Faculty Bibliography

Increasingly, laws are being proposed and passed by governments around the world to regulate artificial intelligence (AI) systems implemented into the public and private sectors. Many of these regulations address the transparency of AI systems, and related citizen-aware issues like allowing individuals to have the right to an explanation about how an AI system makes a decision that impacts them. Yet, almost all AI governance documents to date have a significant drawback: they have focused on what to do (or what not to do) with respect to making AI systems transparent, but have left the brunt of the work to technologists to figure out how to build transparent systems. We fill this gap by proposing a stakeholder-first approach that assists technologists in designing transparent, regulatory-compliant systems. We also describe a real-world case study that illustrates how this approach can be used in practice.

BACKGROUND:The surge of telemedicine use during the early stages of the coronavirus-19 (COVID-19) pandemic has been well documented. However, scarce evidence considers the utilization of telemedicine in the subsequent period. OBJECTIVE:This study aims to evaluate utilization patterns of video-based telemedicine visits for ambulatory care and urgent care provision over the course of recurring pandemic waves in one large health system in New York City, and what this means for healthcare delivery. METHODS:Retrospective electronic health record (EHR) data of patients between January 1st, 2020, and February 28th, 2022 were used to longitudinally track and analyze telemedicine and in-person visit volumes across ambulatory care specialties and urgent care, as well as compare them to a pre-pandemic baseline (June to November 2019). Diagnosis codes to differentiate COVID-19 suspected visits from non-COVID-19 visits, as well as evaluating COVID-19 based telemedicine utilization over time, were compared to the total number of COVID-19 positive cases in the same geographic region (city-level). The time-series data was segmented based on change-point analysis and variances in visit trends were compared between the segments. RESULTS:The emergence of COVID-19 prompted an early increase in the number of telemedicine visits across the urgent care and ambulatory care settings. This utilization continued throughout the pandemic at a much higher level than the pre-pandemic baseline for both COVID-19 and non-COVID suspected visits, despite fluctuation in COVID-19 cases throughout the pandemic and the resumption of in-person clinical services. Utilization of telemedicine-based urgent care services for COVID-19 suspected visits showed more variance in response to each pandemic wave, but telemedicine visits for ambulatory care have remained relatively steady after the initial crisis period. During the Omicron wave, the utilization of all visit types including in-person activities decreased. Patients between 25 and 34 years of age were the largest users of telemedicine-based urgent care. Patient satisfaction with telemedicine-based urgent care remained high despite the rapid scaling of services to meet increased demand. CONCLUSIONS:The trend of increased use of telemedicine as a means of healthcare delivery relative to the pre-COVID-19 baseline has been maintained throughout the later pandemic periods despite fluctuating COVID-19 cases and the resumption of in-person care delivery. Overall satisfaction with telemedicine-based care is also high. The trends in telemedicine utilization suggest that telemedicine-based healthcare delivery has become a mainstream and sustained supplement to in-person-based ambulatory care, particularly for younger patients, for both urgent and non-urgent care needs. These findings have implications for the healthcare delivery system, including practice leaders, insurers, and policymakers. Further investigation is needed to evaluate telemedicine adoption by key demographics, identify ongoing barriers to adoption, and explore the impacts of sustained use of telemedicine on healthcare outcomes and experience.

BACKGROUND:Telemedicine as a mode of health care work has grown dramatically during the COVID-19 pandemic; the impact of this transition on clinicians' after-hours electronic health record (EHR)-based clinical and administrative work is unclear. OBJECTIVE:This study assesses the impact of the transition to telemedicine during the COVID-19 pandemic on physicians' EHR-based after-hours workload (ie, "work outside work") at a large academic medical center in New York City. METHODS:We conducted an EHR-based retrospective cohort study of ambulatory care physicians providing telemedicine services before the pandemic, during the acute pandemic, and after the acute pandemic, relating EHR-based after-hours work to telemedicine intensity (ie, percentage of care provided via telemedicine) and clinical load (ie, patient load per provider). RESULTS:A total of 2129 physicians were included in this study. During the acute pandemic, the volume of care provided via telemedicine significantly increased for all physicians, whereas patient volume decreased. When normalized by clinical load (ie, average appointments per day by average clinical days per week), telemedicine intensity was positively associated with work outside work across time periods. This association was strongest after the acute pandemic. CONCLUSIONS:Taking physicians' clinical load into account, physicians who devoted a higher proportion of their clinical time to telemedicine throughout various stages of the pandemic engaged in higher levels of EHR-based after-hours work compared to those who used telemedicine less intensively. This suggests that telemedicine, as currently delivered, may be less efficient than in-person-based care and may increase the after-hours work burden of physicians.

