Department of Radiology, TongDe Hospital of ZheJiang Province, No. 234, Gucui Road, Hangzhou, 310012, Zhejiang, China.
Department of Radiology, The First Affiliated Hospital of Nanjing Medical University, No. 300, Guangzhou Road, Nanjing, 210009, Jiangsu, China.
Abdom Radiol (NY). 2019 Sep;44(9):3019-3029. doi: 10.1007/s00261-019-02098-w.
Controversy still exists on the optimal surgical resection for potentially curable gastric cancer (GC). Use of radiologic evaluation and machine learning algorithms might predict extent of lymphadenectomy to limit unnecessary surgical treatment. We purposed to design a machine learning-based clinical decision-support model for predicting extent of lymphadenectomy (D1 vs. D2) in local advanced GC.
Clinicoradiologic features available from routine clinical assignments in 557 patients with GCs were retrospectively interpreted by an expert panel blinded to all histopathologic information. All patients underwent surgery using standard D2 resection. Decision models were developed with a logistic regression (LR), support vector machine (SVM) and auto-encoder (AE) algorithm in 371 training and tested in 186 test data, respectively. The primary end point was to measure diagnostic performance of decision model and a Japanese gastric cancer treatment guideline version 4th (JPN 4th) criteria for discriminate D1 (pT1 + pN0) versus D2 (≥ pT1 + ≥ pN1) lymphadenectomy.
The decision model with AE analysis produced highest area under ROC curve (train: 0.965, 95% confidence interval (CI) 0.948-0.978; test: 0.946, 95% CI 0.925-0.978), followed by SVM (train: 0.925, 95% CI 0.902-0.944; test: 0.942, 95% CI 0.922-0.973) and LR (train: 0.886, 95% CI 0.858-0.910; test: 0.891, 95% CI 0.891-0.952). By this improvement, overtreatment was reduced from 21.7% (121/557) by treat-all pattern, to 15.1% (84/557) by JPN 4th criteria, and to 0.7-0.9% (4-5/557) by the new approach.
The decision model with machine learning analysis demonstrates high accuracy for identifying patients who are candidates for D1 versus D2 resection. Its approximate 14-20% improvements in overtreatment compared to treat-all pattern and JPN 4th criteria potentially increase the number of patients with local advanced GCs who can safely avoid unnecessary lymphadenectomy.
对于潜在可治愈的胃癌（GC），最佳手术切除范围仍存在争议。放射学评估和机器学习算法的使用可能有助于预测淋巴结清扫范围，以限制不必要的手术治疗。我们旨在设计一种基于机器学习的临床决策支持模型，用于预测局部晚期 GC 的淋巴结清扫范围（D1 与 D2）。
回顾性分析了 557 例 GC 患者的临床影像学特征，这些特征均由一个专家小组从常规临床任务中获得，该专家小组对所有组织病理学信息均不知情。所有患者均采用标准的 D2 切除术进行手术。使用逻辑回归（LR）、支持向量机（SVM）和自动编码器（AE）算法分别在 371 例训练数据和 186 例测试数据中建立决策模型。主要终点是衡量决策模型和日本胃癌治疗指南第 4 版（JPN 4th）标准对 D1（pT1+ pN0）与 D2（≥ pT1+ ≥ pN1）淋巴结清扫的诊断性能。
AE 分析的决策模型产生了最高的 ROC 曲线下面积（训练：0.965，95%置信区间 [CI] 0.948-0.978；测试：0.946，95% CI 0.925-0.978），其次是 SVM（训练：0.925，95% CI 0.902-0.944；测试：0.942，95% CI 0.922-0.973）和 LR（训练：0.886，95% CI 0.858-0.910；测试：0.891，95% CI 0.891-0.952）。通过这种改进，与全治疗模式相比，过度治疗从 21.7%（121/557）减少到 JPN 4 版标准的 15.1%（84/557），再减少到新方法的 0.7-0.9%（4-5/557）。
基于机器学习分析的决策模型在识别适合 D1 与 D2 切除的患者方面具有很高的准确性。与全治疗模式和 JPN 4 版标准相比，其在过度治疗方面提高了约 14-20%，这可能会增加可以安全避免不必要淋巴结清扫的局部晚期 GC 患者数量。
