Canonical text (the exact bytes fingerprinted)

TRUVACE RECORD VERSION
record: TRV-2026-1034
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-09-09T06:07:14.212120Z
status: published
lens: g_space
sector: health
headline: Causal adiposity and clinical validation of regional fat distribution in PMOS: a multi-modal analysis integrating GBD 2021, Mendelian randomization, and machine learning
dek: Background Polyendocrine metabolic ovarian syndrome (PMOS), previously named polycystic ovary syndrome (PCOS), is the most common endocrine disorder among women of reproductive age and a leading cause of anovulatory infertility. However, the evolving global burden of PMOS and the role of adiposity, particularly regional fat distribution, remain incompletely understood. We integrated global epidemiological analysis, genetic causal inference, and clinical prediction modeling to investigate the burden and adiposity…
gain_title: XGBoost model predicted PMOS status from detailed body-composition measures with AUC 0.701 in testing, with SHAP highlighting regional fat masses as top predictors.
problem_title: (none)
trace_subject: (none)
gain_reading: XGBoost model predicted PMOS status from detailed body-composition measures with AUC 0.701 in testing, with SHAP highlighting regional fat masses as top predictors.
gain_evidence: XGBoost achieved the best predictive performance in the testing set (AUC = 0.701) | SHAP identified left-leg fat mass, trunk fat mass, percent body fat, and protein mass as the most influential predictors
problem_reading: (none)
problem_evidence: (none)
quick_read: Using GBD 2021 data, Mendelian randomization, and a clinical cohort, the study quantified global PMOS burden and tested adiposity as a causal determinant, then built machine-learning models from body-composition measures to predict PMOS, with XGBoost reaching AUC 0.701 and SHAP highlighting left-leg and trunk fat mass.

The convergence of population burden, genetic evidence, and ML prediction suggests body-composition phenotypes beyond BMI matter for PMOS risk and infertility, but predictive performance was modest and validation was internal only, leaving uncertainty about clinical utility across diverse populations and settings.
limitation: Models were only internally validated in a single clinical cohort, limiting generalizability beyond that population.
tag: Evidence-backed gain
key_points: GBD 2021 analysis estimated PMOS affected approximately 69.5 million women globally in 2021 with age-standardized prevalence 1,757.8 per 100,000. | Two-sample Mendelian randomization linked adiposity traits to PMOS, including body mass index OR 2.60 and left-leg fat mass OR 3.44, while fat-free mass showed no significant association. | Clinical ML cohort used detailed body-composition measurements to train models, with XGBoost outperforming others and SHAP used to interpret predictions.
rundown: Researchers combined GBD 2021 epidemiology, two-sample Mendelian randomization for overall and regional adiposity, and a clinical cohort with detailed body-composition data to train ML models for PMOS.

GBD results showed highest modeled incidence in the 10-14-year age group and about 12.5 million infertility cases attributable to PMOS worldwide, while MR found trunk fat mass OR 2.54 and body fat percentage OR 3.32 as risk factors.
sources:
- peer_reviewed | International Journal of Obesity | https://doi.org/10.1038/s41366-026-02208-x | 2026-09-07
prev: 0000000000000000000000000000000000000000000000000000000000000000

sha256

bd9e98378c328704b76b318d877c2b8b3bf57559780c5d7d085991e79e0df894

previous

0000000000000000000000000000000000000000000000000000000000000000