AbstractInherited dyslipidemias, including familial hypercholesterolemia (FH) and sitosterolemia (STSL), present overlapping clinical manifestations but require distinct therapeutic interventions. Current diagnosis relies on genetic sequencing and mass spectrometry-based sterol profiling, which are accurate but costly, time-consuming, and unsuitable for large-scale first-line screening. Infrared (IR) microspectroscopy offers a rapid, label-free alternative for biochemical phenotyping; however, its clinical application is challenged by the spatial heterogeneity of biological samples and spectral variability. Here, we present an IR microspectroscopic phenotyping framework for rapid screening and etiological stratification of inherited dyslipidemias. The framework combines multi-pixel IR microspectroscopy, batch-effect correction, and participant-level spectral integration to achieve reproducible cross-batch phenotyping while preserving disease-associated biochemical signatures. Quantitative models predicted FH-associated lipid biomarkers, whereas qualitative classifiers identified STSL-associated sterol biomarkers. A cascade decision framework integrating IR-based phenotypic classification with biomarker profiling further enabled etiological discrimination among healthy controls, FH, and STSL. The framework achieved an average participant-level sensitivity of approximately 93% and yielded detection rates of 97.7% for FH and 85.7% for STSL, compared with 93.2% and 60.7%, respectively, with conventional clinical screening. Co-importance and consensus feature analyses identified 2837 cm−1 as a prominent discriminatory spectral feature. Measurements of β-sitosterol reference standards revealed a characteristic absorption band near 2834 cm−1, supporting a possible association between this spectral region and β-sitosterol-related biochemical variation. This internally validated proof-of-concept study supports the potential of IR microspectroscopy for rapid etiological stratification of inherited dyslipidemias, complementing genetic sequencing and mass spectrometry.

Precision phenotyping combining audiometric data and genetics predicts sensorineural hearing loss far more accurately than standard billing codes.

A new end-to-end framework extracts and ranks rare disease terms directly from clinical text to streamline undiagnosed case reviews.

npj Digital Medicine - An artificial intelligence framework for end-to-end rare disease phenotyping from clinical notes using large language models

npj Digital Medicine - An artificial intelligence framework for end-to-end rare disease phenotyping from clinical notes using large language models

The authors are grateful to the patients for participating in the Undiagnosed Diseases Network. This work was supported in part by the National Institutes of Health Common Fund, grant 15-HG-0130 from the National Human Genome Research Institute, U01NS134349 from the National Institute of Neurological Disorders and Stroke, R00LM014429 from the National Library of Medicine, T32GM082773 from the National Institute of General Medical Sciences, and the Potocsnak Center for Undiagnosed and Rare Disorders. The NIH Undiagnosed Diseases Program is supported by the Intramural Research Program of the National Human Genome Research Institute. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article...

What can more than 100,000 medical scans reveal about the fat that burns energy, and why are researchers examining an everyday blood pressure medication? Study: Artificial intelligence-enabled clinical phenotyping of human brown fat. Image Credit: Love Employee / Shutterstock In a recent study published in the journal Nature, researchers used artificial intelligence (AI) and large clinical datasets to systematically map the clinical phenotypes associated with human brown adipose tissue (BAT). The researchers analyzed 18F-fluorodeoxyglucose positron emission tomography-computed tomography (18F-FDG-PET-CT) scans from 101,243 adult patients to examine associations between brown fat and health outcomes. Detectable BAT was associated with a lower prevalence of cardiometabolic and psychiatric conditions, while higher measured activity in individual BAT depots was linked to more favorable cardiometabolic outcomes. The study’s pharmacological profiling identified the common antihypertensive drug losartan as a candidate for further investigation. Experiments in cultured fat cells and mice showed increased thermogenic activity and energy expenditure, but the drug’s effects on human BAT have not been tested in clinical trials. Background Decades of research have established that mammalian adipose tissue plays a physiologically important role in maintaining systemic energy balance. In humans, adipose tissue primarily consists of white adipose tissue (WAT), which stores calories as...

What can more than 100,000 medical scans reveal about the fat that burns energy, and why are researchers examining an everyday blood pressure medication? Study: Artificial intelligence-enabled clinical phenotyping of human brown fat. Image Credit: Love Employee / Shutterstock In a recent study published in the journal Nature, researchers used artificial intelligence (AI) and large clinical datasets to systematically map the clinical phenotypes associated with human brown adipose tissue (BAT).

Artificial intelligence-enabled quantification and analyses of brown adipose tissue (BAT) activity uncover links between BAT and a healthier metabolic profile along with lower rates of psychiatric dis...

Artificial intelligence-enabled quantification and analyses of brown adipose tissue (BAT) activity uncover links between BAT and a healthier metabolic profile along with lower rates of psychiatric dis...

This retrospective case–control study followed institutional guidelines and was approved by the Institutional Review Board of Memorial Sloan Kettering Cancer Center (MSK), protocol number 17-581. The requirement for informed consent was waived because the study involved secondary analysis of previously acquired PET–CT imaging and clinical data, posed no more than minimal risk to participants and could not be conducted practically with individual informed consent. All patient data were deidentified before use. We previously established a cohort of patients with BAT by analysing 18F-FDG-PET–CT scans performed at MSK from June 2009 through March 20184. Patients who were positive for BAT in at least one scan were termed BAT+, whereas the remaining patients were grouped as BAT−. To expand this cohort, we established a new Extension BAT Cohort by following a similar strategy. We reviewed all 18F-FDG-PET–CT scans performed at MSK from patients aged 18 years and above from April 2018 through July 2024. People who were positive for BAT on at least one scan were termed BAT+, whereas the remaining patients were grouped as BAT− (Extended Data Fig. 1a and Supplementary Fig. 2). For each patient in both groups, an index scan was designated: in patients with BAT reported on at...

Artificial intelligence-enabled quantification and analyses of brown adipose tissue (BAT) activity uncover links between BAT and a healthier metabolic profile along with lower rates of psychiatric dis...