Necessary for seizure forecasting outcome metrics: seizure frequency and benchmark model.

Work is ongoing to advance seizure forecasting, but the performance metrics used to evaluate model effectiveness can sometimes lead to misleading outcomes. For example, some metrics improve when tested on patients with a particular range of seizure frequencies (SF). This study illustrates the connection between SF and metrics. Additionally, we compared benchmarks for testing performance: a moving average (MA) or the commonly used permutation benchmark. Three data sets were used for the evaluations: (1) Self-reported seizure diaries of 3,994 Seizure Tracker patients; (2) Automatically detected (and sometimes manually reported or edited) generalized tonic-clonic seizures from 2,350 Empatica Embrace 2 and Mate App seizure diary users, and (3) Simulated datasets with varying SFs. Metrics of calibration and discrimination were computed for each dataset, comparing MA and permutation performance across SF values. Most metrics were found to depend on SF. The MA model outperformed or matched the permutation model in all cases. The findings highlight SF's role in seizure forecasting accuracy and the MA model's suitability as a benchmark. This underscores the need for considering patient SF in forecasting studies and suggests the MA model may provide a better standard for evaluating future seizure forecasting models.

Brief Report: Quantifying Speech Production Coordination from Non- and Minimally-Speaking Individuals.

PURPOSE: Non-verbal utterances are an important tool of communication for individuals who are non- or minimally-speaking. While these utterances are typically understood by caregivers, they can be challenging to interpret by their larger community. To date, there has been little work done to detect and characterize the vocalizations produced by non- or minimally-speaking individuals. This paper aims to characterize five categories of utterances across a set of 7 non- or minimally-speaking individuals. METHODS: The characterization is accomplished using a correlation structure methodology, acting as a proxy measurement for motor coordination, to localize similarities and differences to specific speech production systems. RESULTS: We specifically find that frustrated and dysregulated utterances show similar correlation structure outputs, especially when compared to self-talk, request, and delighted utterances. We additionally witness higher complexity of coordination between articulatory and respiratory subsystems and lower complexity of coordination between laryngeal and respiratory subsystems in frustration and dysregulation as compared to self-talk, request, and delight. Finally, we observe lower complexity of coordination across all three speech subsystems in the request utterances as compared to self-talk and delight. CONCLUSION: The insights from this work aid in understanding of the modifications made by non- or minimally-speaking individuals to accomplish specific goals in non-verbal communication.

Deep learning-aided decision support for diagnosis of skin disease across skin tones.

Although advances in deep learning systems for image-based medical diagnosis demonstrate their potential to augment clinical decision-making, the effectiveness of physician-machine partnerships remains an open question, in part because physicians and algorithms are both susceptible to systematic errors, especially for diagnosis of underrepresented populations. Here we present results from a large-scale digital experiment involving board-certified dermatologists (n = 389) and primary-care physicians (n = 459) from 39 countries to evaluate the accuracy of diagnoses submitted by physicians in a store-and-forward teledermatology simulation. In this experiment, physicians were presented with 364 images spanning 46 skin diseases and asked to submit up to four differential diagnoses. Specialists and generalists achieved diagnostic accuracies of 38% and 19%, respectively, but both specialists and generalists were four percentage points less accurate for the diagnosis of images of dark skin as compared to light skin. Fair deep learning system decision support improved the diagnostic accuracy of both specialists and generalists by more than 33%, but exacerbated the gap in the diagnostic accuracy of generalists across skin tones. These results demonstrate that well-designed physician-machine partnerships can enhance the diagnostic accuracy of physicians, illustrating that success in improving overall diagnostic accuracy does not necessarily address bias.

Testing a Behavioral Activation Gaming App for Depression During Pregnancy: Multimethod Pilot Study.

