{"id":1177,"date":"2026-02-13T22:53:28","date_gmt":"2026-02-13T22:53:28","guid":{"rendered":"https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/?p=1177"},"modified":"2026-02-13T22:53:29","modified_gmt":"2026-02-13T22:53:29","slug":"week-5-2","status":"publish","type":"post","link":"https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/2026\/02\/13\/week-5-2\/","title":{"rendered":"Week 5"},"content":{"rendered":"\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"579\" src=\"https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/wp-content\/uploads\/sites\/209\/2026\/02\/Screenshot-2026-02-10-at-3.45.11\u202fPM-1024x579.jpg\" alt=\"\" class=\"wp-image-1179\" srcset=\"https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/wp-content\/uploads\/sites\/209\/2026\/02\/Screenshot-2026-02-10-at-3.45.11\u202fPM-1024x579.jpg 1024w, https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/wp-content\/uploads\/sites\/209\/2026\/02\/Screenshot-2026-02-10-at-3.45.11\u202fPM-300x170.jpg 300w, https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/wp-content\/uploads\/sites\/209\/2026\/02\/Screenshot-2026-02-10-at-3.45.11\u202fPM-768x435.jpg 768w, https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/wp-content\/uploads\/sites\/209\/2026\/02\/Screenshot-2026-02-10-at-3.45.11\u202fPM-1536x869.jpg 1536w, https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/wp-content\/uploads\/sites\/209\/2026\/02\/Screenshot-2026-02-10-at-3.45.11\u202fPM-2048x1159.jpg 2048w, https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/wp-content\/uploads\/sites\/209\/2026\/02\/Screenshot-2026-02-10-at-3.45.11\u202fPM-371x210.jpg 371w\" sizes=\"(max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n\n<p><\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Refining Stability and Clinical Plausibility: Validating the Single-Lead ECG-FM<\/h2>\n\n\n\n<p id=\"p-rc_7472fa17c07e354d-94\">This week marked a transition from initial adaptation to rigorous validation and system stabilization<sup><\/sup>. Our focus shifted toward evaluating the fine-tuned ECG foundation model on real-world single-lead inputs and enforcing physiological constraints to ensure our outputs meet clinical standards<sup><\/sup><sup><\/sup>.+2<\/p>\n\n\n\n<p id=\"p-rc_7472fa17c07e354d-95\">Rather than just proving feasibility, we are now hardening the system\u2014moving from experimental code to a reproducible, high-performance diagnostic pipeline<sup><\/sup><sup><\/sup>.+1<\/p>\n\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\" \/>\n\n\n\n<h3 class=\"wp-block-heading\">Key Accomplishments This Week<\/h3>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Validated High-Performance Single-Lead Adaptation<\/strong> We evaluated the fine-tuned ECG-FM on duplicated single-lead inputs using real-world samples provided by our sponsor. The model achieved a strong AUROC of approximately 0.96, demonstrating that single-lead adaptation can reach performance levels comparable to dual-lead baselines.<\/li>\n\n\n\n<li><strong>Optimized Temporal Aggregation for Prediction Stability<\/strong> By comparing 10-second and 30-second inference windows, we observed that longer temporal aggregation significantly improves stability. This strategy reduces inconsistent arrhythmia predictions, providing a more reliable output for clinical review.<\/li>\n\n\n\n<li><strong>Advanced PQRST Delineation with Physiological Constraints<\/strong> We improved our post-processing by enforcing rules such as temporal ordering and minimum inter-peak intervals. By clustering nearby candidate peaks, we successfully reduced duplicate detections and began addressing over-generation issues in T-wave localization.+1<\/li>\n\n\n\n<li><strong>Refined Clinical Plausibility and Thresholding<\/strong> In collaboration with sponsor feedback, we analyzed multi-label outputs\u2014including PVC, tachycardia, and bundle branch blocks\u2014to assess their clinical plausibility. We investigated thresholding strategies to suppress low-confidence or physiologically impossible diagnoses in a deployment setting.+1<\/li>\n\n\n\n<li><strong>Initiated Model Interpretability Research<\/strong> We began exploring saliency and attribution maps to localize PVC-related regions. This work supports the debugging of both classification and fiducial detection, ensuring the model is looking at the correct features for its predictions.<\/li>\n<\/ul>\n\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\" \/>\n\n\n\n<h3 class=\"wp-block-heading\">Next Steps: Toward a Deployment-Ready System<\/h3>\n\n\n\n<p>In the coming week, we will:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Benchmark window aggregation<\/strong> (10s vs. 30s) alongside tuned confidence thresholds to quantify the trade-offs between system responsiveness and prediction stability.<\/li>\n\n\n\n<li><strong>Refine T-wave localization<\/strong> by further adjusting temporal windows and confidence filtering within the post-processing pipeline.<\/li>\n\n\n\n<li><strong>Finalize patient-level data splits<\/strong> and document preprocessing workflows to support future regulatory-oriented documentation.<\/li>\n\n\n\n<li><strong>Package the fine-tuning pipeline<\/strong> and inference scripts into a clean, reproducible repository for team sharing and review.<\/li>\n\n\n\n<li><strong>Draft a structured technical report<\/strong> detailing the single-lead adaptation approach, experimental setup, and key performance results.<\/li>\n<\/ul>\n\n\n\n<p id=\"p-rc_7472fa17c07e354d-106\">With our core performance validated and physiological constraints in place, we are moving closer to a cohesive, deployment-oriented system<sup><\/sup>.<\/p>\n\n\n\n<p>See you next week!<\/p>\n\n\n\n<p><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Refining Stability and Clinical Plausibility: Validating the Single-Lead ECG-FM This week marked a transition from initial adaptation to rigorous validation and system stabilization. Our focus shifted toward evaluating the fine-tuned ECG foundation model on real-world single-lead inputs and enforcing physiological&hellip; <a href=\"https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/2026\/02\/13\/week-5-2\/\" aria-label=\"Read \\\"Week 5\\\" class=\"read-more\">Read&nbsp;More<\/a><\/p>\n","protected":false},"author":849,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"inline_featured_image":false,"footnotes":"","_links_to":"","_links_to_target":""},"categories":[12],"tags":[],"class_list":["post-1177","post","type-post","status-publish","format-standard","hentry","category-springsemester"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/wp-json\/wp\/v2\/posts\/1177"}],"collection":[{"href":"https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/wp-json\/wp\/v2\/users\/849"}],"replies":[{"embeddable":true,"href":"https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/wp-json\/wp\/v2\/comments?post=1177"}],"version-history":[{"count":2,"href":"https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/wp-json\/wp\/v2\/posts\/1177\/revisions"}],"predecessor-version":[{"id":1183,"href":"https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/wp-json\/wp\/v2\/posts\/1177\/revisions\/1183"}],"wp:attachment":[{"href":"https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/wp-json\/wp\/v2\/media?parent=1177"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/wp-json\/wp\/v2\/categories?post=1177"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/wp-json\/wp\/v2\/tags?post=1177"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}