Healthcare AI Daily

Right now, doctors wait one to three days to learn which antibiotics will work for a serious infection. A new deep learning system called ANTIBIOTIC reads mass spectrometry data hospitals already collect and predicts resistance in minutes. Antimicrobial resistance kills over a million people a year. The core problem is speed - conventional testing takes one to three days. During that window, doctors prescribe broad-spectrum antibiotics, driving more resistance. Published in NPJ Digital Medicine by Wang, Tsao, Hsieh et al. from National Taiwan University Hospital. Key findings: - Bacterial identification AUC: 0.99 (internal), 0.96 (external) - Resistance prediction AUC: 0.94 (internal), 0.61 after fine-tuning on recent data - 89,026 mass spectrometry records, 274 deep learning models - 26 models for bacterial ID, 248 for antibiotic resistance prediction - Integrated chatbot for antibiotic recommendation with kidney-function dosing A resistance AUC of 0.61 is not a slam dunk - but it is decision support, not a replacement for lab culture. The open-source pipeline means any hospital can adapt it locally. Paper: https://doi.org/10.1038/s41746-026-02879-w Like and subscribe for daily Healthcare AI episodes. New videos every day on the latest in clinical AI research. #HealthcareAI #MachineLearning #AntibioticResistance #AI #DeepLearning #ClinicalAI #MALDITOF #DigitalMedicine #AntimicrobialResistance #InfectiousDisease Watch this episode on YouTube: https://youtu.be/FBFC-zD0Vps 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.

What happens when healthcare AI hallucinates? A new systematic review of 44 studies identifies 7 strategies that actually reduce AI errors in medicine. Sanjay Basu and Benjamin Huynh from UC San Francisco and Johns Hopkins reviewed every major approach - from retrieval-augmented generation to human-in-the-loop oversight - and found that combining strategies consistently beats any single method. Key findings: - RAG reduced hallucinations by 30-50% - Human-in-the-loop achieved up to 95% reduction - Combined approaches outperformed all single methods - Proposed MediHall severity scale for prioritizing AI errors Paper: Basu S, Huynh B. Mitigating Hallucinations in Healthcare AI: A Systematic Review of Evidence-Based Strategies. BMC Health Services Research (2026). DOI: 10.1186/s12913-026-14851-1 PMID: 42251377 What strategies has your team tried to reduce AI hallucinations? Share in the comments. Like and subscribe for daily Healthcare AI videos. #HealthcareAI #ArtificialIntelligence #LLM #PatientSafety #AIMedical #RAG #HumanInTheLoop #AISafety #ClinicalAI #HealthInformatics Watch this episode on YouTube: https://youtu.be/nVDuULt65XQ 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.

Can contactless sensors in your home detect cerebrovascular disease before a stroke happens? A new study from South Korea placed IoT sensors in the homes of 1,224 older adults and trained an AI to watch for subtle changes in movement, sleep, and activity patterns. The AI identified people in the pre-stroke phase with 85% precision, and predicted who was within four weeks of diagnosis with 95% sensitivity. Evening activity patterns were the strongest signal. Published in NPJ Digital Medicine (2026). DOI: 10.1038/s41746-026-02836-7 Authors: Baek J, Cho K-H, Lim L, Chong JW Key findings: - 1,224 adults aged 65+, 13,362 two-week observation windows - AUPRC 0.85 for prodromal (pre-diagnosis) identification - AUROC 0.91 for classifying diagnosed patients - 95% sensitivity, 97% specificity for imminent diagnostic risk - Bedtime hours (10 PM - 2 AM) and evening hours (6-10 PM) most informative - Even indoor humidity correlated with cerebrovascular risk This is retrospective research. Prospective validation is the critical next step. What do you think about AI monitoring in the home? Drop your thoughts in the comments. Like and subscribe for daily Healthcare AI episodes. New videos every day on the latest in clinical AI research. #HealthcareAI #Stroke #CerebrovascularDisease #SmartHome #AI #DigitalHealth #MachineLearning #ElderlyCare #IoT #NPJDigitalMedicine #StrokePrevention #RemotePatientMonitoring #AgingInPlace Watch this episode on YouTube: https://youtu.be/uv32UDNGMxs 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.

