arXiv subjects

Conor Heneghan

Publications and source records attributed to Conor Heneghan.

7 recordsLinked to original sources

arXiv2026

Towards a General Intelligence and Interface for Wearable Health Data

While ubiquitous wearable sensors capture a wealth of behavioral and physiological information, effectively transforming these signals into personalized health insights is challenging. Specifically, converting low-level sensor data into representations capable of characterizing higher-level states is difficult due to high phenotypic diversity and variation in individual baseline health, physiology, and lifestyle factors. Moreover, collecting wearable data paired with health outcome annotations is laborious and expensive, and retrospective annotation remains practically unfeasible, contributing to a scarcity of data with high-quality labels. To overcome these limitations, we propose a foundation model for wearable health that is pretrained on more than one trillion minutes of unlabeled sensor signals drawn from a large cohort of five million participants. We demonstrate that the joint scaling of model capacity and pretraining data volume leads to systematic improvements in performance, as evaluated on a diverse set of 35 health prediction tasks, spanning cardiovascular, metabolic, sleep, and mental health, as well as lifestyle choices and demographic factors. We find that this population scale representation unlocks label-efficient few-shot learning and generative capabilities for robust daily metric estimation. To further leverage this learned representation, we deploy a classroom of LLM agents to autonomously search the space of downstream predictive heads built on the model embeddings, showing broad performance improvements that increase with LLM model capacity. Finally, we show how integrating these downstream predictors into a Personal Health Agent can support model responses that are more relevant, contextually aware, and safe, and we validate this via 1,860 ratings from a cohort of clinicians.

cs.AI↗

arXiv2026

Quantifying Cardiovascular-Kidney-Metabolic(CKM) Health Age Using Wearables and lab Biomarkers

Cardiovascular-Kidney-Metabolic (CKM) syndrome represents a growing public health crisis, yet its subclinical heterogeneity remains underexplored. Early detection of physiological deviation is critical for preventing irreversible organ damage. Here, we characterize the prevalence and interplay of CKM impairment in a US cohort (N=819) by integrating continuous wearable data with clinical biomarkers. We assessed cardiovascular, kidney via clinical biomarkers, namely Chol/HDL, eGFR, as well as metabolic health risk through Homeostatic Model Assessment of Insulin Resistance (HOMA-IR). We show that while metabolic and cardiovascular disruptions are significantly associated (r=0.26, $p<0.001$), early-stage kidney impairment manifests independently. Utilizing a normalized deviance score, we identified significant health impairments in 29.0% of the cohort. Cardiovascular deviation was the most prevalent singular phenotype (13.3%), followed by metabolic (9.1%) and renal (6.25%) deviations, with dual metabolic-cardiovascular impairment occurring in only 2.2% of participants. These findings suggest that high system-specific deviance may serve as an indicator for accelerated physiological aging within the respective organ system. Furthermore, Step count, Active Zone Minutes, and resting heart rate were identified as the key wearable predictors of cardiovascular and metabolic decline. Leveraging these and CKM metrics, we developed a Health Age model that tracks physiological health, aligning with established clinical standards (Metabolic syndrome severity and modified AHA Life's Essential 8) and user-reported overall health. These findings underscore the necessity of a multi-system subtyping approach, demonstrating that wearable-derived phenotypes can facilitate the early, targeted interventions required to manage the complex landscape of CKM syndrome.

q-bio.QM↗

arXiv2025

Passive Measurement of Autonomic Arousal in Real-World Settings

The autonomic nervous system (ANS) is activated during stress, which can have negative effects on cardiovascular health, sleep, the immune system, and mental health. While there are ways to quantify ANS activity in laboratories, there is a paucity of methods that have been validated in real-world contexts. We present the Fitbit Body Response Algorithm, an approach to continuous remote measurement of ANS activation through widely available remote wrist-based sensors. The design was validated via two experiments, a Trier Social Stress Test (n = 45) and ecological momentary assessments (EMA) of perceived stress (n=87), providing both controlled and ecologically valid test data. Model performance predicting perceived stress when using all available sensor modalities was consistent with expectations (accuracy=0.85) and outperformed models with access to only a subset of the signals. We discuss and address challenges to sensing that arise in real world settings that do not present in conventional lab environments.

