The Opportunities and Risks of Large Language Models in Mental Health.

Unlabelled: Global rates of mental health concerns are rising, and there is increasing realization that existing models of mental health care will not adequately expand to meet the demand. With the emergence of large language models (LLMs) has come great optimism regarding their promise to create novel, large-scale solutions to support mental health. Despite their nascence, LLMs have already been applied to mental health-related tasks. In this paper, we summarize the extant literature on efforts to use LLMs to provide mental health education, assessment, and intervention and highlight key opportunities for positive impact in each area. We then highlight risks associated with LLMs' application to mental health and encourage the adoption of strategies to mitigate these risks. The urgent need for mental health support must be balanced with responsible development, testing, and deployment of mental health LLMs. It is especially critical to ensure that mental health LLMs are fine-tuned for mental health, enhance mental health equity, and adhere to ethical standards and that people, including those with lived experience with mental health concerns, are involved in all stages from development through deployment. Prioritizing these efforts will minimize potential harms to mental health and maximize the likelihood that LLMs will positively impact mental health globally.

Examining the challenges of blood pressure estimation via photoplethysmogram.

The use of observed wearable sensor data (e.g., photoplethysmograms [PPG]) to infer health measures (e.g., glucose level or blood pressure) is a very active area of research. Such technology can have a significant impact on health screening, chronic disease management and remote monitoring. A common approach is to collect sensor data and corresponding labels from a clinical grade device (e.g., blood pressure cuff) and train deep learning models to map one to the other. Although well intentioned, this approach often ignores a principled analysis of whether the input sensor data have enough information to predict the desired metric. We analyze the task of predicting blood pressure from PPG pulse wave analysis. Our review of the prior work reveals that many papers fall prey to data leakage and unrealistic constraints on the task and preprocessing steps. We propose a set of tools to help determine if the input signal in question (e.g., PPG) is indeed a good predictor of the desired label (e.g., blood pressure). Using our proposed tools, we found that blood pressure prediction using PPG has a high multi-valued mapping factor of 33.2% and low mutual information of 9.8%. In comparison, heart rate prediction using PPG, a well-established task, has a very low multi-valued mapping factor of 0.75% and high mutual information of 87.7%. We argue that these results provide a more realistic representation of the current progress toward the goal of wearable blood pressure measurement via PPG pulse wave analysis. For code, see our project page: https://github.com/lirus7/PPG-BP-Analysis.

An overview of diagnostics and therapeutics using large language models.

There is an acute need for solutions to treat stress and trauma-related sequelae, and there are well-documented shortages of qualified human professionals. Artificial intelligence (AI) presents an opportunity to create advanced screening, diagnosis, and treatment solutions that relieve the burden on people and can provide just-in-time interventions. Large language models (LLMs), in particular, are promising given the role language plays in understanding and treating traumatic stress and other mental health conditions. In this article, we provide an overview of the state-of-the-art LLMs applications in diagnostic assessments, clinical note generation, and therapeutic support. We discuss the open research direction and challenges that need to be overcome to realize the full potential of deploying language models for use in clinical contexts. We highlight the need for increased representation in AI systems to ensure there are no disparities in access. Public datasets and models will help lead progress toward better models; however, privacy-preserving model training will be necessary for protecting patient data.

The Google Health Digital Well-Being Study: Protocol for a Digital Device Use and Well-Being Study.