Raphael T. Malikian, MBBS, BSc (Hons) translates healthcare AI research into practical, clinically grounded questions for builders, clinicians, researchers, and governance teams. GitHub: https://github.com/rtmalikian LinkedIn: http://www.linkedin.com/in/raphael-t-malikian-mbbs-bsc-hons-71075436a --- Healthcare AI Daily translates one healthcare AI paper into a short practical briefing for builders, clinicians, researchers, and governance teams. Today: how three frontier AI models performed when interpreting real blood count reports from patients with blood diseases, and where each one stumbled. Source article Title: Performance Evaluation of GPT-5, Grok 4, and DeepSeek R1 in Interpreting Complete Blood Count Reports for Hematologic Diseases: Retrospective Comparative Study Authors: Xianfei Ye, Xinglun Qi, Lina Fan, Qian Yu, Suming Zhou, Chunyun Ren, Dagan Yang Journal: Journal of Medical Internet Research (JMIR) Published: 5 Jun 2026 DOI: https://doi.org/10.2196/87802 Article: https://www.jmir.org/2026/1/e87802 Keywords: healthcare AI, medical AI, large language models, GPT-5, Grok 4, DeepSeek R1, blood count, CBC, hematology, clinical validation, AI hallucinations, lab medicine, AI safety, clinical deployment. This video is educational commentary, not medical advice. Source screenshots and figures are used for attributed research discussion. Like, subscribe, and enable notifications for daily Healthcare AI episodes. Healthcare AI Weekly releases every Friday at 9 AM. 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.

In rural Brazil, there’s one dermatologist for every 17,000 people. A new AI app that works completely offline just helped health workers catch skin cancers far more accurately. Published in npj Digital Medicine (Nature), Pacheco et al. validated the tool with 131 healthcare professionals across 9 cities. Sensitivity jumped from 64% to 80% with AI assistance. In rural areas, unnecessary specialist referrals dropped 30% while cancer surgeries stayed the same. Key findings: • AI model: fine-tuned MobileNet-V3, trained on 13,569 images • 5 priority levels for malignancy risk • Offline-capable mobile app for rural deployment • Sensitivity: 0.648 → 0.804 with AI assistance • Triage effectiveness: 58.6% → 75.7% Limitations: single Brazilian state, not all skin tones represented, triage tool not diagnosis. Paper: Pacheco et al., "Towards a clinically integrated artificial intelligence tool for triage of skin cancer," npj Digital Medicine, 2026. DOI: 10.1038/s41746-026-02851-8 If this kind of breakdown is useful to you, take a second to like and subscribe — it really does help. #HealthcareAI #SkinCancer #AI #DigitalHealth #MachineLearning #ClinicalValidation #MobileHealth Watch this episode on YouTube: https://youtu.be/fdC5XCkZPjE 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.

Can ChatGPT, Gemini, or Claude actually read an electrocardiogram? A new peer-reviewed study tested six major vision-language models on 70 real clinical ECG images. The results are a wake-up call for anyone assuming AI can replace ECG interpretation. Key findings: • Best generalist model (ChatGPT-5): 62% balanced accuracy • Atrial fibrillation sensitivity: 11% or less — missed nearly every case • ST-segment deviation sensitivity: below 25% • Specialized ECG model PULSE-7B reached 89% for rhythm classification Paper: "Performance of Vision-Enabled Large Language Models in Image-Based Electrocardiogram Interpretation: Exploratory Evaluation" Authors: Soubh, Rasenack, Haarmann, Wiedmann, Zabel, Schmidt, Suliman, Bergau Journal: Journal of Medical Internet Research, June 3, 2026 DOI: 10.2196/86692 #HealthcareAI #ECG #AIinMedicine #Cardiology #VisionLanguageModels #ClinicalAI If this kind of breakdown is useful to you, take a second to like and subscribe — it really does help. Watch this episode on YouTube: https://youtu.be/yWO7ouwb-sc 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.