cs.HC↗

arXiv2025

Insulin Resistance Prediction From Wearables and Routine Blood Biomarkers

Insulin resistance, a precursor to type 2 diabetes, is characterized by impaired insulin action in tissues. Current methods for measuring insulin resistance, while effective, are expensive, inaccessible, not widely available and hinder opportunities for early intervention. In this study, we remotely recruited the largest dataset to date across the US to study insulin resistance (N=1,165 participants, with median BMI=28 kg/m2, age=45 years, HbA1c=5.4%), incorporating wearable device time series data and blood biomarkers, including the ground-truth measure of insulin resistance, homeostatic model assessment for insulin resistance (HOMA-IR). We developed deep neural network models to predict insulin resistance based on readily available digital and blood biomarkers. Our results show that our models can predict insulin resistance by combining both wearable data and readily available blood biomarkers better than either of the two data sources separately (R2=0.5, auROC=0.80, Sensitivity=76%, and specificity 84%). The model showed 93% sensitivity and 95% adjusted specificity in obese and sedentary participants, a subpopulation most vulnerable to developing type 2 diabetes and who could benefit most from early intervention. Rigorous evaluation of model performance, including interpretability, and robustness, facilitates generalizability across larger cohorts, which is demonstrated by reproducing the prediction performance on an independent validation cohort (N=72 participants). Additionally, we demonstrated how the predicted insulin resistance can be integrated into a large language model agent to help understand and contextualize HOMA-IR values, facilitating interpretation and safe personalized recommendations. This work offers the potential for early detection of people at risk of type 2 diabetes and thereby facilitate earlier implementation of preventative strategies.

cs.LG↗

arXiv2024

Towards a Personal Health Large Language Model

In health, most large language model (LLM) research has focused on clinical tasks. However, mobile and wearable devices, which are rarely integrated into such tasks, provide rich, longitudinal data for personal health monitoring. Here we present Personal Health Large Language Model (PH-LLM), fine-tuned from Gemini for understanding and reasoning over numerical time-series personal health data. We created and curated three datasets that test 1) production of personalized insights and recommendations from sleep patterns, physical activity, and physiological responses, 2) expert domain knowledge, and 3) prediction of self-reported sleep outcomes. For the first task we designed 857 case studies in collaboration with domain experts to assess real-world scenarios in sleep and fitness. Through comprehensive evaluation of domain-specific rubrics, we observed that Gemini Ultra 1.0 and PH-LLM are not statistically different from expert performance in fitness and, while experts remain superior for sleep, fine-tuning PH-LLM provided significant improvements in using relevant domain knowledge and personalizing information for sleep insights. We evaluated PH-LLM domain knowledge using multiple choice sleep medicine and fitness examinations. PH-LLM achieved 79% on sleep and 88% on fitness, exceeding average scores from a sample of human experts. Finally, we trained PH-LLM to predict self-reported sleep quality outcomes from textual and multimodal encoding representations of wearable data, and demonstrate that multimodal encoding is required to match performance of specialized discriminative models. Although further development and evaluation are necessary in the safety-critical personal health domain, these results demonstrate both the broad knowledge and capabilities of Gemini models and the benefit of contextualizing physiological data for personal health applications as done with PH-LLM.

cs.AI↗

arXiv2000

Heart Rate Variability: Measures and Models

We focus on various measures of the fluctuations of the sequence of intervals between beats of the human heart, and how such fluctuations can be used to assess the presence or likelihood of cardiovascular disease. We examine sixteen such measures and their suitability for correctly classifying heartbeat records of various lengths as normal or revealing the presence of cardiac dysfunction, particularly congestive heart failure. Using receiver-operating-characteristic analysis we demonstrate that scale-dependent measures prove substantially superior to scale-independent ones. The wavelet-transform standard deviation at a scale near 32 heartbeat intervals, and its spectral counterpart near 1/32 cycles/interval, turn out to provide reliable results using heartbeat records just minutes long. We further establish for all subjects that the human heartbeat has an underlying stochastic origin rather than arising from a chaotic attractor. Finally, we develop a mathematical point process that emulates the human heartbeat time series for both normal subjects and heart-failure patients.

physics.bio-ph↗

arXiv1997

Analysis, Synthesis, and Estimation of Fractal-Rate Stochastic Point Processes

Fractal and fractal-rate stochastic point processes (FSPPs and FRSPPs) provide useful models for describing a broad range of diverse phenomena, including electron transport in amorphous semiconductors, computer-network traffic, and sequences of neuronal action potentials. A particularly useful statistic of these processes is the fractal exponent $α$, which may be estimated for any FSPP or FRSPP by using a variety of statistical methods. Simulated FSPPs and FRSPPs consistently exhibit bias in this fractal exponent, however, rendering the study and analysis of these processes non-trivial. In this paper, we examine the synthesis and estimation of FRSPPs by carrying out a systematic series of simulations for several different types of FRSPP over a range of design values for $α$. The discrepancy between the desired and achieved values of $α$ is shown to arise from finite data size and from the character of the point-process generation mechanism. In the context of point-process simulation, reduction of this discrepancy requires generating data sets with either a large number of points, or with low jitter in the generation of the points. In the context of fractal data analysis, the results presented here suggest caution when interpreting fractal exponents estimated from experimental data sets.

adap-org↗