CoreSlicer与SliceOmatic在胃癌CT体成分分析中的一致性验证
本研究回顾性纳入261例局部进展期胃腺癌患者,对比CoreSlicer(AI辅助分割+人工校准)与SliceOmatic(手动分割)在L3层面CT体成分分析中的一致性。结果显示两种软件对骨骼肌面积、皮下及内脏脂肪面积的测量一致性极佳,对肌萎缩的诊断一致性优秀,对主要病理缓解的预测效能相似。
要点
- CoreSlicer测量骨骼肌面积与SliceOmatic一致性极佳,可用于替代手动分割。
- CoreSlicer对内脏脂肪面积存在系统性高估,报告时需注意。
- 两种软件对主要病理缓解的预测AUC均为0.675,临床效能相当。
核心结论
- CoreSlicer结合人工校准在CT体成分测量中与SliceOmatic具有极佳一致性,可作为临床实践中手动分割的可靠替代方案。
研究设计与样本
- 回顾性研究,纳入261例局部进展期胃腺癌患者,分析治疗前后L3层面CT图像。
- 使用CoreSlicer(AI辅助分割+人工校准)与SliceOmatic(手动分割)两种软件进行体成分分析。
- 测量骨骼肌面积(SMA)、内脏脂肪面积(VAT)、皮下脂肪面积(SAT)及相应指数(SMI、SAI、VAI)。
关键影像表现/征象
- 骨骼肌面积治疗前ICC为0.997,治疗后ICC为0.997,两种软件间一致性极佳。
- 皮下脂肪面积治疗前ICC为0.990,治疗后ICC为0.990,一致性极佳。
- 内脏脂肪面积治疗前ICC为0.964,治疗后ICC为0.942,Bland-Altman分析显示CoreSlicer存在系统性高估。
- Bland-Altman分析显示骨骼肌面积与皮下脂肪面积的偏倚极小。
诊断标准与关键数值
- 骨骼肌面积ICC:治疗前0.997,治疗后0.997。
- 皮下脂肪面积ICC:治疗前0.990,治疗后0.990。
- 内脏脂肪面积ICC:治疗前0.964,治疗后0.942。
- 肌萎缩诊断一致性Cohen's kappa为0.923。
- 两种方法对主要病理缓解(MPR)的预测AUC均为0.675。
临床应用与阅片要点
- CoreSlicer可用于胃癌患者L3层面CT体成分的骨骼肌和皮下脂肪测量,与手动分割具有可比性。
- 使用CoreSlicer测量内脏脂肪面积时需注意其对VAT存在系统性高估。
- 肌萎缩的自动评估与手动评估诊断一致性优秀,可用于临床筛查。
局限性
- 本研究为回顾性单中心设计。
- CoreSlicer对内脏脂肪面积存在系统性高估。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
英文原文(摘要)· 点击展开
BACKGROUND: This study aimed to validate the agreement between CoreSlicer (AI-assisted segmentation with manual calibration) and SliceOmatic (manual segmentation) for CT-based body composition analysis in gastric cancer patients.
METHODS: This retrospective study included 261 patients with locally advanced gastric adenocarcinoma. Pre- and post-treatment L3-level CT images were analyzed using both software programs. Skeletal muscle area (SMA), visceral adipose tissue area (VAT), and subcutaneous adipose tissue area (SAT) were measured, and corresponding indices (SMI, SAI, VAI) were calculated. Agreement was assessed using intraclass correlation coefficients (ICC), Bland-Altman analysis, and Cohen's kappa. Multivariable logistic regression models were constructed to predict major pathological response (MPR), and model discrimination was compared using the area under the curve (AUC).
RESULTS: Excellent agreement was observed for SMA (ICC: pre-treatment 0.997, post-treatment 0.997), SAT (ICC: 0.990, 0.990), and VAT (ICC: 0.964, 0.942). Bland-Altman analysis showed minimal bias for SMA and SAT, whereas VAT was systematically overestimated by CoreSlicer. Diagnostic agreement for muscle atrophy was excellent (kappa: 0.923). Both methods showed similar predictive performance for MPR (AUC = 0.675).
CONCLUSIONS: CoreSlicer with manual calibration demonstrates excellent agreement with SliceOmatic for body composition measurements, supporting its use as a reliable alternative to manual segmentation in clinical practice.
引用
- Lu, Huang, Wang, et al. Validation of an AI‑Assisted Open‑Source software (CoreSlicer) for CT‑Based body composition analysis in locally advanced gastric Cancer: A Comparative study with SliceOmatic. Eur J Radiol. 2026. PMID: 42447772. doi:10.1016/j.ejrad.2026.113073
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。