BACKGROUND: The impact of digital device use on health and well-being is a pressing question. However, the scientific literature on this topic, to date, is marred by small and unrepresentative samples, poor measurement of core constructs, and a limited ability to address the psychological and behavioral mechanisms that may underlie the relationships between device use and well-being. Recent authoritative reviews have made urgent calls for future research projects to address these limitations. The critical role of research is to identify which patterns of use are associated with benefits versus risks and who is more vulnerable to harmful versus beneficial outcomes, so that we can pursue evidence-based product design, education, and regulation aimed at maximizing benefits and minimizing the risks of smartphones and other digital devices. OBJECTIVE: The objective of this study is to provide normative data on objective patterns of smartphone use. We aim to (1) identify how patterns of smartphone use impact well-being and identify groups of individuals who show similar patterns of covariation between smartphone use and well-being measures across time; (2) examine sociodemographic and personality or mental health predictors and which patterns of smartphone use and well-being are associated with pre-post changes in mental health and functioning; (3) discover which nondevice behavior patterns mediate the association between device use and well-being; (4) identify and explore recruitment strategies to increase and improve the representation of traditionally underrepresented populations; and (5) provide a real-world baseline of observed stress, mood, insomnia, physical activity, and sleep across a representative population. METHODS: This is a prospective, nonrandomized study to investigate the patterns and relationships among digital device use, sensor-based measures (including both behavioral and physiological signals), and self-reported measures of mental health and well-being. The study duration is 4 weeks per participant and includes passive sensing based on smartphone sensors, and optionally a wearable (Fitbit), for the complete study period. The smartphone device will provide activity, location, phone unlocks and app usage, and battery status information. RESULTS: At the time of submission, the study infrastructure and app have been designed and built, the institutional review board of the University of Oregon has approved the study protocol, and data collection is underway. Data from 4182 enrolled and consented participants have been collected as of March 27, 2023. We have made many efforts to sample a study population that matches the general population, and the demographic breakdown we have been able to achieve, to date, is not a perfect match. CONCLUSIONS: The impact of digital devices on mental health and well-being raises important questions. The Digital Well-Being Study is designed to help answer questions about the association between patterns of smartphone use and well-being. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/49189.

MMPD: Multi-Domain Mobile Video Physiology Dataset.

Remote photoplethysmography (rPPG) is an attractive method for noninvasive, convenient and concomitant measurement of physiological vital signals. Public benchmark datasets have served a valuable role in the development of this technology and improvements in accuracy over recent years. However, there remain gaps in the public datasets. First, despite the ubiquity of cameras on mobile devices, there are few datasets recorded specifically with mobile phone cameras. Second, most datasets are relatively small and therefore are limited in diversity, both in appearance (e.g., skin tone), behaviors (e.g., motion) and environment (e.g., lighting conditions). In an effort to help the field advance, we present the Multi-domain Mobile Video Physiology Dataset (MMPD), comprising 11 hours of recordings from mobile phones of 33 subjects. The dataset is designed to capture videos with greater representation across skin tone, body motion, and lighting conditions. MMPD is comprehensive with eight descriptive labels and can be used in conjunction with the rPPG-toolbox [1]. The reliability of the dataset is verified by mainstream unsupervised methods and neural methods. The GitHub repository of our dataset: https://github.com/THU-CS-PI/MMPD_rPPG_dataset.

Video-based sympathetic arousal assessment via peripheral blood flow estimation.

Electrodermal activity (EDA) is considered a standard marker of sympathetic activity. However, traditional EDA measurement requires electrodes in steady contact with the skin. Can sympathetic arousal be measured using only an optical sensor, such as an RGB camera? This paper presents a novel approach to infer sympathetic arousal by measuring the peripheral blood flow on the face or hand optically. We contribute a self-recorded dataset of 21 participants, comprising synchronized videos of participants' faces and palms and gold-standard EDA and photoplethysmography (PPG) signals. Our results show that we can measure peripheral sympathetic responses that closely correlate with the ground truth EDA. We obtain median correlations of 0.57 to 0.63 between our inferred signals and the ground truth EDA using only videos of the participants' palms or foreheads or PPG signals from the foreheads or fingers. We also show that sympathetic arousal is best inferred from the forehead, finger, or palm.

Objective Communication Patterns Associated With Team Member Effectiveness in Real-World Virtual Teams.

OBJECTIVE: We explore the relationships between objective communication patterns displayed during virtual team meetings and established, qualitative measures of team member effectiveness. BACKGROUND: A key component of teamwork is communication. Automated measures of objective communication patterns are becoming more feasible and offer the ability to measure and monitor communication in a scalable, consistent and continuous manner. However, their validity in reflecting meaningful measures of teamwork processes are not well established, especially in real-world settings. METHOD: We studied real-world virtual student teams working on semester-long projects. We captured virtual team meetings using the Zoom video conferencing platform throughout the semester and periodic surveys comprising peer ratings of team member effectiveness. Leveraging audio transcripts, we examined relationships between objective measures of speaking time, silence gap duration and vocal turn-taking and peer ratings of team member effectiveness. RESULTS: Speaking time, speaking turn count, degree centrality and (marginally) speaking turn duration, but not silence gap duration, were positively related to individual-level team member effectiveness. Time in dyadic interactions and interaction count, but not interaction length, were positively related to dyad-level team member effectiveness. CONCLUSION: Our study highlights the relevance of objective measures of speaking time and vocal turn-taking to team member effectiveness in virtual project-based teams, supporting the validity of these objective measures and their use in future research. APPLICATION: Our approach offers a scalable, easy-to-use method for measuring communication patterns and team member effectiveness in virtual teams and opens the opportunity to study these patterns in a more continuous and dynamic manner.

