The human blood system is maintained through the differentiation and massive amplification of a limited number of long-lived haematopoietic stem cells (HSCs)1. Perturbations to this process underlie diverse diseases, but the clonal contributions to human haematopoiesis and how this changes with age remain incompletely understood. Although recent insights have emerged from barcoding studies in model systems2–5, simultaneous detection of cell states and phylogenies from natural barcodes in humans remains challenging. Here we introduce an improved, single-cell lineage-tracing system based on deep detection of naturally occurring mitochondrial DNA mutations with simultaneous readout of transcriptional states and chromatin accessibility. We use this system to define the clonal architecture of HSCs and map the physiological state and output of clones. We uncover functional heterogeneity in HSC clones, which is stable over months and manifests as both differences in total HSC output and biases towards the production of different mature cell types. We also find that the diversity of HSC clones decreases markedly with age, leading to an oligoclonal structure with multiple distinct clonal expansions. Our study thus provides a clonally resolved and cell-state-aware atlas of human haematopoiesis at single-cell resolution, showing an unappreciated functional diversity of human HSC clones and, more broadly, paving the way for refined studies of clonal dynamics across a range of tissues in human health and disease.
The United States Department of Health and Human Services (HHS) pledged $90 million to help reduce health disparities with data-driven solutions. The funds are being distributed to 1400 community health centers, serving over 30 million Americans. Given these developments, our piece examines the reasons behind the delayed adoption of big data for healthcare equity, recent efforts embracing big data tools, and methods to maximize potential without overburdening physicians. We additionally propose a public database for anonymized patient data, introducing diverse metrics and equitable data collection strategies, providing valuable insights for policymakers and health systems to better serve communities.
Artificial intelligence technologies are rapidly advancing and significantly impacting healthcare, particularly in critical care environments where rapid, precise decision-making is crucial. They promise reductions in clinical errors, enhanced diagnostic accuracy, optimized treatment plans, and better resource allocation. Artificial intelligence applications are widespread across medical fields, with numerous artificial intelligence/machine learning-enabled medical devices approved by […] ABSTRACT Artificial intelligence technologies are rapidly advancing and significantly impacting healthcare, particularly in critical care environments where rapid, precise decision-making is crucial. They promise reductions in clinical errors, enhanced diagnostic accuracy, optimized treatment plans, and better resource allocation. Artificial intelligence applications are widespread across medical fields, with numerous artificial intelligence/machine learning-enabled medical devices approved by […]
Early and accurate prediction of Alzheimer’s disease (AD) from accessible clinical data remains a significant challenge in healthcare. This study proposes ALTARN, a tabular attention residual neural network architecture for robust classification of AD through heterogeneous patient data from a publicly available dataset of 2149 subjects, with medical, demographic, and lifestyle variables organized as structured tabular data. With sigmoid attention mechanisms for the dynamic reweighing of input variables for each patient, deep residual connections capturing complex, and non-linear relationships in tabular features, we propose ALTARN-an early Alzheimer prediction tool. ALTARN achieved an average cross-validated training accuracy of roughly 92.73%, alongside robust validation metrics, including an average accuracy of 85.06%, average precision of 80.73%, mean recall of 76.18%, and a mean validation F1-score of 78.32% when evaluated through five-fold stratified validation. When tested against other approaches, ALTARN meets and also exceeds the performance of supervised deep learning models for non-imaging AD classification. This further illustrates that deep neural network based approaches with tabular attention offer a promising direction for interpretable diagnosis of AD via non-imaging medical data.