Machine learning algorithms successfully identified subtle cardiac changes in pre-symptomatic carriers of the TTR Val142Ile variant, which affects 4% of African Americans and confers 40-60% lifetime risk of developing cardiac amyloidosis. The Random Forest model achieved 76% discrimination accuracy using 15 echocardiographic features, including relative apical sparing and inferolateral strain patterns, despite no individual measurements showing significant differences between 49 carriers and 45 matched controls. This represents a meaningful advance in precision cardiology for inherited amyloidosis. Early detection could enable preventive interventions before irreversible cardiac damage occurs, potentially transforming outcomes for the estimated 1.6 million Americans carrying this variant. The approach addresses a critical clinical gap, as current guidelines lack tools for risk-stratifying pre-symptomatic carriers identified through expanding genetic screening programs. However, the small cohort size and single-variant focus limit generalizability, and external validation remains incomplete in this preprint awaiting peer review. While promising for personalized cardiac surveillance, larger multi-center studies across diverse TTR variants will be essential before clinical implementation.
Machine Learning Detects Pre-Symptomatic Heart Changes in TTR Gene Carriers
📄 Based on research published in medRxiv preprint
Read the original research →⚠️ This is a preprint — it has not yet been peer-reviewed. Results should be interpreted with caution and may change following peer review.
For informational, non-clinical use. Synthesized analysis of published research — may contain errors. Not medical advice. Consult original sources and your physician.