Division of Cardiology, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, USA
Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge, Massachusetts, USA.
Open Heart. 2022 May;9(1). doi: 10.1136/openhrt-2022-001990.
To use echocardiographic and clinical features to develop an explainable clinical risk prediction model in patients with aortic stenosis (AS), including those with low-gradient AS (LGAS), using machine learning (ML).
In 1130 patients with moderate or severe AS, we used bootstrap lasso regression (BLR), an ML method, to identify echocardiographic and clinical features important for predicting the combined outcome of all-cause mortality or aortic valve replacement (AVR) within 5 years after the initial echocardiogram. A separate hold out set, from a different centre (n=540), was used to test the generality of the model. We also evaluated model performance with respect to each outcome separately and in different subgroups, including patients with LGAS.
Out of 69 available variables, 26 features were identified as predictive by BLR and expert knowledge was used to further reduce this set to 9 easily available and input features without loss of efficacy. A ridge logistic regression model constructed using these features had an area under the receiver operating characteristic curve (AUC) of 0.74 for the combined outcome of mortality/AVR. The model reliably identified patients at high risk of death in years 2-5 (HRs ≥2.0, upper vs other quartiles, for years 2-5, p<0.05, p=not significant in year 1) and was also predictive in the cohort with LGAS (n=383, HRs≥3.3, p<0.05). The model performed similarly well in the independent hold out set (AUC 0.78, HR ≥2.5 in years 1-5, p<0.05).
In two separate longitudinal databases, ML identified prognostic features and produced an algorithm that predicts outcome for up to 5 years of follow-up in patients with AS, including patients with LGAS. Our algorithm, the Aortic Stenosis Risk (ASteRisk) score, is available online for public use.
使用超声心动图和临床特征，通过机器学习（ML）为主动脉瓣狭窄（AS）患者（包括低梯度 AS [LGAS]患者）开发一个可解释的临床风险预测模型。
在 1130 名中重度 AS 患者中，我们使用了一种 ML 方法——Bootstrap 套索回归（BLR），来识别对预测初始超声心动图后 5 年内全因死亡率或主动脉瓣置换（AVR）的复合结局有重要意义的超声心动图和临床特征。使用来自不同中心的另一个独立验证集（n=540）来测试模型的通用性。我们还分别评估了模型在不同亚组中的表现，包括 LGAS 患者。
在 69 个可用变量中，有 26 个特征通过 BLR 被确定为具有预测性，然后使用专家知识进一步将该组特征减少至 9 个易于获取且不影响疗效的输入特征。使用这些特征构建的岭逻辑回归模型对死亡率/AVR 的复合结局的受试者工作特征曲线下面积（AUC）为 0.74。该模型可靠地识别出 2-5 年内死亡风险较高的患者（2-5 年内 HR≥2.0，上四分位数比其他四分位数，p<0.05，p=第 1 年无统计学意义），并且在 LGAS 患者队列中也具有预测性（n=383，HR≥3.3，p<0.05）。该模型在独立验证集中的表现也相当出色（AUC 0.78，HR≥2.5，1-5 年，p<0.05）。
在两个独立的纵向数据库中，ML 确定了预后特征，并生成了一种算法，可预测 AS 患者长达 5 年的随访结果，包括 LGAS 患者。我们的算法，即主动脉瓣狭窄风险（ASteRisk）评分，可在线供公众使用。
