A Human-Centered Systems Approach to AI-Enhanced VR Training for Home-Based Peritoneal Dialysis

AI for UI: Designing Error-tolerant Interfaces for Home Dialysis using Artificial Intelligence

GenAI-assisted processes can provide rapid, actionable design mitigations that reduce error likelihood and enhance patient autonomy, establishing a replicable pipeline for producing heuristic-driven design libraries across diverse medical device contexts.

Beyond the Therapeutic Cliff: A Clinical Engineering Framework for AI-Driven Home-Based Neurorehabilitation in Rural Veterans

The therapeutic cliff, defined as the steep drop in rehabilitation intensity once a patient leaves inpatient care, is a documented driver of poor functional recovery after stroke and traumatic brain injury. For rural veterans, distance to specialised care, broadband disparity, and fragmented access amplify the drop. We propose a clinical engineering framework for AI-driven home-based neurorehabilitation grounded in SEIPS 2.0 sociotechnical principles. The framework integrates a wearable inertial measurement layer, an artificial intelligence exercise-quality and dose-tracking layer, a fall-prediction subsystem, a caregiver interface layer, and a Veterans Health Administration Care Coordination Home Telehealth integration layer. We synthesise eighteen peer-reviewed sources, map architecture components to validated evidence, and specify deployment constraints for rural cohorts including a bandwidth-tiered profile and a polytrauma-compatible cognitive guardrail. Validation pathways and policy implications for telerehabilitation reimbursement parity are outlined.

Hybrid Rehabilitation Models Combining In-Clinic and Home-Based Therapy for Complex Neurological Conditions

Hybrid rehabilitation models combining in-clinic and home-based therapy address resource constraints and the heterogeneous needs of patients with complex neurological conditions. Each arm has independent support: telerehabilitation, early supported discharge, and caregiver-mediated exercise carry Cochrane-level evidence, while in-clinic intensity and task-specific training remain standard for motor recovery. No integrative framework, however, prescribes how dose, content, timing, and outcome measurement should be allocated across settings when stroke, traumatic brain injury, progressive disease, and fall risk overlap in a single patient. This paper synthesises fifteen high-quality references and proposes a five-component hybrid framework: phase-gate architecture, dose-allocation logic, caregiver-integration tiers, a remote-capable outcomes battery, and safety guardrails. Application to stroke, polytrauma, progressive neurological disease, and geriatric falls populations is mapped. Implications for civilian home-health physical therapy practice and cost-effectiveness modelling are discussed.

Human, hybrid and automated education in paramedic practice

Which elements of paramedic training should remain inherently human, which are best delivered through hybrid human–AI models, and which tasks can be ethically and safely automated with oversight are examined.

The error room : a simulation-based tool for enhancing patient in critical care

An Interactive, Project-Based AI in Healthcare Curriculum for Clinicians and Medical Educators

This curriculum offers a generalizable, no-cost model for closing the AI evaluation skills gap in healthcare education, combining interactive correction with a mandatory, learner-defined capstone project.