Artificial intelligence of arterial Doppler waveforms to predict major adverse outcomes among patients evaluated for peripheral artery disease

Abstract

ABSTRACTBackgroundPatients with peripheral arterial disease (PAD) are at increased risk for major adverse cardiac (MACE), limb (MALE) events and all-cause mortality. Developing tools capable of identifying those patients with PAD at greatest risk for major adverse events is the first step for outcome prevention. This study aimed to determine whether computer assisted analysis of a resting Doppler waveform using deep neural networks can accurately identify PAD patients at greatest risk for adverse outcome events.MethodsConsecutive patients (4/1/2015-12/31/2020) undergoing ankle brachial index (ABI) testing were included. Patients were randomly allocated to training, validation and testing subsets (60%/20%/20%). Deep neural networks were trained on resting posterior tibial arterial Doppler waveforms to predict MACE, MALE and all-cause mortality at 5 years. Patients were then analyzed in quartiles based on the distribution of each prediction score.ResultsAmong 11,384 total patients, 10,437 patients met study inclusion criteria (mean age 65.8±14.8 years; 40.6% female). The test subset included 2,084 patients. During 5 years of follow up, there were 447 deaths, 585 MACE and 161 MALE events. After adjusting for age, sex, and Charlson index, deep neural network analysis of the posterior tibial artery waveform provided independent prediction of death (Hazard ratio 2.45 95% confidence interval 1.79-3.36), MACE (HR 1.98, 95%CI 1.50-2.62) and MALE (HR 11.65 95%CI 5.65-24.04) at 5 years with similar results at 1 year.ConclusionAn artificial intelligence enabled analysis of a resting Doppler arterial waveform enables identification of major adverse outcomes including all-cause mortality, MACE and MALE among PAD patients.