An official website of the United States government
Substance use disorders affect 17.3% of Americans. Digital health solutions that use machine learning to detect substance use from wearable biosignal data can eventually pave the way for real-time digital interventions. However, difficulties in addressing severe between-subject data heterogeneity have hampered the adaptation of machine learning approaches for substance use detection, necessitating more robust technological solutions. We tested the utility of personalized machine learning using participant-specific convolutional neural networks (CNNs) enhanced with self-supervised learning (SSL) to detect drug use. In a pilot feasibility study, we collected data from 9 participants using Fitbit Charge 5 devices, supplemented by ecological momentary assessments to collect real-time labels of substance use. We implemented a baseline 1D-CNN model with traditional supervised learning and an experimental SSL-enhanced model to improve individualized feature extraction under limited label conditions. Results: Among the 9 participants, we achieved an average area under the receiver operating characteristic curve score across participants of 0.695 for the supervised CNNs and 0.729 for the SSL models. Strategic selection of an optimal threshold enabled us to optimize either sensitivity or specificity while maintaining reasonable performance for the other metric. Conclusion: These findings suggest that Fitbit data have the potential to enhance substance use monitoring systems. However, the small sample size in this study limits its generalizability to diverse populations, so we call for future research that explores SSL-powered personalization at a larger scale.
more »
« less
Personalization of AI Using Personal Foundation Models Can Lead to More Precise Digital Therapeutics
Digital health interventions often use machine learning (ML) models to make predictions of repeated adverse health events. For example, models may be used to analyze patient data to identify patterns that can anticipate the likelihood of disease exacerbations, enabling timely interventions and personalized treatment plans. However, many digital health applications require the prediction of highly heterogeneous and nuanced health events. The cross-subject variability of these events makes traditional ML approaches, where a single generalized model is trained to classify a particular condition, unlikely to generalize to patients outside of the training set. A natural solution is to train a separate model for each individual or subgroup, essentially overfitting the model to the unique characteristics of the individual without negatively overfitting in terms of the desired prediction task. Such an approach has traditionally required extensive data labels from each individual, a reality that has rendered personalized ML infeasible for precision health care. The recent popularization of self-supervised learning, however, provides a solution to this issue: by pretraining deep learning models on the vast array of unlabeled data streams arising from patient-generated health data, personalized models can be fine-tuned to predict the health outcome of interest with fewer labels than purely supervised approaches, making personalization of deep learning models much more achievable from a practical perspective. This perspective describes the current state-of-the-art in both self-supervised learning and ML personalization for health care as well as growing efforts to combine these two ideas by conducting self-supervised pretraining on an individual’s data. However, there are practical challenges that must be addressed in order to fully realize this potential, such as human-computer interaction innovations to ensure consistent labeling practices within a single participant.
more »
« less
- Award ID(s):
- 2516767
- PAR ID:
- 10687542
- Publisher / Repository:
- JMIR Publications
- Date Published:
- Journal Name:
- JMIR AI
- Volume:
- 4
- ISSN:
- 2817-1705
- Page Range / eLocation ID:
- e55530
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Electronic health records (EHRs) have been heavily used in modern healthcare systems for recording patients' admission information to health facilities. Many data-driven approaches employ temporal features in EHR for predicting specific diseases, readmission times, and diagnoses of patients. However, most existing predictive models cannot fully utilize EHR data, due to an inherent lack of labels in supervised training for some temporal events. Moreover, it is hard for the existing methods to simultaneously provide generic and personalized interpretability. To address these challenges, we propose Sherbet, a self-supervised graph learning framework with hyperbolic embeddings for temporal health event prediction. We first propose a hyperbolic embedding method with information flow to pretrain medical code representations in a hierarchical structure. We incorporate these pretrained representations into a graph neural network (GNN) to detect disease complications and design a multilevel attention method to compute the contributions of particular diseases and admissions, thus enhancing personalized interpretability. We present a new hierarchy-enhanced historical prediction proxy task in our self-supervised learning framework to fully utilize EHR data and exploit medical domain knowledge. We conduct a comprehensive set of experiments on widely used publicly available EHR datasets to verify the effectiveness of our model. Our results demonstrate the proposed model's strengths in both predictive tasks and interpretable abilities.more » « less
-
Diabetes management requires constant monitoring and individualized adjustments. This study proposes a novel approach that leverages digital twins and personal health knowledge graphs (PHKGs) to revolutionize diabetes care. Our key contribution lies in developing a real-time, patient-centric digital twin framework built on PHKGs. This framework integrates data from diverse sources, adhering to HL7 standards and enabling seamless information access and exchange while ensuring high levels of accuracy in data representation and health insights. PHKGs offer a flexible and efficient format that supports various applications. As new knowledge about the patient becomes available, the PHKG can be easily extended to incorporate it, enhancing the precision and accuracy of the care provided. This dynamic approach fosters continuous improvement and facilitates the development of new applications. As a proof of concept, we have demonstrated the versatility of our digital twins by applying it to different use cases in diabetes management. These include predicting glucose levels, optimizing insulin dosage, providing personalized lifestyle recommendations, and visualizing health data. By enabling real-time, patient-specific care, this research paves the way for more precise and personalized healthcare interventions, potentially improving long-term diabetes management outcomes.more » « less
-
A significant proportion of clinical physiologic monitoring alarms are false. This often leads to alarm fatigue in clinical personnel, inevitably compromising patient safety. To combat this issue, researchers have attempted to build Machine Learning (ML) models capable of accurately adjudicating Vital Sign (VS) alerts raised at the bedside of hemodynamically monitored patients as real or artifact. Previous studies have utilized supervised ML techniques that require substantial amounts of hand-labeled data. However, manually harvesting such data can be costly, time-consuming, and mundane, and is a key factor limiting the widespread adoption of ML in healthcare (HC). Instead, we explore the use of multiple, individually imperfect heuristics to automatically assign probabilistic labels to unlabeled training data using weak supervision. Our weakly supervised models perform competitively with traditional supervised techniques and require less involvement from domain experts, demonstrating their use as efficient and practical alternatives to supervised learning in HC applications of ML.more » « less
-
Human Activity Recognition (HAR) based on wearable inertial sensors plays a critical role in remote health monitoring. In patients with movement disorders, the ability to detect abnormal patient movements in their home environments can enable continuous optimization of treatments and help alert caretakers as needed. Machine learning approaches have been proposed for HAR tasks using Inertial Measurement Unit (IMU) data; however, most rely on application-specific labels and lack generalizability to data collected in different environments or populations. To address this limitation, we propose a new cross-modal self-supervised pretraining approach to learn representations from large-sale unlabeled IMU-video data and demonstrate improved generalizability in HAR tasks on out of distribution (OOD) IMU datasets, including a dataset collected from patients with Parkinson’s disease. Specifically, our results indicate that the proposed cross-modal pretraining approach outperforms the current state-of-the-art IMU-video pretraining approach and IMU-only pretraining under zero-shot and few-shot evaluations. Broadly, our study provides evidence that in highly dynamic data modalities, such as IMU signals, cross-modal pretraining may be a useful tool to learn generalizable data representations.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript
- Journal Article:
- https://doi.org/10.2196/55530
