验证心肌缺血损伤指数机器学习算法在异质人群中指导心肌梗死诊断的准确性：一项预设的探索性分析。
Validation of the myocardial-ischaemic-injury-index machine learning algorithm to guide the diagnosis of myocardial infarction in a heterogenous population: a prespecified exploratory analysis.
机构信息
British Heart Foundation Centre for Cardiovascular Science, University of Edinburgh, Edinburgh, UK; Usher Institute, University of Edinburgh, Edinburgh, UK.
British Heart Foundation Centre for Cardiovascular Science, University of Edinburgh, Edinburgh, UK.
出版信息
Lancet Digit Health. 2022 May;4(5):e300-e308. doi: 10.1016/S2589-7500(22)00025-5.
BACKGROUND
Diagnostic pathways for myocardial infarction rely on fixed troponin thresholds, which do not recognise that troponin varies by age, sex, and time within individuals. To overcome this limitation, we recently introduced a machine learning algorithm that predicts the likelihood of myocardial infarction. Our aim was to evaluate whether this algorithm performs well in routine clinical practice and predicts subsequent events.
METHODS
The myocardial-ischaemic-injury-index (MI) algorithm was validated in a prespecified exploratory analysis using data from a multi-centre randomised trial done in Scotland, UK that included consecutive patients with suspected acute coronary syndrome undergoing serial high-sensitivity cardiac troponin I measurement. Patients with ST-segment elevation myocardial infarction were excluded. MI incorporates age, sex, and two troponin measurements to compute a value (0-100) reflecting an individual's likelihood of myocardial infarction during the index visit and estimates diagnostic performance metrics (including area under the receiver-operating-characteristic curve, and the sensitivity, specificity, negative predictive value, and positive predictive value) at the computed score. Model performance for an index diagnosis of myocardial infarction (type 1 or type 4b), and for subsequent myocardial infarction or cardiovascular death at 1 year was determined using the previously defined low-probability threshold (1·6) and high-probability MI threshold (49·7). The trial is registered with ClinicalTrials.gov, NCT01852123.
FINDINGS
In total, 20 761 patients (64 years [SD 16], 9597 [46%] women) enrolled between June 10, 2013, and March 3, 2016, were included from the High-STEACS trial cohort, of whom 3272 (15·8%) had myocardial infarction. MI had an area under the receiver-operating-characteristic curve of 0·949 (95% CI 0·946-0·952) identifying 12 983 (62·5%) patients as low-probability for myocardial infarction at the pre-specified threshold (MI score <1·6; sensitivity 99·3% [95% CI 99·0-99·6], negative predictive value 99·8% [99·8-99·9]), and 2961 (14·3%) as high-probability at the pre-specified threshold (MI score ≥49·7; specificity 95·0% [94·6-95·3], positive predictive value 70·4% [68·7-72·0]). At 1 year, subsequent myocardial infarction or cardiovascular death occurred more often in high-probability patients than low-probability patients (520 [17·6%] of 2961 vs 197 [1·5%] of 12 983], p<0·0001).
INTERPRETATION
In consecutive patients undergoing serial cardiac troponin measurement for suspected acute coronary syndrome, the MI algorithm accurately estimated the likelihood of myocardial infarction and predicted subsequent adverse cardiovascular events. By providing individual probabilities the MI algorithm could improve the diagnosis and assessment of risk in patients with suspected acute coronary syndrome.
FUNDING
Medical Research Council, British Heart Foundation, National Institute for Health Research, and NHSX.
背景
心肌梗死的诊断途径依赖于固定的肌钙蛋白阈值，而这些阈值无法识别肌钙蛋白因个体年龄、性别和时间而异。为了克服这一局限性，我们最近引入了一种机器学习算法，可以预测心肌梗死的可能性。我们的目的是评估该算法在常规临床实践中的表现，并预测随后的事件。
方法
我们使用来自英国苏格兰一项多中心随机试验的前瞻性分析数据，对心肌缺血损伤指数（MI）算法进行了验证。该试验纳入了连续就诊的疑似急性冠状动脉综合征患者，他们接受了连续的高敏肌钙蛋白 I 检测。排除 ST 段抬高型心肌梗死患者。MI 结合了年龄、性别和两次肌钙蛋白测量值，计算出一个反映个体在就诊期间发生心肌梗死可能性的数值（0-100），并估计诊断性能指标（包括接受者操作特征曲线下面积，以及敏感性、特异性、阴性预测值和阳性预测值）在计算得分。使用之前定义的低概率阈值（1.6）和高概率 MI 阈值（49.7），确定指数诊断为心肌梗死（1 型或 4b 型）和随后 1 年内发生心肌梗死或心血管死亡的模型性能。该试验在 ClinicalTrials.gov 注册，NCT01852123。
结果
总共纳入了 20761 例患者（64 岁[标准差 16]，9597 例[46%]为女性），这些患者来自 2013 年 6 月 10 日至 2016 年 3 月 3 日进行的高 STEACS 试验队列，其中 3272 例（15.8%）发生了心肌梗死。MI 的受试者工作特征曲线下面积为 0.949（95%置信区间 0.946-0.952），该曲线将 12983 例（62.5%）患者确定为低概率发生心肌梗死（MI 评分<1.6；敏感性 99.3%[99.0-99.6]，阴性预测值 99.8%[99.8-99.9]），将 2961 例（14.3%）确定为高概率发生心肌梗死（MI 评分≥49.7；特异性 95.0%[94.6-95.3]，阳性预测值 70.4%[68.7-72.0]）。在 1 年内，高概率患者比低概率患者更常发生随后的心肌梗死或心血管死亡（2961 例中的 520 例[17.6%]与 12983 例中的 197 例[1.5%]，p<0.0001）。
解释
在连续接受高敏肌钙蛋白检测的疑似急性冠状动脉综合征患者中，MI 算法能够准确估计心肌梗死的可能性，并预测随后发生的不良心血管事件。通过提供个体概率，MI 算法可以改善疑似急性冠状动脉综合征患者的诊断和风险评估。
资金
医学研究理事会、英国心脏基金会、国家卫生研究院和 NHSX。
Zotero 插件