A study of 853 adults aged 20 to 94 years found antibody architecture broadly preserved, with age-associated signals concentrated in focal targets.A study of 853 adults aged 20 to 94 years found antibody architecture broadly preserved, with age-associated signals concentrated in focal targets. In a preprint, researchers analyzed clinically screened Chinese adults aged 20 to 94 years to profile antigen-resolved humoral immunity across…
People with intramuscular fat fractions above 10% showed significantly lower paraspinal muscle mass and strength than those with levels below 5%.People with intramuscular fat fractions above 10% showed significantly lower paraspinal muscle mass and strength than those with levels below 5%. In people with spinal sarcopenia, researchers found that the intramuscular fat fraction, rather than perimuscular fat fraction, was an…
An XGBoost model reached an AUC of 0.782 in an external cohort of 2,066 older adults to predict basic activities of daily living disability.An XGBoost model reached an AUC of 0.782 in an external cohort of 2,066 older adults to predict basic activities of daily living disability. In a study of older adults aged 65 and older, researchers developed and externally validated a machine learning model to predict disability risk.
In individuals with conventional zero scores, AI-detected calcium tracked a 7.7% 20-year coronary heart disease incidence compared with 3.8% in those…In individuals with conventional zero scores, AI-detected calcium tracked a 7.7% 20-year coronary heart disease incidence compared with 3.8% in those without. In a pooled analysis of 3,965 human participants from the Multi-Ethnic Study of Atherosclerosis and the Framingham Heart Study with…
In 70,473 adults followed for up to 10 years, 210 wearable activity phenotypes were linked to 406 diseases, with step intensity and circadian rhythm…In 70,473 adults followed for up to 10 years, 210 wearable activity phenotypes were linked to 406 diseases, with step intensity and circadian rhythm showing prominent ties. In a cohort of 70,473 human participants followed for up to 10 years, researchers analyzed 11.8 million person-hours…
Tested across six dataset and task conditions, the bilingual framework outperformed three baselines while directly generating natural-language…Tested across six dataset and task conditions, the bilingual framework outperformed three baselines while directly generating natural-language explanations for clinicians. In human speech datasets from individuals with Alzheimer's disease and mild cognitive impairment, researchers…
In data from 720,000 people, the self-evolving system evaluated 2,031 causal questions and identified blood pressure as a convergence node for…In data from 720,000 people, the self-evolving system evaluated 2,031 causal questions and identified blood pressure as a convergence node for vascular outcomes. In data from over 720,000 human participants across 48 countries, researchers tested an automated discovery platform called…
Trained on 1,088 participants, the machine learning framework predicted flortaucipir-PET with R² values of 0.34 to 0.71 and classified tau stages with…Trained on 1,088 participants, the machine learning framework predicted flortaucipir-PET with R² values of 0.34 to 0.71 and classified tau stages with 79% to 95% accuracy. In a study of 1,088 human participants spanning normal aging and Alzheimer’s disease and related disorders…
Across six human cohorts and a 2.3-year follow-up, neural networks identified brain structural patterns tied to cardiometabolic and pulmonary…Across six human cohorts and a 2.3-year follow-up, neural networks identified brain structural patterns tied to cardiometabolic and pulmonary conditions. In human participants from six independent cohorts, researchers applied deep learning to T1-weighted MRI scans to isolate structural…
In models trained on 53,026 UK Biobank participants, 13 of 15 incident disease models maintained performance across FinnGen and All of Us cohorts.In models trained on 53,026 UK Biobank participants, 13 of 15 incident disease models maintained performance across FinnGen and All of Us cohorts. In a preprint analyzing human data across three major cohorts, researchers tested how well blood proteomic disease risk models transfer across…
In 81,318 UK Biobank adults, vertebral bone marrow fat rose 2.95% to 3.83% per decade across spinal regions, with higher levels in females.In 81,318 UK Biobank adults, vertebral bone marrow fat rose 2.95% to 3.83% per decade across spinal regions, with higher levels in females. In a retrospective study of 81,318 UK Biobank participants, researchers mapped bone marrow adipose tissue across the human spine.