BACKGROUND: Depression during pregnancy is increasingly recognized as a worldwide public health problem. If untreated, there can be detrimental outcomes for the mother and child. Anxiety is also often comorbid with depression. Although effective treatments exist, most women do not receive treatment. Technology is a mechanism to increase access to and engagement in mental health services. OBJECTIVE: The Guardians is a mobile app, grounded in behavioral activation principles, which seeks to leverage mobile game mechanics and in-game rewards to encourage user engagement. This study seeks to assess app satisfaction and engagement and to explore changes in clinical symptoms of depression and anxiety among a sample of pregnant women with elevated depressive symptoms. METHODS: This multimethod pilot test consisted of a single-arm, proof-of-concept trial to examine the feasibility and acceptability of The Guardians among a pregnant sample with depression (N=18). Participation included two web-based study visits: (1) a baseline assessment to collect demographic and obstetric information and to assess clinical symptoms and (2) an exit interview to administer follow-up measures and explore user experience. Participants completed biweekly questionnaires (ie, Patient Health Questionnaire-9 and Generalized Anxiety Disorder-7) during the trial to assess depression and anxiety symptom severity. App satisfaction was measured using 2 self-report scales (ie, Mobile Application Rating Scale and Player Experience of Needs Satisfaction scale). Engagement with The Guardians was captured using game interaction metric data. We used backward-eliminated mixed effects longitudinal models to examine the effects of app engagement and satisfaction and length of time in the study on symptoms of depression and anxiety. Content analysis was conducted on qualitative data from exit interviews. RESULTS: The 15-day and 30-day overall app retention rates were 26.6% and 15.1%, respectively. Mixed effects models found significant negative main effects of week in study (ß=-.35; t61=-3.05; P=.003), number of activities completed (ß=-.12; t61=-2.05; P=.04), days played (ß=-.12; t58=-2.9; P=.005), and satisfaction, according to the Mobile Application Rating Scale (ß=-3.05; t45=-2.19; P=.03) on depressive symptoms. We have reported about similar analyses for anxiety. There is preliminary evidence suggesting harder activities are associated with greater mood improvement than easier activities. Qualitative content analysis resulted in feedback falling under the following themes: activities, app design, engagement, fit of the app with lifestyle, perceived impact of the app on mood, and suggestions for app modifications. CONCLUSIONS: Preliminary results from this multimethod study of The Guardians indicate feasibility and acceptability among pregnant women with depression. Retention and engagement levels were more than double those of previous public mental health apps, and use of the app was associated with significant decrease in depressive symptom scores over the 10-week trial. The Guardians shows promise as an effective and scalable digital intervention to support women experiencing depression.

Prospective clinical validation of the Empatica EmbracePlus wristband as a reflective pulse oximeter.

Introduction: Respiratory diseases such as chronic obstructive pulmonary disease, obstructive sleep apnea syndrome, and COVID-19 may cause a decrease in arterial oxygen saturation (SaO2). The continuous monitoring of oxygen levels may be beneficial for the early detection of hypoxemia and timely intervention. Wearable non-invasive pulse oximetry devices measuring peripheral oxygen saturation (SpO2) have been garnering increasing popularity. However, there is still a strong need for extended and robust clinical validation of such devices, especially to address topical concerns about disparities in performances across racial groups. This prospective clinical validation aimed to assess the accuracy of the reflective pulse oximeter function of the EmbracePlus wristband during a controlled hypoxia study in accordance with the ISO 80601-2-61:2017 standard and the Food & Drug Administration (FDA) guidance. Methods: Healthy adult participants were recruited in a controlled desaturation protocol to reproduce mild, moderate, and severe hypoxic conditions with SaO2 ranging from 100% to 70% (ClinicalTrials.gov registration #NCT04964609). The SpO2 level was estimated with an EmbracePlus device placed on the participant's wrist and the reference SaO2 was obtained from blood samples analyzed with a multiwavelength co-oximeter. Results: The controlled hypoxia study yielded 373 conclusive measurements on 15 subjects, including 30% of participants with dark skin pigmentation (V-VI on the Fitzpatrick scale). The accuracy root mean square (Arms) error was found to be 2.4%, within the 3.5% limit recommended by the FDA. A strong positive correlation between the wristband SpO2 and the reference SaO2 was observed (r = 0.96, P < 0.001), and a good concordance was found with Bland-Altman analysis (bias, 0.05%; standard deviation, 1.66; lower limit, -4.7%; and upper limit, 4.8%). Moreover, acceptable accuracy was observed when stratifying data points by skin pigmentation (Arms 2.2% in Fitzpatrick V-VI, 2.5% in Fitzpatrick I-IV), and sex (Arms 1.9% in females, and 2.9% in males). Discussion: This study demonstrates that the EmbracePlus wristband could be used to assess SpO2 with clinically acceptable accuracy under no-motion and high perfusion conditions for individuals of different ethnicities across the claimed range. This study paves the way for further accuracy evaluations on unhealthy subjects and during prolonged use in ambulatory settings.