Using High-Fidelity Avatars to Advance Camera-Based Cardiac Pulse Measurement.

Non-contact physiological measurement has the potential to provide low-cost, non-invasive health monitoring. However, machine vision approaches are often limited by the availability and diversity of annotated video datasets resulting in poor generalization to complex real-life conditions. To address these challenges, this work proposes the use of synthetic avatars that display facial blood flow changes and allow for systematic generation of samples under a wide variety of conditions. Our results show that training on both simulated and real video data can lead to performance gains under challenging conditions. We show strong performance on three large benchmark datasets and improved robustness to skin type and motion. These results highlight the promise of synthetic data for training camera-based pulse measurement; however, further research and validation is needed to establish whether synthetic data alone could be sufficient for training models.

Synthetic Data for Multi-Parameter Camera-Based Physiological Sensing.

Synthetic data is a powerful tool in training data hungry deep learning algorithms. However, to date, camera-based physiological sensing has not taken full advantage of these techniques. In this work, we leverage a high-fidelity synthetics pipeline for generating videos of faces with faithful blood flow and breathing patterns. We present systematic experiments showing how physiologically-grounded synthetic data can be used in training camera-based multi-parameter cardiopulmonary sensing. We provide empirical evidence that heart and breathing rate measurement accuracy increases with the number of synthetic avatars in the training set. Furthermore, training with avatars with darker skin types leads to better overall performance than training with avatars with lighter skin types. Finally, we discuss the opportunities that synthetics present in the domain of camera-based physiological sensing and limitations that need to be overcome.

Systematic analysis of video-based pulse measurement from compressed videos.

Camera-based physiological measurement enables vital signs to be captured unobtrusively without contact with the body. Remote, or imaging, photoplethysmography involves recovering peripheral blood flow from subtle variations in video pixel intensities. While the pulse signal might be easy to obtain from high quality uncompressed videos, the signal-to-noise ratio drops dramatically with video bitrate. Uncompressed videos incur large file storage and data transfer costs, making analysis, manipulation and sharing challenging. To help address these challenges, we use compression specific supervised models to mitigate the effect of temporal video compression on heart rate estimates. We perform a systematic evaluation of the performance of state-of-the-art algorithms across different levels, and formats, of compression. We demonstrate that networks trained on compressed videos consistently outperform other benchmark methods, both on stationary videos and videos with significant rigid head motions. By training on videos with the same, or higher compression factor than test videos, we achieve improvements in signal-to-noise ratio (SNR) of up to 3 dB and mean absolute error (MAE) of up to 6 beats per minute (BPM).

Non-contact imaging of peripheral hemodynamics during cognitive and psychological stressors.

Peripheral hemodynamics, measured via the blood volume pulse and vasomotion, provide a valuable way of monitoring physiological state. Camera imaging-based systems can be used to measure these peripheral signals without contact with the body, at distances of multiple meters. While researchers have paid attention to non-contact imaging photoplethysmography, the study of peripheral hemodynamics and the effect of autonomic nervous system activity on these signals has received less attention. Using a method, based on a tissue-like model of the skin, we extract melanin [Formula: see text] and hemoglobin [Formula: see text] concentrations from videos of the hand and face and show that significant decreases in peripheral pulse signal power (by 36% ± 29%) and vasomotion signal power (by 50% ± 26%) occur during periods of cognitive and psychological stress. Via three experiments we show that similar results are achieved across different stimuli and regions of skin (face and hand). While changes in peripheral pulse and vasomotion power were significant the changes in pulse rate variability were less consistent across subjects and tasks.

Simple and affordable imaging of multiple physiological parameters with RGB camera-based diffuse reflectance spectroscopy.

