JMIR Human Factors

Robotic-assisted surgery (RAS) is increasingly being integrated into surgical services worldwide. Successful implementation of robotic surgical technologies depends not only on clinical effectiveness but also on public trust, perceptions of safety, and willingness to use technology-enabled care. In the United Arab Emirates, the adoption of RAS is expanding rapidly; however, evidence regarding public perceptions and factors influencing acceptance remains limited.

AI is now embedded in the infrastructure of perioperative care. Risk stratification algorithms, hemodynamic prediction tools, and clinical decision support systems are active in operating rooms at major health systems, and their adoption is accelerating. However, the field has studied model performance and organizational implementation while largely bypassing the moment between them: the real-time encounter in which an anesthesia provider must decide, under active case conditions, what to do with an AI-generated output. We term this the cognitive transaction and argue that it is the fundamental unit of perioperative AI implementation. The perioperative environment presents a specific constellation of conditions that existing human-AI interaction research was not designed to address. Continuous real-time decision demands, extreme time compression, high cognitive load, and consequences that unfold in seconds distinguish the operating room from the clinical contexts where most provider-AI interaction research has been conducted. What we know about AI adoption in radiology, oncology, or ambulatory care does not readily translate to this setting. The cognitive moment in anesthesia has its own structure, its own failure modes, and its own research requirements. This paper examines what those requirements are. We analyze how the operating room functions as a pre-existing human-machine cognitive system into which AI is now being inserted, and why the conditions of that system generate predictable vulnerabilities: miscalibrated trust, automation bias, and cognitive friction produced by interfaces optimized for technical accuracy rather than clinical usability. We argue that these failure modes are not incidental but structural and that they will persist regardless of model performance until the provider-AI interaction is itself treated as a research object. We identify 4 priority research domains. The first concerns the structure of provider-AI disagreement and the methods needed to distinguish automation bias from legitimate clinical insight. The second concerns the longitudinal dynamics of trust calibration across repeated clinical encounters rather than single-session experimental designs. The third concerns interface design for high-acuity workflows, specifically what constitutes a usable AI output for a provider managing a patient in real time. The fourth concerns the need for ecologically valid study designs capable of capturing provider reasoning under actual perioperative conditions rather than retrospective or survey-based proxies. The anesthesia and perioperative research community is positioned to lead this work. The clinical specificity, domain knowledge, and professional stake required to design meaningful studies are all present within the field. Evaluating the cognitive transaction under perioperative conditions, not the computational model in isolation, is both a methodological imperative and a patient safety priority.

SSEndoSim, a free, open access, mobile, web-based endodontic diagnostic simulation application for undergraduate dental education, demonstrated favorable usability rated on a System Usability Scale (SUS) witht a composite 100-point score (mean 75.30, SD 14.72) and high perceived educational value (PEV) on a 5-point scale (mean 4.62, SD 0.40) in a cross-sectional usability evaluation of final-year bachelor of dental surgery (BDS) students at Liaquat University of Medical and Health Sciences (LUMHS), Jamshoro, Pakistan, in April 2026, establishing usability and learner acceptability.

mHealth (mobile health) interventions that integrate psychoeducation with structured problem-solving training (PST) hold strong potential for improving self-management of chronic conditions. Evaluating the usability of these interventions requires assessing technological, pedagogical, and sociocultural fit. However, most usability evaluations remain narrowly technocentric, focusing on interface-level metrics while neglecting pedagogical coherence, cultural responsiveness, and patient learning needs.

Neonatal mortality remains a leading contributor to under-5 deaths globally, particularly in low- and middle-income countries (LMICs). While machine learning (ML)–based risk prediction models show promise for identifying high-risk neonates, published evidence describing the real-world usability and practical implementation of ML-derived neonatal risk prediction tools within routine clinical workflows in LMIC neonatal units remains limited.

After the emergence of telemedicine as a major health care delivery mode during the COVID-19 pandemic, Sheba Medical Center (SMC) rapidly transitioned its outpatient clinics to telemedicine. A postimplementation satisfaction study revealed that while nearly 90% of patients reported a positive experience, only 38% of clinicians expressed high satisfaction. SMC soon established Sheba BEYOND, a dedicated virtual hospital. This initiative was accompanied by extensive efforts to improve the telemedicine experience. In 2025, a follow-up survey addressing clinician satisfaction and acceptance was conducted.

Self-injurious thoughts and behaviors, including suicidal ideation, planning, attempts, and nonsuicidal self-injury, are a significant public health concern among college students. Community college students face elevated mental health risks and persistent barriers to traditional services, particularly those from underserved and ethnoracially minoritized backgrounds, underscoring the need for scalable, accessible digital mental health interventions.

Large language models can generate fluent summaries of longitudinal medical records, but in high-stakes clinical settings, verification burden remains a barrier to trust. Existing provenance mechanisms such as document-level citations and section references often require manual search within long, fragmented notes, limiting their usefulness during time-constrained workflows for clinicians.

Virtual clinics allow doctors to connect to patients in difficult-to-reach locations. While this can result in improved access to care, implementing existing virtual clinics in these locations is difficult due to their contextual constraints. To address these challenges, a virtual clinic system for remote, rural, and underserved areas was developed based on user-centered design principles. To complete the user-centered design cycle, the developed system was then implemented and evaluated in the target contexts.

Mental health challenges such as stress and depression are prevalent among employees. Mobile health platforms that deliver exercise or yoga interventions offer a promising approach to improve mental health outcomes in this population.

Despite increasing recognition of human factors in technologically advanced manufacturing, psychological resilience remains an underexplored dimension. Existing studies often rely on generic instruments that fail to capture the specific adaptive challenges faced by workers in high-tech environments. There is a growing need for context-sensitive tools capable of assessing resilience in line with the human-centric vision of Industry 5.0.

Patient safety classification systems are fundamental to surveillance, organizational learning, research, and governance because they enable adverse events and near misses to be organized into standardized categories for comparison and analysis. However, the increasing complexity of health information technology (HIT)–related patient safety incidents challenges the assumptions underpinning conventional classification approaches, as these incidents often emerge from dynamic, distributed, and evolving sociotechnical interactions rather than discrete, time-bounded events. In this Viewpoint, I argue that many of the challenges associated with classifying HIT-related patient safety incidents arise not simply from limitations of individual classification systems but from the inherent representational logic of classification itself. By viewing classification as a knowledge practice rather than merely a technical tool for organizing incident data, I contend that abstraction, boundary-setting, and standardization inevitably simplify complex sociotechnical processes and constrain how safety problems are represented, interpreted, and acted upon. I discuss 4 recurring representational limitations that characterize the application of patient safety classification systems to HIT-related incidents: fragmentation of sociotechnical interactions, loss of temporality and evolving processes, inadequate representation of scale and propagation across systems, and normalization of “use error” through simplified attribution of responsibility. These limitations can contribute to incomplete organizational learning, misaligned safety interventions, and challenges in interpreting and comparing classified patient safety data across health care settings. Rather than arguing against the continued use of patient safety classification systems, I propose that their strengths and limitations should be recognized simultaneously. Classification remains indispensable for surveillance, learning, and governance, but it should be interpreted as one component of a broader sociotechnical understanding of patient safety. Recognizing the representational limits of classification can support more reflexive interpretation of classification-based evidence and encourage complementary approaches that better capture the complexity of HIT-related patient safety.