Healthcare AI Weekly

I’m Raphael Malikian, a healthcare AI builder with 20+ years of domain experience, from family medicine practice to bootstrapping a direct primary care startup. This channel documents my healthcare-AI pivot: building in public, sharing insights on clinical workflows, digital health, LLM evaluation, automation, research ops, and turning real projects into a practical portfolio. Every week I release Healthcare AI Weekly which is a journal club discussing an important peer-reviewed healthcare AI article, and one you might have missed. Do you have a healthcare problem that you help with solving? Reach out. My email is [email protected] Links Github github.com/rtmalikian LinkedIn linkedin.com/in/raphael-t-malikian-mbbs-bsc-hons-71075436a --- This week in Healthcare AI Weekly: a source-grounded look at two new peer-reviewed healthcare AI papers. 1) Bavali-Gazik et al. — Developing and validating an AI-based electronic triage model for cardiac-suspected ED patients. PMID: 42182049; PMCID: PMC13195029. 2) Hiratsuka et al. — AI/ML in Alaska Native healthcare systems: symposium perspectives. PMID: 42176020; PMCID: PMC13202658. Comment question: which layer matters most for healthcare AI right now — model performance, workflow fit, or governance? Synthetic voice disclosure: narration uses edge-tts en-US-AndrewNeural synthetic voice for review production. Watch this episode on YouTube: https://youtu.be/gQbft5MLFTA YouTube Channel: https://www.youtube.com/@RaphaelMalikian-g4h Created by Raphael T. Malikian ([email protected]). In true AI fashion, this podcast was created with AI tools including text-to-speech using Microsoft Edge TTS and Hermes Agent by Nous Research.

Healthcare AI Weekly - your bridge between clinical medicine and AI research. This week we examine two critical healthcare AI papers: Paper 1 - AI-powered skin cancer detection in clinical settings Paper 2 - The deskilling effect: how AI assistance in colonoscopy may reduce physician competency over time One paper shows AI outperforming clinicians in detection. The other raises an uncomfortable question: what happens to human skill when we rely on AI too much? MEDICAL DISCLAIMER: This podcast is for educational and informational purposes only. It is not intended to diagnose, treat, cure, or prevent any medical condition. Nothing discussed in this episode constitutes medical advice. If you have a medical concern or health problem, you must seek attention from a licensed medical professional immediately. Created by Raphael T. Malikian ([email protected]). In true AI fashion, this podcast was created with AI tools including text-to-speech using Microsoft Edge TTS and Hermes Agent by Nous Research. #HealthcareAI #ClinicalAI #MedicalAI #AIinMedicine #DigitalHealth #MachineLearning #HealthTech #AIResearch #ClinicalDecisionSupport #HealthcareInnovation Watch this episode on YouTube: https://youtu.be/kauhiLUZJrA YouTube Channel: https://www.youtube.com/@RaphaelMalikian-g4h Created by Raphael T. Malikian ([email protected]). In true AI fashion, this podcast was created with AI tools including text-to-speech using Microsoft Edge TTS and Hermes Agent by Nous Research.

Healthcare AI Weekly - your bridge between clinical medicine and AI research. This week we cover two papers on AI detecting what current methods miss: Paper 1 - RCAID: Renal Cell Carcinoma AI Detector Huang C, Wang G, Yuan Y, et al. European Urology (2026) A plasma metabolomic model using 7 blood metabolites detected kidney cancer with AUROCs of 0.91-0.99 across 6 validation cohorts. Paper 2 - Video-based Detection of Delirium in Hospitalized Adults Mendu M, Tesh RA, Pellerin K, et al. PLOS Digital Health (2026) A video-based AI system using deep learning pose estimation detected delirium with AUC 0.79. MEDICAL DISCLAIMER: This podcast is for educational and informational purposes only. It is not intended to diagnose, treat, cure, or prevent any medical condition. Nothing discussed in this episode constitutes medical advice. If you have a medical concern or health problem, you must seek attention from a licensed medical professional immediately. Created by Raphael T. Malikian ([email protected]). In true AI fashion, this podcast was created with AI tools including text-to-speech using Microsoft Edge TTS and Hermes Agent by Nous Research. #HealthcareAI #ClinicalAI #MedicalAI #AIinMedicine #DigitalHealth #MachineLearning #HealthTech #AIResearch #ClinicalDecisionSupport #HealthcareInnovation Watch this episode on YouTube: https://youtu.be/E9EPRWN89LM YouTube Channel: https://www.youtube.com/@RaphaelMalikian-g4h Created by Raphael T. Malikian ([email protected]). In true AI fashion, this podcast was created with AI tools including text-to-speech using Microsoft Edge TTS and Hermes Agent by Nous Research.

Welcome to Healthcare AI Weekly with Raphael Malikian. This episode breaks down two recent peer-reviewed healthcare AI papers: one major paper on large language models for diagnostic reasoning in epilepsy, and one overlooked but important paper on nurse concern signals for inpatient deterioration. Papers covered: 1. "Evaluating large language models for diagnostic reasoning from unstructured clinical narratives in epilepsy" - Dani et al. 2. "Enhancing prediction of inpatient deterioration by combining nursing concern signals with machine learning" Both papers explore how AI can augment clinical judgment - one at the diagnostic reasoning level, the other at the bedside monitoring level. MEDICAL DISCLAIMER: This podcast is for educational and informational purposes only. It is not intended to diagnose, treat, cure, or prevent any medical condition. Nothing discussed in this episode constitutes medical advice. If you have a medical concern or health problem, you must seek attention from a licensed medical professional immediately. Created by Raphael T. Malikian ([email protected]). In true AI fashion, this podcast was created with AI tools including text-to-speech using Microsoft Edge TTS and Hermes Agent by Nous Research. #HealthcareAI #ClinicalAI #MedicalAI #AIinMedicine #DigitalHealth #MachineLearning #HealthTech #AIResearch #ClinicalDecisionSupport #HealthcareInnovation YouTube Channel: https://www.youtube.com/@RaphaelMalikian-g4h Created by Raphael T. Malikian ([email protected]). In true AI fashion, this podcast was created with AI tools including text-to-speech using Microsoft Edge TTS and Hermes Agent by Nous Research.