Performance metrics of AI-enhanced single lead EKG maintained after entry of organised clustered data

Abstract Background Our experience in creating innovative Artificial Intelligence-guided single lead EKG methodologies for ST-Elevation Myocardial Infarction (STEMI) detection within complex records has been previously validated. Purpose By expanding the intricate variables of our tested algorithm input, we seek to further improve STEMI detecting tool. Methods 11,567 12-lead (10-s length, 500 Hz sample frequency) derived from Latin America Telemedicine Infarct Network database April 2014 December 2019. From these records, included following balanced classes: angiographically confirmed and unconfirmed (divided by wall affected), branch blocks, non-specific ST-T changes, normal, abnormal (Remaining 200+ CPT codes). Cardiologist annotations ensured precision (Ground truth). Determined classes were “STEMI” “Not-STEMI”. A 1-D Convolutional Neural model was trained each with dataset proportions 90/10, respectively. The last dense layer outputs a probability record being STEMI/Not-STEMI. analysis also performance metrics false-negative reports. Results Overall, most promising Single V2 (91.2% Accuracy, 89.6% Sensitivity, 92.9% Specificity). 55% false negatives inferior (Table 1). Conclusion Appreciable progress new methodology compared previous experiences AI-guided Lead detection, especially V2. performing thorough reports, aspire identify potential areas weakness which will become focus future ventures. Funding Acknowledgement Type funding sources: None.