ReCANVo: A database of real-world communicative and affective nonverbal vocalizations.

Nonverbal vocalizations, such as sighs, grunts, and yells, are informative expressions within typical verbal speech. Likewise, individuals who produce 0-10 spoken words or word approximations ("minimally speaking" individuals) convey rich affective and communicative information through nonverbal vocalizations even without verbal speech. Yet, despite their rich content, little to no data exists on the vocal expressions of this population. Here, we present ReCANVo: Real-World Communicative and Affective Nonverbal Vocalizations - a novel dataset of non-speech vocalizations labeled by function from minimally speaking individuals. The ReCANVo database contains over 7000 vocalizations spanning communicative and affective functions from eight minimally speaking individuals, along with communication profiles for each participant. Vocalizations were recorded in real-world settings and labeled in real-time by a close family member who knew the communicator well and had access to contextual information while labeling. ReCANVo is a novel database of nonverbal vocalizations from minimally speaking individuals, the largest available dataset of nonverbal vocalizations, and one of the only affective speech datasets collected amidst daily life across contexts.

Interventions to Modify Psychological Well-Being: Progress, Promises, and an Agenda for Future Research.

Psychological well-being, characterized by feelings, cognitions, and strategies that are associated with positive functioning (including hedonic and eudaimonic well-being), has been linked with better physical health and greater longevity. Importantly, psychological well-being can be strengthened with interventions, providing a strategy for improving population health. But are the effects of well-being interventions meaningful, durable, and scalable enough to improve health at a population-level? To assess this possibility, a cross-disciplinary group of scholars convened to review current knowledge and develop a research agenda. Here we summarize and build on the key insights from this convening, which were: (1) existing interventions should continue to be adapted to achieve a large-enough effect to result in downstream improvements in psychological functioning and health, (2) research should determine the durability of interventions needed to drive population-level and lasting changes, (3) a shift from individual-level care and treatment to a public-health model of population-level prevention is needed and will require new infrastructure that can deliver interventions at scale, (4) interventions should be accessible and effective in racially, ethnically, and geographically diverse samples. A discussion examining the key future research questions follows.

Rising early warning signals in affect associated with future changes in depression: a dynamical systems approach.

BACKGROUND: Predicting future states of psychopathology such as depressive episodes has been a hallmark initiative in mental health research. Dynamical systems theory has proposed that rises in certain 'early warning signals' (EWSs) in time-series data (e.g. auto-correlation, temporal variance, network connectivity) may precede impending changes in disorder severity. The current study investigates whether rises in these EWSs over time are associated with future changes in disorder severity among a group of patients with major depressive disorder (MDD). METHODS: Thirty-one patients with MDD completed the study, which consisted of daily smartphone-delivered surveys over 8 weeks. Daily positive and negative affect were collected for the time-series analyses. A rolling window approach was used to determine whether rises in auto-correlation of total affect, temporal standard deviation of total affect, and overall network connectivity in individual affect items were predictive of increases in depression symptoms. RESULTS: Results suggested that rises in auto-correlation were significantly associated with worsening in depression symptoms (r = 0.41, p = 0.02). Results indicated that neither rises in temporal standard deviation (r = -0.23, p = 0.23) nor in network connectivity (r = -0.12, p = 0.59) were associated with changes in depression symptoms. CONCLUSIONS: This study more rigorously examines whether rises in EWSs were associated with future depression symptoms in a larger group of patients with MDD. Results indicated that rises in auto-correlation were the only EWS that was associated with worsening future changes in depression.

Real-time digital monitoring of a suicide attempt by a hospital patient.

