Digital Phenotyping of Neuromuscular\u2013Cognitive Aging Using Portable Ultrasound and Multidomain

This dataset contains multidomain clinical and functional measurements collected from 40 community-dwelling older women to investigate neuromuscular\u2013cognitive aging phenotypes. The dataset includes portable ultrasound\u2013derived quadriceps muscle thickness, hand-grip strength, bioimpedance-based adjusted skeletal muscle index (ASMI), Montreal Cognitive Assessment (MoCA) scores, anthropometric variables (age, height, weight, BMI), and lower-extremity function indicators (gait speed, chair-stand time, and SPPB total score). All measurements were obtained using standardized clinical protocols performed by trained examiners.The dataset was originally developed for an explainable unsupervised machine-learning study aimed at identifying latent phenotypes representing distinct combinations of muscle morphology, strength, body composition, and cognitive performance. These data support research in digital phenotyping, geriatric assessment, sarcopenia classification, physical function modeling, and multimodal clustering. The dataset is suitable for PCA, clustering, feature importance analysis, predictive modeling, and validation of digital biomarker frameworks.All data are fully anonymized and contain no personally identifiable information. The study procedures were approved by an Institutional Review Board, and written informed consent was obtained from all participants. This dataset provides a valuable benchmark for researchers developing interpretable machine-learning models, digital health tools, or multimodal assessment systems for aging populations