College of Computer Science and Engineering at Yanbu, Taibah University, Medina, KSA, Saudi Arabia; Department of Computer Science, The University of Manchester, Manchester, UK.
Division of Informatics, Imaging and Data Sciences, School of Health Sciences, The University of Manchester, Manchester, UK.
J Electrocardiol. 2023 Nov-Dec;81:218-223. doi: 10.1016/j.jelectrocard.2023.09.012. Epub 2023 Oct 7.
Drug-induced QT-prolongation increases the risk of TdP arrhythmia attacks and sudden cardiac death. However, measuring the QT-interval and determining a precise cut-off QT/QTc value that could put a patient at risk of TdP is challenging and influenced by many factors including female sex, drug-free baseline, age, genetic predisposition, and bradycardia.
This paper presents a novel approach for intuitively and visually monitoring QT-prolongation showing a potential risk of TdP, which can be adjusted according to patient-specific risk factors, using a pseudo-coloring technique and explainable artificial intelligence (AI).
We extended the development and evaluation of an explainable AI-based technique- visualized using pseudo-color on the ECG signal, thus intuitively 'explaining' how its decision was made -to detect QT-prolongation showing a potential risk of TdP according to a cut-off personalized QTc value (using Bazett's ∆QTc > 60 ms relative to drug-free baseline and Bazett's QTc > 500 ms as examples), and validated its performance using a large number of ECGs (n = 5050), acquired from a clinical trial assessing the effects of four known QT-prolonging drugs versus placebo on healthy subjects. We compared this new personalized approach to our previous study that used a more general approach using the QT-nomogram.
The explainable AI-based algorithm can accurately detect QT-prolongation when adjusted to a personalized patient-specific cut-off QTc value showing a potential risk of TdP. Using ∆QTc > 60 ms relative to drug-free baseline and QTc > 500 ms as examples, the algorithm yielded a sensitivity of 0.95 and 0.79, and a specificity of 0.95 and 0.98, respectively. We found that adjusting pseudo-coloring according to Bazett's ∆QTc > 60 ms relative to a drug-free baseline personalized to each patient provides better sensitivity than using Bazett's QTc > 500 ms, which could underestimate a potentially clinically significant QT-prolongation with bradycardia.
药物引起的 QT 延长会增加尖端扭转型室性心动过速（TdP）心律失常发作和心源性猝死的风险。然而，测量 QT 间期并确定精确的 QT/QTc 截断值，以评估患者发生 TdP 的风险，是具有挑战性的，这受到许多因素的影响，包括女性、无药物基线、年龄、遗传易感性和心动过缓。
本文提出了一种新的方法，通过伪彩色技术和可解释的人工智能（AI）直观地监测 QT 延长，以显示 TdP 的潜在风险，并根据患者的特定危险因素进行调整。
我们扩展了一种基于可解释 AI 的技术的开发和评估——在心电图信号上使用伪彩色直观地显示——从而直观地“解释”其决策是如何做出的——根据个性化的 QTc 截断值（例如，Bazett 的 ∆QTc>60ms 相对于无药物基线和 Bazett 的 QTc>500ms）检测显示 TdP 潜在风险的 QT 延长，并使用大量从评估四种已知 QT 延长药物与安慰剂对健康受试者的影响的临床试验中获得的心电图（n=5050）验证其性能。我们将这种新的个性化方法与我们之前使用 QT 诺模图的更通用方法进行了比较。
基于可解释 AI 的算法可以在调整为显示 TdP 潜在风险的个性化患者特定截断 QTc 值时准确检测 QT 延长。使用 Bazett 的 ∆QTc>60ms 相对于无药物基线和 QTc>500ms 作为示例，该算法的灵敏度分别为 0.95 和 0.79，特异性分别为 0.95 和 0.98。我们发现，根据 Bazett 的 ∆QTc>60ms 相对于每个患者的无药物基线进行伪彩色调整提供了比使用 Bazett 的 QTc>500ms 更好的灵敏度，后者可能低估伴有心动过缓的潜在临床显著 QT 延长。