Suicide is among the most devastating problems facing clinicians, who currently have limited tools to predict and prevent suicidal behavior. Here we report on real-time, continuous smartphone and sensor data collected before, during, and after a suicide attempt made by a patient during a psychiatric inpatient hospitalization. We observed elevated and persistent sympathetic nervous system arousal and suicidal thinking leading up to the suicide attempt. This case provides the highest resolution data to date on the psychological, psychophysiological, and behavioral markers of imminent suicidal behavior and highlights new directions for prediction and prevention efforts.

A Robotic Companion for Psychological Well-being: A Long-term Investigation of Companionship and Therapeutic Alliance.

Social support plays a crucial role in managing and enhancing one's mental health and well-being. In order to explore the role of a robot's companion-like behavior on its therapeutic interventions, we conducted an eight-week-long deployment study with seventy participants to compare the impact of (1) a control robot with only assistant-like skills, (2) a coach-like robot with additional instructive positive psychology interventions, and (3) a companion-like robot that delivered the same interventions in a peer-like and supportive manner. The companion-like robot was shown to be the most effective in building a positive therapeutic alliance with people, enhancing participants' well-being and readiness for change. Our work offers valuable insights into how companion AI agents could further enhance the efficacy of the mental health interventions by strengthening their therapeutic alliance with people for long-term mental health support.

Deploying a robotic positive psychology coach to improve college students' psychological well-being.

Despite the increase in awareness and support for mental health, college students' mental health is reported to decline every year in many countries. Several interactive technologies for mental health have been proposed and are aiming to make therapeutic service more accessible, but most of them only provide one-way passive contents for their users, such as psycho-education, health monitoring, and clinical assessment. We present a robotic coach that not only delivers interactive positive psychology interventions but also provides other useful skills to build rapport with college students. Results from our on-campus housing deployment feasibility study showed that the robotic intervention showed significant association with increases in students' psychological well-being, mood, and motivation to change. We further found that students' personality traits were associated with the intervention outcomes as well as their working alliance with the robot and their satisfaction with the interventions. Also, students' working alliance with the robot was shown to be associated with their pre-to-post change in motivation for better well-being. Analyses on students' behavioral cues showed that several verbal and nonverbal behaviors were associated with the change in self-reported intervention outcomes. The qualitative analyses on the post-study interview suggest that the robotic coach's companionship made a positive impression on students, but also revealed areas for improvement in the design of the robotic coach. Results from our feasibility study give insight into how learning users' traits and recognizing behavioral cues can help an AI agent provide personalized intervention experiences for better mental health outcomes.

Building a culture of responsible neurotech: Neuroethics as socio-technical challenges.

Scientists around the globe are joining the race to achieve engineering feats to read, write, modulate, and interface with the human brain in a broadening continuum of invasive to non-invasive ways. The expansive implications of neurotechnology for our conception of health, mind, decision-making, and behavior has raised social and ethical considerations that are inextricable from neurotechnological progress. We propose "socio-technical" challenges as a framing to integrate neuroethics into the engineering process. Intentionally aligning societal and engineering goals within this framework offers a way to maximize the positive impact of next-generation neurotechnologies on society.

Combining Electrodermal Activity With the Peak-Pain Time to Quantify Three Temporal Regions of Pain Experience.