Pretrained on 10 million images, the model assessed 62 biomarkers across eight physiological systems, reaching a median Pearson's r of 0.172 across…Pretrained on 10 million images, the model assessed 62 biomarkers across eight physiological systems, reaching a median Pearson's r of 0.172 across four cohorts. In human datasets, researchers developed MedicalFaceFound, a facial foundation model trained on over 10 million images to…
Algorithm scores correlated strongly with expert assessments at r = 0.99, identifying moderate-to-severe aortic stenosis with 0.92 sensitivity and…Algorithm scores correlated strongly with expert assessments at r = 0.99, identifying moderate-to-severe aortic stenosis with 0.92 sensitivity and 0.98 specificity. In nongated chest CT scans from patients across 33 sites in the United States and Brazil, an automated deep learning…
Adding automated CT biomarkers to an ICD-based frailty score raised one-year mortality prediction AUROC from 0.75 to 0.79 in 7,638 surgery patients.Adding automated CT biomarkers to an ICD-based frailty score raised one-year mortality prediction AUROC from 0.75 to 0.79 in 7,638 surgery patients. In a retrospective cohort of 7,672 patients undergoing non-emergent surgery at a tertiary academic center, researchers evaluated whether…
A reinforcement learning model alerted to Alzheimer's disease 20.9 months before diagnosis and autism 8.7 months before diagnosis, both at 90%…A reinforcement learning model alerted to Alzheimer's disease 20.9 months before diagnosis and autism 8.7 months before diagnosis, both at 90% specificity. In clinical electronic health record cohorts, researchers developed a model-free reinforcement learning framework to optimize the…
Across eight external cohorts, the score yielded C-indices of 0.67 to 0.86, outperforming recalibrated conventional comorbidity indices by absolute…Across eight external cohorts, the score yielded C-indices of 0.67 to 0.86, outperforming recalibrated conventional comorbidity indices by absolute gains of 0.12 to 0.21. In 770,439 adults from nine international longitudinal cohorts, researchers evaluated a large-language-model-based…
In people with osteoporosis, higher undercarboxylated osteocalcin correlated with greater medial prefrontal cortex volume, which predicted one-year…In people with osteoporosis, higher undercarboxylated osteocalcin correlated with greater medial prefrontal cortex volume, which predicted one-year cognitive decline with an AUC of 0.84.
Across about 80,000 human participants, simple anatomical measurements performed as well as complex AI models while improved pretraining boosted…Across about 80,000 human participants, simple anatomical measurements performed as well as complex AI models while improved pretraining boosted biological age estimation. In a preprint analyzing structural brain MRI scans from approximately 80,000 human participants across 18 datasets…
Compact models trained on synthetic UK Biobank cohorts predicted patient survival with concordance indices reaching 0.91 without exposing real…Compact models trained on synthetic UK Biobank cohorts predicted patient survival with concordance indices reaching 0.91 without exposing real clinical data. In health records from the UK Biobank, researchers developed a privacy-preserving computational framework to evaluate multimorbidity…
Adding brain MRI to baseline traits predicted conversion from subjective decline to mild impairment with an AUC of 0.923 at five years.Adding brain MRI to baseline traits predicted conversion from subjective decline to mild impairment with an AUC of 0.923 at five years. In an observational study of 1,352 human participants with subjective cognitive decline, researchers evaluated machine learning models to forecast…
The model counted teeth with a mean absolute error of 1.96 and matched manual associations with cognitive scores in a 233-person cohort.The model counted teeth with a mean absolute error of 1.96 and matched manual associations with cognitive scores in a 233-person cohort. In human brain MRI scans from the OASIS-3 study, researchers developed an automated pipeline to extract dental information from routine neuroimaging.
Corticosterone treatment shifts DNA methylation profiles in young chicks toward those of older birds and reshapes longevity pathways.Corticosterone treatment shifts DNA methylation profiles in young chicks toward those of older birds and reshapes longevity pathways. In zebra finch chicks (Taeniopygia guttata), experimental early-life stress accelerates epigenetic aging, according to a new preprint.
Researchers identified a conserved neural progenitor pool that drives weekly brain regeneration in Botryllus schlosseri but collapses during aging.Researchers identified a conserved neural progenitor pool that drives weekly brain regeneration in Botryllus schlosseri but collapses during aging. In the colonial chordate Botryllus schlosseri, the entire central nervous system resorbs and regenerates de novo each week.
Researchers linked poor auditory nerve synchrony and myelin degradation to amplified communication difficulties in older humans, beyond standard…Researchers linked poor auditory nerve synchrony and myelin degradation to amplified communication difficulties in older humans, beyond standard hearing thresholds. In a human study detailed in a preprint, researchers investigated how auditory nerve changes affect hearing in older adults.