We propose a simple and affordable imaging technique to evaluate transcutaneously multiple physiological parameters by using a digital red-green-blue camera. In this method, the RGB-values were converted into tristimulus values in the CIE (Commission Internationale de l'Eclairage) XYZ color space, which is compatible with the common color spaces. Monte Carlo simulation for light transport in biological tissue was then performed to specify the relationship among the XYZ-values and the concentrations of oxygenated hemoglobin, deoxygenated hemoglobin, bilirubin, and melanin. The concentration of total hemoglobin and tissue oxygen saturation were also calculated from the estimated concentrations of oxygenated and deoxygenated hemoglobin. In vivo experiments with bile duct ligation in rats demonstrated that the estimated bilirubin concentration increased after ligation of the bile duct and reached around 22 mg/dl at 116 h after the onset of ligation, which corresponds to the ground truth value of bilirubin measured by a commercially available transcutaneous bilirubinometer. Experiments with rats while varying the fraction of inspired oxygen demonstrated that oxygenated hemoglobin and deoxygenated hemoglobin decreased and increased, respectively, as the fraction of inspired oxygen decreased. Consequently, tissue oxygen saturation dramatically decreased. We further extended the method to a non-contact imaging photo-plethysmograph and estimation of the percutaneous oxygen saturation. An empirical formula to estimate percutaneous oxygen saturation was derived from the pulse wave amplitudes of oxygenated and deoxygenated hemoglobin. The estimated percutaneous oxygen saturation dropped remarkably when a faction of inspired oxygen was below 19%, indicating the onset of hypoxemia due to hypoxia, whereas the tissue oxygen saturation decreased gradually according to the reduction of the faction of inspired oxygen. The results in this study indicate the potential of this method for imaging of multiple physiological parameters in skin tissue and evaluating an optical biomedical imaging technique that enables cost-effective, easy-to-use, portable, remotely administered, and/or point-of-care solutions.

iPhys: An Open Non-Contact Imaging-Based Physiological Measurement Toolbox.

Imaging-based, non-contact measurement of physiology (including imaging photoplethysmography and imaging ballistocardiography) is a growing field of research. There are several strengths of imaging methods that make them attractive. They remove the need for uncomfortable contact sensors and can enable spatial and concomitant measurement from a single sensor. Furthermore, cameras are ubiquitous and often low-cost solutions for sensing. Open source toolboxes help accelerate the progress of research by providing a means to compare new approaches against standard implementations of the state-of-the-art. We present an open source imaging-based physiological measurement toolbox with implementations of many of the most frequently employed computational methods. We hope that this toolbox will contribute to the advancement of noncontact physiological sensing methods. Code: https://github.com/danmcduff/iphys-toolbox.

Using Non-Contact Imaging Photoplethysmography to Recover Diurnal Patterns in Heart Rate.

Daily patterns in cardiovascular signals can reveal important information about physiological processes, health and well-being. Traditionally, contact sensors have been used to collect longitudinal data of this kind. However, recent advances in non-contact imaging techniques have led to algorithms that can be used to measure vital signs unobtrusively. Imaging methods are highly scalable due to the availability of webcams and computing devices making them attractive for longitudinal, in-situ measurement. Using a software tool we captured over 1,000 hours of non-contact heart rate measurements, via imaging photoplethysmography. Using these data we were able to recover diurnal patterns in heart rate during the working day. Non-contact sensing techniques hold much promise but also raise ethical issues that need to be addressed seriously within the biomedical engineering community.

Wearable Motion-Based Heart Rate at Rest: A Workplace Evaluation.

This paper studies the feasibility of using low-cost motion sensors to provide opportunistic heart rate assessments from ballistocardiographic signals during restful periods of daily life. Three wearable devices were used to capture peripheral motions at specific body locations (head, wrist, and trouser pocket) of 15 participants during five regular workdays each. Three methods were implemented to extract heart rate from motion data and their performance was compared to those obtained with an FDA-cleared device. With a total of 1358 h of naturalistic sensor data, our results show that providing accurate heart rate estimations from peripheral motion signals is possible during relatively "still" moments. In our real-life workplace study, the head-mounted device yielded the most frequent assessments (22.98% of the time under 5 beats per minute of error) followed by the smartphone in the pocket (5.02%) and the wrist-worn device (3.48%). Most importantly, accurate assessments were automatically detected by using a custom threshold based on the device jerk. Due to the pervasiveness and low cost of wearable motion sensors, this paper demonstrates the feasibility of providing opportunistic large-scale low-cost samples of resting heart rate.

Spectral Estimation Methods for Evaluating iPPG Pulse Rate Variability.

Non-contact measurement of physiological parameters, like pulse rate variability (PRV), has numerous applications in medicine and affective computing. PRV is an informative measure of autonomic nervous system activity. Spectral estimation from unevenly sampled, non-stationary data is integral to pulse rate variability frequency-domain analysis. We present the first comparison of results of PRV computation using the Lomb-Scargle method and Bayesian Spectral Estimation. The Lomb-Scargle method performs well, even in the presence of missing beats. However, the Bayesian Spectral Estimation method has advantages when tracking changes in amplitude and frequency. We illustrate these characteristics with results from synthetic data and real non-contact imaging photoplethysmography measurements.

