{"id":1187,"date":"2026-02-21T03:51:40","date_gmt":"2026-02-21T03:51:40","guid":{"rendered":"https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/?p=1187"},"modified":"2026-02-21T03:53:03","modified_gmt":"2026-02-21T03:53:03","slug":"week-6-2","status":"publish","type":"post","link":"https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/2026\/02\/21\/week-6-2\/","title":{"rendered":"Week 6"},"content":{"rendered":"\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"549\" src=\"https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/wp-content\/uploads\/sites\/209\/2026\/02\/Screenshot-2026-02-20-at-9.48.53\u202fPM-1024x549.jpg\" alt=\"\" class=\"wp-image-1191\" srcset=\"https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/wp-content\/uploads\/sites\/209\/2026\/02\/Screenshot-2026-02-20-at-9.48.53\u202fPM-1024x549.jpg 1024w, https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/wp-content\/uploads\/sites\/209\/2026\/02\/Screenshot-2026-02-20-at-9.48.53\u202fPM-300x161.jpg 300w, https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/wp-content\/uploads\/sites\/209\/2026\/02\/Screenshot-2026-02-20-at-9.48.53\u202fPM-768x412.jpg 768w, https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/wp-content\/uploads\/sites\/209\/2026\/02\/Screenshot-2026-02-20-at-9.48.53\u202fPM-1536x823.jpg 1536w, https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/wp-content\/uploads\/sites\/209\/2026\/02\/Screenshot-2026-02-20-at-9.48.53\u202fPM-2048x1097.jpg 2048w, https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/wp-content\/uploads\/sites\/209\/2026\/02\/Screenshot-2026-02-20-at-9.48.53\u202fPM-392x210.jpg 392w\" sizes=\"(max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n\n<p><\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Enhancing Interpretability and Broadening Horizons: Advancing the Single-Lead ECG-FM<\/h2>\n\n\n\n<p id=\"p-rc_fc53abe37cb4319d-77\">This week represented a major leap from validating performance to establishing deep system interpretability and planning for expanded clinical capabilities. While previous milestones focused on adaptation, we are now pulling back the curtain on model behavior\u2014using advanced visualization tools to ensure our AI aligns with clinical intuition while preparing for even more complex diagnostic tasks.<\/p>\n\n\n\n<p id=\"p-rc_fc53abe37cb4319d-78\">Rather than just reporting high numbers, we are now investigating the &#8220;why&#8221; behind the predictions and hardening our pipeline for real-time application.<\/p>\n\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\" \/>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>Key Accomplishments This Week<\/strong><\/h2>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Integrated Model Interpretability Tools<\/strong> We successfully implemented saliency maps using attention weights to visualize where the model focuses during ECG classification. We observed the model correctly highlighting clinical features, such as deep S-waves and high T-waves, when identifying conditions like left bundle branch block.<\/li>\n\n\n\n<li><strong>Benchmarked High-Performance Single-Lead Accuracy<\/strong> The team achieved exceptional AUROC scores of approximately 0.98 for PVC and Tachycardia detection using our fine-tuned single-lead data. We also identified a specific performance gap in Infarction detection (.73 vs. the original .92), prompting a new strategy to fine-tune on Lead II data to capture more distinct clinical markers.<\/li>\n\n\n\n<li><strong>Refined PQRST Delineation and Post-Processing<\/strong> We enhanced our U-Net model by incorporating ResNet and DCN 1D layers and resolved previous stability issues by integrating a missing script for QRS peak derivation. New post-processing techniques\u2014including moving averages and minimum distance constraints\u2014have significantly refined our peak-picking cleanliness.<\/li>\n\n\n\n<li><strong>Initiated Research and Technical Documentation<\/strong> As we move toward a final product, we have begun drafting a formal research paper. This documentation follows the structure of leading foundation model references, detailing our specific architecture, loss functions, and hyperparameters to ensure our work is scientifically reproducible.<\/li>\n<\/ul>\n\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\" \/>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>Next Steps: Toward a Unified Clinical Product<\/strong><\/h2>\n\n\n\n<p>In the coming week, we will:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Validate Data Integrity and Rule Out Leakage<\/strong>: Perform a rigorous patient-level review of our testing sets to verify that our high accuracy scores are not influenced by data leakage.<\/li>\n\n\n\n<li><strong>Fine-Tune for Ejection Fraction (EF)<\/strong>: Utilize a new sponsor-provided dataset of 200\u2013300 patients to train the model on categorical heart-pumping efficiency thresholds.<\/li>\n\n\n\n<li><strong>Experiment with Multi-Lead Training<\/strong>: Test a hybrid approach using both Lead I and Lead II data to increase sample volume and model robustness.<\/li>\n\n\n\n<li><strong>Consolidate System Integration<\/strong>: Merge our classification, fiducial detection, and saliency mapping into a single, unified application.<\/li>\n\n\n\n<li><strong>Prepare for Real-Time Presentation<\/strong>: Develop the backend logic to process binary files from wearable devices instantly, moving us from offline analysis to a live diagnostic demo.<\/li>\n<\/ul>\n\n\n\n<p id=\"p-rc_fc53abe37cb4319d-88\">By bridging the gap between raw performance and clinical interpretability, we are ensuring that our single-lead system is not only accurate but also trustworthy for the medical professionals who will use it.<\/p>\n\n\n\n<p>See you next week!<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Enhancing Interpretability and Broadening Horizons: Advancing the Single-Lead ECG-FM This week represented a major leap from validating performance to establishing deep system interpretability and planning for expanded clinical capabilities. While previous milestones focused on adaptation, we are now pulling back&hellip; <a href=\"https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/2026\/02\/21\/week-6-2\/\" aria-label=\"Read \\\"Week 6\\\" 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-1187","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\/1187"}],"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=1187"}],"version-history":[{"count":3,"href":"https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/wp-json\/wp\/v2\/posts\/1187\/revisions"}],"predecessor-version":[{"id":1197,"href":"https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/wp-json\/wp\/v2\/posts\/1187\/revisions\/1197"}],"wp:attachment":[{"href":"https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/wp-json\/wp\/v2\/media?parent=1187"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/wp-json\/wp\/v2\/categories?post=1187"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.ippd.ufl.edu\/blogs\/ay2526team03\/wp-json\/wp\/v2\/tags?post=1187"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}