformed viral load, procalcitonin (PCT), and T12-IMFI was constructed, which exhibited satisfactory discrimination performance with an area under the ROC curve (AUC) of 0.877 (95% CI 0.805-0.950) and good calibration (Hosmer-Lemeshow P = 0.278). DCA demonstrated that the model provided superior net benefit across the clinically relevant threshold range of 0.05 to 0.68. The addition of T12-IMFI significantly improved individual risk stratification compared with the clinical-only model (net reclassification improvement [NRI] = 0.470, 95% CI 0.078- 0.862, P = 0.019). Patients with myosteatosis alone showed the highest 28-day mortality rate (26.1%), and significant differences in survival were observed among the four muscle phenotypes (log-rank P = 0.012).Conclusions CT-defined myosteatosis, rather than sarcopenia, is independently associated with 28-day mortality in SFTS patients. T12-IMFI is a valuable prognostic biomarker, and the model incorporating T12-IMFI provides a practical tool for early risk stratification in SFTS patients at admission.Clinical trial number Not applicable.
SN - 1471-2334
ER -