ClinGeX agent lifts concordance on held-out clinical genomic cases
Original title: Advancing Clinical Genomic Analysis with a Skill-Centred Agent
On 223 held-out cases it reached 71.7% task concordance versus 45.7% for the strongest comparator.
Researchers built ClinGeX, a large-language-model agent that composes 105 curated executable skills through task-specific retrieval and dependency-aware orchestration for clinical genomic analysis. On 223 temporally held-out benchmark cases it reached 71.7% macro-averaged task-level concordance with expert classifications, against 45.7% for the strongest comparator. In a counterbalanced study, 45 professionals assessed 135 cases under routine, DeepSeek-assisted and ClinGeX-assisted conditions; ClinGeX raised raw concordance and cut reported interpretation time. An exploratory one-month non-interventional review drew feedback from 11 professionals on 96 self-selected episodes.
Unconfirmed: Real-world time savings and conclusion adoption are self-reported, not independently measured.
- ClinGeX composes 105 curated executable skills.
- Assisted raw concordance was 51.1%, 57.8% and 64.4% across three scenarios.
- Comparator conditions were 15.6%, 31.1% and 20.0%.
- Eleven professionals reviewed 96 self-selected analysis episodes.
- The real-world assessment was non-interventional and exploratory.
From the original article
1 min read at arxiv.orgClinical genomic analysis requires experts to integrate evolving guidelines, databases, and computational evidence. As sequencing has become increasingly accessible, this complex and labor-intensive process has become a major bottleneck to the timely use of genomic information in clinical decision-making.
By Qiaoyu Zheng · arxiv.org
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.