AI Watch
Remote patient monitoring was once a simple data-collection exercise — a blood pressure reading transmitted, a nurse alerted if the number crossed a threshold. Artificial intelligence is fundamentally changing that model, shifting from reactive alerts to individualized predictive risk scoring that can flag deterioration before any vital sign crosses a static cutoff.
From thresholds to predictions
Today’s AI-enabled RPM platforms build individualized baselines for each enrolled patient and apply machine learning to detect patterns that precede clinical events — not just values that exceed fixed limits. A 2026 preprint published on arXiv demonstrated an autonomous AI triage agent that compressed clinician review time from days to minutes in a live RPM deployment, achieving validated accuracy across a diverse patient population. A separate multi-modal AI study in cancer care RPM collected 2.1 million data points across 6,000 patient-days and achieved 84% accuracy (AUC 0.70) combining wearable sensor data, daily patient-reported surveys, and clinical event history.
The market reflects this clinical momentum. The AI-in-RPM sector was valued at approximately $2 billion in 2024 and is projected to grow at a 27% compound annual rate through 2032, reaching an estimated $13 billion. Hospitals integrating AI-assisted workflow automation into RPM programs report 20–30% reductions in routine clinical task burden — freeing clinician capacity for patients with the highest acuity.
The FDA and CMS regulatory landscape
The FDA has cleared more than 950 AI/ML-enabled medical devices as of mid-2024 — up from fewer than 400 in 2020. In 2025 alone, 295 AI/ML devices received 510(k) clearance. In December 2024, the FDA finalized Predetermined Change Control Plan (PCCP) guidance, allowing manufacturers to pre-specify algorithm update pathways post-market — a significant evolution for adaptive AI-RPM software that learns from new patient data over time.
On the payment side, CMS’s 2024 Physician Fee Schedule confirmed that RPM can be billed alongside Chronic Care Management without double-counting. The 2026 Hospital OPPS Final Rule established national Medicare reimbursement for AI-assisted cardiac analysis. Most significantly, CMS’s new ACCESS model — launching July 2026 — uses AI diagnostics to identify chronic disease patients suited for technology-enabled care management, directly tying Medicare payment to measurable health outcomes rather than service volume.
Policy implications
For Medicare beneficiaries managing heart failure, COPD, hypertension, or diabetes, AI-enhanced RPM offers a credible path to fewer emergency visits and reduced hospitalizations. But the clinical promise depends entirely on whether the algorithms powering these systems were trained on data that accurately represents the patients they serve — a question that CMS, FDA, and Congress have not yet answered with binding standards. See our companion post on the algorithmic bias problem in AI-assisted monitoring.