BACKGROUND:Sustained engagement is essential for the success of telerehabilitation programs. However, patients' lack of motivation and adherence could undermine these goals. To overcome this challenge, physical exercises have often been gamified. Building on the advantages of serious games, we propose a citizen science-based approach in which patients perform scientific tasks by using interactive interfaces and help advance scientific causes of their choice. This approach capitalizes on human intellect and benevolence while promoting learning. To further enhance engagement, we propose performing citizen science activities in immersive media, such as virtual reality (VR). OBJECTIVE:This study aims to present a novel methodology to facilitate the remote identification and classification of human movements for the automatic assessment of motor performance in telerehabilitation. The data-driven approach is presented in the context of a citizen science software dedicated to bimanual training in VR. Specifically, users interact with the interface and make contributions to an environmental citizen science project while moving both arms in concert. METHODS:In all, 9 healthy individuals interacted with the citizen science software by using a commercial VR gaming device. The software included a calibration phase to evaluate the users' range of motion along the 3 anatomical planes of motion and to adapt the sensitivity of the software's response to their movements. During calibration, the time series of the users' movements were recorded by the sensors embedded in the device. We performed principal component analysis to identify salient features of movements and then applied a bagged trees ensemble classifier to classify the movements. RESULTS:The classification achieved high performance, reaching 99.9% accuracy. Among the movements, elbow flexion was the most accurately classified movement (99.2%), and horizontal shoulder abduction to the right side of the body was the most misclassified movement (98.8%). CONCLUSIONS:Coordinated bimanual movements in VR can be classified with high accuracy. Our findings lay the foundation for the development of motion analysis algorithms in VR-mediated telerehabilitation.

With recurring waves of the Covid-19 pandemic, a dilemma facing public health leadership is whether to provide public advice that is medically optimal (e.g., most protective against infection if followed), but unlikely to be adhered to, or advice that is less protective but is more likely to be followed. To provide insight about this dilemma, we examined and quantified public perceptions about the tradeoff between (a) the stand-alone value of health behavior advice, and (b) the advice's adherence likelihood. In a series of studies about preference for public health leadership advice, we asked 1061 participants to choose between (5) strict advice that is medically optimal if adhered to but which is less likely to be broadly followed, and (2) relaxed advice, which is less medically effective but more likely to gain adherence-given varying infection expectancies. Participants' preference was consistent with risk aversion. Offering an informed choice alternative that shifts volition to advice recipients only strengthened risk aversion, but also demonstrated that informed choice was preferred as much or more than the risk-averse strict advice.

Objective/UNASSIGNED:The widespread deployment of electronic health records (EHRs) has introduced new sources of error and inefficiencies to the process of ordering medications in the hospital setting. Existing work identifies orders that require pharmacy intervention by comparing them to a patient's medical records. In this work, we develop a machine learning model for identifying medication orders requiring intervention using only provider behavior and other contextual features that may reflect these new sources of inefficiencies. Materials and Methods/UNASSIGNED:Data on providers' actions in the EHR system and pharmacy orders were collected over a 2-week period in a major metropolitan hospital system. A classification model was then built to identify orders requiring pharmacist intervention. We tune the model to the context in which it would be deployed and evaluate global and local feature importance. Results/UNASSIGNED:The resultant model had an area under the receiver-operator characteristic curve of 0.91 and an area under the precision-recall curve of 0.44. Conclusions/UNASSIGNED:Providers' actions can serve as useful predictors in identifying medication orders that require pharmacy intervention. Careful model tuning for the clinical context in which the model is deployed can help to create an effective tool for improving health outcomes without using sensitive patient data.

BACKGROUND:Self-focused augmented reality (AR) technologies are growing in popularity and present an opportunity to address health communication and behavior change challenges. OBJECTIVE:We aimed to examine the impact of self-focus AR and vicarious reinforcement on psychological predictors of behavior change during the COVID-19 pandemic. In addition, our study included measures of fear and message minimization to assess potential adverse reactions to the design interventions. METHODS:A web-based between-subjects experiment (n = 335) was conducted to compare the health perceptions of participants in self-focus AR & vicarious reinforcement design conditions to those in a control condition. RESULTS:We found that participants who experienced self-focus AR & vicarious reinforcement scored higher in perceived threat severity (P = 0.03) and susceptibility (P = 0.01) when compared to the control. A significant indirect effect of self-focus AR & vicarious reinforcement on intention was found with perceived threat severity as a mediator (b = .06, 95% CI= [.02, .12], SE = .02). Self-focus AR & vicarious reinforcement did not result in higher levels of fear (P = 0.32) or message minimization (P = 0.42) when compared to the control. CONCLUSIONS:Augmenting one's reflection with vicarious reinforcement may be an effective strategy for health communication designers. While our study's results did not show adverse effects in regards to fear and message minimization, utilization of self-focus AR should be done with care due to possible adverse effects of heightened levels of fear as a health communication strategy. CLINICALTRIAL/UNASSIGNED/:

Psychological ownership defines how we behave in and interact with the social world and the objects around us. Shared Augmented Reality (shared AR) may challenge conventional understanding of psychological ownership because virtual objects created by one user in a social place are available for other participants to see, interact with, and edit. Moreover, confusion may arise when one user attaches a virtual object in a shared AR environment onto the physical object that is owned by a different user. The goal of this study is to investigate tensions around psychological ownership in shared AR. Drawing on prior work, we developed a conceptualization of psychological ownership in shared AR in terms of five underlying dimensions: possession, control, identity, responsibility, and territoriality. We studied several shared AR scenarios through a laboratory experiment that was intended to highlight normative tensions. We divided participants into pairs, whereby one participant in each pair created the virtual object (object-creator) and placed it over the other person's (space proprietor) physical object or space. We recorded participants"™ perceptions of psychological ownership along the 5 dimensions through surveys and interviews. Our results reveal that the paired participants failed to form a mutual understanding of ownership over the virtual objects. In addition, the introduction of virtual objects called into question participants"™ sense of psychological ownership over the physical articles to which the virtual objects were attached. Building on our results, we offer a set of design principles for shared AR environments, intended specifically to alleviate psychological ownership-related concerns. Herein, we also discuss the implications of our findings for research and practice in this field.