Background: Self-reported pain levels, while easily measured, are often not reliable for quantifying pain. More objective methods are needed that supplement self-report without adding undue burden or cost to a study. Methods that integrate multiple measures, such as combining self-report with physiology in a structured and specific-to-pain protocol may improve measures. Method: We propose and study a novel measure that combines the timing of the peak pain measured by an electronic visual-analog-scale (eVAS) with continuously-measured changes in electrodermal activity (EDA), a physiological measure quantifying sympathetic nervous system activity that is easily recorded with a skin-surface sensor. The new pain measure isolates and specifically quantifies three temporal regions of dynamic pain experience: I. Anticipation preceding the onset of a pain stimulus, II. Response rising to the level of peak pain, and III. Recovery from the peak pain level. We evaluate the measure across two pain models (cold pressor, capsaicin), and four types of treatments (none, A=pregabalin, B=oxycodone, C=placebo). Each of 24 patients made four visits within 8 weeks, for 96 visits total: A training visit (TV), followed by three visits double-blind presenting A, B, or C (randomized order). Within each visit, a participant experienced the cold pressor, followed by an hour of rest during which one of the four treatments was provided, followed by a repeat of the cold pressor, followed by capsaicin. Results: The novel method successfully discriminates the pain reduction effects of the four treatments across both pain models, confirming maximal pain for no-treatment, mild pain reduction for placebo, and the most pain reduction with analgesics. The new measure maintains significant discrimination across the test conditions both within a single-day's visit (for relative pain relief within a visit) and across repeated visits spanning weeks, reducing different-day-physiology affects, and providing better discriminability than using self-reported eVAS. Conclusion: The new method combines the subjectively-identified time of peak pain with capturing continuous physiological data to quantify the sympathetic nervous system response during a dynamic pain experience. The method accurately discriminates, for both pain models, the reduction of pain with clinically effective analgesics.

Deepfake detection by human crowds, machines, and machine-informed crowds.

The recent emergence of machine-manipulated media raises an important societal question: How can we know whether a video that we watch is real or fake? In two online studies with 15,016 participants, we present authentic videos and deepfakes and ask participants to identify which is which. We compare the performance of ordinary human observers with the leading computer vision deepfake detection model and find them similarly accurate, while making different kinds of mistakes. Together, participants with access to the model's prediction are more accurate than either alone, but inaccurate model predictions often decrease participants' accuracy. To probe the relative strengths and weaknesses of humans and machines as detectors of deepfakes, we examine human and machine performance across video-level features, and we evaluate the impact of preregistered randomized interventions on deepfake detection. We find that manipulations designed to disrupt visual processing of faces hinder human participants' performance while mostly not affecting the model's performance, suggesting a role for specialized cognitive capacities in explaining human deepfake detection performance.

Prospective Study of a Multimodal Convulsive Seizure Detection Wearable System on Pediatric and Adult Patients in the Epilepsy Monitoring Unit.

Background: Using machine learning to combine wrist accelerometer (ACM) and electrodermal activity (EDA) has been shown effective to detect primarily and secondarily generalized tonic-clonic seizures, here termed as convulsive seizures (CS). A prospective study was conducted for the FDA clearance of an ACM and EDA-based CS-detection device based on a predefined machine learning algorithm. Here we present its performance on pediatric and adult patients in epilepsy monitoring units (EMUs). Methods: Patients diagnosed with epilepsy participated in a prospective multi-center clinical study. Three board-certified neurologists independently labeled CS from video-EEG. The Detection Algorithm was evaluated in terms of Sensitivity and false alarm rate per 24 h-worn (FAR) on all the data and on only periods of rest. Performance were analyzed also applying the Detection Algorithm offline, with a less sensitive but more specific parameters configuration ("Active mode"). Results: Data from 152 patients (429 days) were used for performance evaluation (85 pediatric aged 6-20 years, and 67 adult aged 21-63 years). Thirty-six patients (18 pediatric) experienced a total of 66 CS (35 pediatric). The Sensitivity (corrected for clustered data) was 0.92, with a 95% confidence interval (CI) of [0.85-1.00] for the pediatric population, not significantly different (p > 0.05) from the adult population's Sensitivity (0.94, CI: [0.89-1.00]). The FAR on the pediatric population was 1.26 (CI: [0.87-1.73]), higher (p < 0.001) than in the adult population (0.57, CI: [0.36-0.81]). Using the Active mode, the FAR decreased by 68% while reducing Sensitivity to 0.95 across the population. During rest periods, the FAR's were 0 for all patients, lower than during activity periods (p < 0.001). Conclusions: Performance complies with FDA's requirements of a lower bound of CI for Sensitivity higher than 0.7 and of a FAR lower than 2, for both age groups. The pediatric FAR was higher than the adult FAR, likely due to higher pediatric activity. The high Sensitivity and precision (having no false alarms) during sleep might help mitigate SUDEP risk by summoning caregiver intervention. The Active mode may be advantageous for some patients, reducing the impact of the FAR on daily life. Future work will examine the performance and usability outside of EMUs.