Quick Buzz Feed
JMIR Publications released two feature News and Perspectives articles covering health tech research developments, and one investigating the rising use of AI in medical education.
Katie Cottingham's article, "Charting a Course for AI in Medical Education," examines how medical schools are integrating AI into their curricula to prepare students for a technology-driven clinical environment. While AI tools offer benefits like personalized and immediate feedback, experts express concerns about "never-skilling," "mis-skilling," and "de-skilling," where over-reliance on AI might prevent students and professionals from developing essential critical thinking abilities. Cottingham includes insights from educators and a medical student, advocating for AI tools to scaffold learning rather than replace fundamental skill development, emphasizing that the human brain requires active engagement with information for deep retention.
Simon Spichak's piece, "AI Models Could Improve Diagnosis and Care for Rare Diseases," highlights the potential of novel machine learning tools to accelerate the detection and treatment discovery for rare diseases. These conditions affect up to 446 million people globally but suffer from limited data and underfunding. Dr. Rose Orenbuch from Harvard details her work with popEVE, a deep generative model that has successfully identified 123 new genetic variants potentially linked to severe developmental disorders. The article notes that larger "frontier and foundation models" could generalize insights from vast datasets to data-scarce rare disease research, potentially aiding diagnosis and drug discovery, though their real-world efficacy is still largely unproven.
Liam Critchley reports on advancements in using electronic noses (e-noses) for non-invasive cancer screening in his article, "Advanced Olfactory Cancer Detection: When E-Noses Sniff the Skin." He discusses a pilot study published in the Journal of Analytical Chemistry where a new quantum-dot-based e-nose, utilizing cadmium sulfide nanocrystals, was developed to detect volatile organic compounds exhaled through the skin. Researchers successfully identified potential biomarkers for the presence or absence of malignant tumors from this data. This highly sensitive device demonstrated 100% accuracy and sensitivity in distinguishing cancer patients from healthy individuals and showed promise in classifying disease severity.