EZMedTech examines practical applications of artificial intelligence in healthcare, grounded in clinical evidence and real-world implementation. We focus on measurable impact across workflows that matter to interdisciplinary clinical teams, from documentation efficiency to medication safety and operational performance.

In each edition, we highlight emerging AI developments, clinically relevant tools, and the practical implications for patient outcomes, workflow efficiency, and care delivery, along with future directions shaping the next phase of clinical innovation.

A new study from the Cleveland Clinic and Dyania Health demonstrates how medically trained large language models can significantly improve the identification of eligible patients for rare disease clinical trials through automated electronic medical record (EMR) review. Published in the Journal of Cardiac Failure, the system screened 1,476 patient charts in one week for a Phase 3 trial targeting transthyretin amyloid cardiomyopathy (ATTR-CM) and achieved 96.2% accuracy across more than 7,700 trial eligibility questions. Notably, 29 of the 30 AI-identified trial matches had not been detected through traditional recruitment methods, and the AI-assisted process enrolled seven patients in six days compared to ten patients over 90 days through standard screening. The model also improved diversity in recruitment, identifying a significantly higher proportion of Black patients and individuals not previously connected to heart failure specialists. By combining structured EMR data with natural language processing of clinical notes and generating auditable reasoning for each eligibility decision, the platform highlights how AI can accelerate trial enrollment while expanding access to underrepresented patient populations.

Takeaway:
AI-powered chart review may help solve one of clinical research’s biggest bottlenecks, efficiently and equitably matching patients to trials, potentially accelerating enrollment timelines while improving representation in rare disease studies

Butterfly iQ3 by Butterfly Network is a portable, AI-powered point-of-care ultrasound device designed to bring medical imaging directly to the bedside. The handheld probe connects to a smartphone or tablet and uses semiconductor-based ultrasound technology and built-in AI tools to guide clinicians in capturing and interpreting scans in real time. Features such as automated image acquisition, anatomical presets, and AI-assisted analysis help clinicians identify structures and assess conditions across multiple organs with a single probe. By enabling high-quality imaging outside traditional radiology settings, such as emergency departments, ambulances, rural clinics, and mobile care environments, the device aims to make diagnostic ultrasound faster, more accessible, and easier for clinicians with varying levels of ultrasound experience.

Takeaway:
AI-guided handheld ultrasound devices like Butterfly iQ3 are helping democratize medical imaging by enabling rapid point-of-care diagnostics in settings where traditional ultrasound systems may not be available.

The product answers patient calls, schedules, reschedules, and cancels appointments, sends reminders, verifies insurance eligibility before visits, and follows up with patients on unpaid balances via voice and SMS, reducing missed calls, no-shows, and delayed collections.

Ezmedtech integrates natively with major EMR, PMS, and RCM systems, including Athena, eClinicalWorks, Elation, Tebra, Open Dental, Dentrix, Dentrix Ascend, pVerify, Google Calendar, and veterinary PIMS systems, allowing clinics to deploy the product without changing existing workflows.