biostudies-literatureDoudesis DBritish Heart FoundationMedical Research CouncilNational Institute for Health Research (NIHR)e300-e308https://www.ebi.ac.uk/biostudies/studies/S-EPMC9052331biostudies-literatureUnknown4(5)<h4>Background</h4>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.<h4>Methods</h4>The myocardial-ischaemic-injury-index (MI³) 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 withThe Lancet. Digital healthValidation of the myocardial-ischaemic-injury-index machine learning algorithm to guide the diagnosis of myocardial infarction in a heterogenous population: a prespecified exploratory analysis.PMC9052331FS/CRTF/21/24273RG/20/10/34966BCDSA/100003SP/12/10/29922CH/09/002/26360CH/F/21/90010MR/N013166/1AI_AWARD02322RE/18/5/34216FS/18/25/33454MR/V007017/1Ford IWeir CJAndrews JPSadat ITaggart CDoudesis DMcAllister DAMacDonald CStrachan FEMcKinlay LFischbacher CLinksted PFox KAYang JVallejos CASandeman DWalker SBerry CParker RAChapman ARMaguire DYoung SAmoils SMendusic FLeslie SJAnand ACharles HStables CLNorrie JHautvast MTsanas AStevens JMalo JStewart SDHung JMcPherson JAlexander BShah ASApple FSWalker AHarkess RLowry MTFindlay ITuck CMills NLAnwar MSHigh-STEACS InvestigatorsPickering JWReid AMarshall LCruikshank AFujisawa TDuncan CCollinson POFinlay FThan MPBularga AKimenai DMNewby DEMoss AFerry AVStirling LAdamson PDShah ASVGray ACroal BLKeerie CLee KKWereski RArmstrong RWackett TfalseValidation of the myocardial-ischaemic-injury-index machine learning algorithm to guide the diagnosis of myocardial infarction in a heterogenous population: a prespecified exploratory analysis.<h4>Background</h4>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.<h4>Methods</h4>The myocardial-ischaemic-injury-index (MI³) 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 with2022-01-01T00:00:00Z2022 May2026-05-10T00:50:53.944Z2024-11-15T17:59:07.623ZS-EPMC90523313546168910.1016/S2589-7500(22)00025-5