InPhysible: Camouflage Against Video-Based Physiological Measurement.

Imaging photoplethysmography (iPPG) is a powerful set of methods for measuring physiological signals from video. Recent advances have shown that a low-cost webcam can be used to measure heart rate, blood flow, respiration, blood oxygen levels and stress. While these methods have many beneficial applications, the unobtrusive and ubiquitous nature of the sensors risk exposing people to unwanted measurement. We present InPhysible the first camouflage system against video- based physiological measurement. The infra-red system can be embedded into any pair of glasses, or other headwear, and disrupts the measurement of the iPPG signal while being imperceptible by the human eye. Our system is flexible and can simulate realistic pulse signals to hinder heart rate measurement. In this paper we present the design of our prototype and a user study validating its efficacy. Finally, we discuss the limitations and implications for data privacy and security.

Fusing Partial Camera Signals for Noncontact Pulse Rate Variability Measurement.

Remote camera-based measurement of physiology has great potential for healthcare and affective computing. Recent advances in computer vision and signal processing have enabled photoplethysmography (PPG) measurement using commercially available cameras. However, there remain challenges in recovering accurate noncontact PPG measurements in the presence of rigid head motion. When a subject is moving, their face may be turned away from one camera, be obscured by an object, or move out of the frame resulting in missing observations. As the calculation of pulse rate variability (PRV) requires analysis over a time window of several minutes, the effect of missing observations on such features is deleterious. We present an approach for fusing partial color-channel signals from an array of cameras that enable physiology measurements to be made from moving subjects, even if they leave the frame of one or more cameras, which would not otherwise be possible with only a single camera. We systematically test our method on subjects ( N=25) using a set of six, 5-min tasks (each repeated twice) involving different levels of head motion. This results in validation across 25 h of measurement. We evaluate pulse rate and PRV parameter estimation including statistical, geometric, and frequency-based measures. The median absolute error in pulse rate measurements was 0.57 beats-per-minute (BPM). In all but two tasks with the greatest motion, the median error was within 0.4 BPM of that from a contact PPG device. PRV estimates were significantly improved using our proposed approach compared to an alternative not designed to handle missing values and multiple camera signals; the error was reduced by over 50%. Without our proposed method, errors in pulse rate would be very high, and estimation of PRV parameters would not be feasible due to significant data loss.

A large-scale analysis of sex differences in facial expressions.

There exists a stereotype that women are more expressive than men; however, research has almost exclusively focused on a single facial behavior, smiling. A large-scale study examines whether women are consistently more expressive than men or whether the effects are dependent on the emotion expressed. Studies of gender differences in expressivity have been somewhat restricted to data collected in lab settings or which required labor-intensive manual coding. In the present study, we analyze gender differences in facial behaviors as over 2,000 viewers watch a set of video advertisements in their home environments. The facial responses were recorded using participants' own webcams. Using a new automated facial coding technology we coded facial activity. We find that women are not universally more expressive across all facial actions. Nor are they more expressive in all positive valence actions and less expressive in all negative valence actions. It appears that generally women express actions more frequently than men, and in particular express more positive valence actions. However, expressiveness is not greater in women for all negative valence actions and is dependent on the discrete emotional state.

A survey of remote optical photoplethysmographic imaging methods.

In recent years researchers have presented a number of new methods for recovering physiological parameters using just low-cost digital cameras and image processing. The ubiquity of digital cameras presents the possibility for many new, low-cost applications of vital sign monitoring. In this paper we present a review of the work on remote photoplethysmographic (PPG) imaging using digital cameras. This review specifically focuses on the state-of-the-art in PPG imaging where: 1) measures beyond pulse rate are evaluated, 2) non-ideal conditions (e.g., the presence of motion artifacts) are explored, and 3) use cases in relevant environments are demonstrated. We discuss gaps within the literature and future challenges for the research community. To aid in the continuing advancement of PPG imaging research, we are making available a website with the references collected for this review as well as information on available code and datasets of interest. It is our hope that this website will become a valuable resource for the PPG imaging community. The site can be found at: http://web.mit.edu/~djmcduff/www/ remote-physiology.html.