<?xml version="1.0" encoding="UTF-8"?><oembed><type>video</type><version>1.0</version><html>&lt;iframe src=&quot;https://www.loom.com/embed/70fb5b77a5ea4221b25fca637da47a3e&quot; frameborder=&quot;0&quot; width=&quot;1990&quot; height=&quot;1492&quot; webkitallowfullscreen mozallowfullscreen allowfullscreen&gt;&lt;/iframe&gt;</html><height>1492</height><width>1990</width><provider_name>Loom</provider_name><provider_url>https://www.loom.com</provider_url><thumbnail_height>1492</thumbnail_height><thumbnail_width>1990</thumbnail_width><thumbnail_url>https://cdn.loom.com/sessions/thumbnails/70fb5b77a5ea4221b25fca637da47a3e-b0a8510f87e7f37c.gif</thumbnail_url><duration>561.64</duration><title>TrialMatch AI Clinical Trial Eligibility Screener Demo</title><description>This Loom demonstrates TrialMatch AI, a clinical trial eligibility screener that matches patients to oncology protocols in minutes. It explains that only about 5% of cancer patients enroll in trials while around 80% of trials fall behind, and that current matching requires manual line-by-line review of 40 to 80 page protocols. The presenter uploads a real xCure protocol PDF and screens four patient scenarios, showing eligible outcomes with per-criterion rationales, geographic ineligibility for patients not getting care in the U.S., a need-review result for insufficient data when biopsy confirmation is missing, and an eligibility edge case using a legally authorized representative for a 9-year-old. The tool is presented as not using protocol-invented thresholds and as flagging when data is unclear.</description></oembed>