Raphael T. Malikian, MBBS, BSc (Hons) translates healthcare AI research into practical, clinically grounded questions for builders, clinicians, researchers, and governance teams. GitHub: https://github.com/rtmalikian LinkedIn: http://www.linkedin.com/in/raphael-t-malikian-mbbs-bsc-hons-71075436a --- Healthcare AI Daily translates one recent peer-reviewed healthcare AI article into a short practical briefing — bridging healthcare and technology for a mixed audience of builders, clinicians, patients, and researchers. Today: a formative RCT shows an LLM-powered chatbot (OPBot) improved osteoporosis knowledge scores and reduced nursing education time by 78% compared to face-to-face education — with selective adherence gains and high response reliability. Source article Title: Preliminary Evaluation of a Large Language Model–Powered Chatbot for Osteoporosis Self-Management Education: Formative Randomized Controlled Trial Authors: Jinling Huang, Xiaolian Xin, Chunyan He, Haoyan Xiong, Wenjun Pang, Shaobo Pang, Wei Sun, Xianghua Ding Journal: JMIR Formative Research Published: 2 June 2026 DOI: https://doi.org/10.2196/85475 Article: https://formative.jmir.org/2026/1/e85475 PDF: https://formative.jmir.org/2026/1/e85475/PDF PubMed PMID: 42228937 Key reported results: Formative RCT (n=100, 88 analyzed); chatbot knowledge scores: median 80.0 vs 75.0 (P=.01); nurse education time: 5 min vs 23 min (78% reduction, P<.001); calcium supplement adherence: OR 1.49 (P=.02); OPBot response reliability: 89.4% rated highly reliable, Cohen κ=0.83; no significant effects for sun exposure, exercise, or total adherence. This video is educational commentary, not medical advice. Source screenshots and figures are used for attributed research discussion. Synthetic voice disclosure: narration generated with edge-tts en-US-AndrewNeural and post-processed for clarity. Watch this episode on YouTube: https://youtu.be/HEL0VZp4hlQ 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.

Raphael T. Malikian, MBBS, BSc (Hons) translates healthcare AI research into practical, clinically grounded questions for builders, clinicians, researchers, and governance teams. GitHub: https://github.com/rtmalikian LinkedIn: http://www.linkedin.com/in/raphael-t-malikian-mbbs-bsc-hons-71075436a --- Healthcare AI Daily translates one recent peer-reviewed healthcare AI article into a short practical briefing. Today: a locally-run large language model matches fine-tuned approaches for detecting self-harm in NHS electronic mental health records — without sending patient data anywhere. Privacy-preserving AI for the most sensitive clinical data. We explain the F1 score, what the NHS is, and the balanced limitations of this single-site study. Source article Title: Detection of Self-Harm in Electronic Mental Health Records Using Privacy-Preserving Local Language Models: Methodological Study Authors: Andrey Kormilitzin, Dan W Joyce, Apostolos Tsiachristas, Rohan Borschmann, Navneet Kapur, Galit Geulayov Journal: JMIR Mental Health Published: 2 June 2026 DOI: https://doi.org/10.2196/87586 Article: https://mental.jmir.org/2026/1/e87586 PDF: https://mental.jmir.org/2026/1/e87586/pdf PubMed PMID: 42227874 Key reported results: 1,000 patients, 1,352 annotated notes from Oxford Health NHS Foundation Trust; Gemma3-27b (local, no data leaves institution) vs fine-tuned RoBERTa; self-harm detection F1=0.92 (both models); global weighted F1: Gemma3-27b=0.88 vs RoBERTa=0.85; recent self-harm (51 notes): Gemma3-27b F1=0.79; prompt engineering only, no gradient-based fine-tuning; privacy-preserving: model runs entirely within NHS secure infrastructure. This video is educational commentary, not medical advice. Source screenshots and figures are used for attributed research discussion. ⚠️ Content disclaimer: This episode discusses self-harm detection in clinical records. If you or someone you know is struggling with thoughts of self-harm, please reach out for help immediately. In the United States, call or text 988 (Suicide and Crisis Lifeline). In the United Kingdom, call 116 123 (Samaritans). Things can get better, and you do not have to face this alone. Synthetic voice disclosure: narration generated with edge-tts en-US-AndrewNeural and post-processed for clarity. 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.

Raphael T. Malikian, MBBS, BSc (Hons) translates healthcare AI research into practical, clinically grounded questions for builders, clinicians, researchers, and governance teams. GitHub: https://github.com/rtmalikian LinkedIn: http://www.linkedin.com/in/raphael-t-malikian-mbbs-bsc-hons-71075436a --- Healthcare AI Daily translates one recent peer-reviewed healthcare AI article into a short practical briefing. Today: a cross-sectional survey reveals that radiologists using an FDA-cleared AI tool for intracranial hemorrhage detection report conditional trust, substantial false-positive burden, and limited workflow savings — with neuroradiologists especially critical. Source article Title: Radiologist Perceptions of an AI Tool for Intracranial Hemorrhage Detection in Teleradiology: Cross-Sectional Survey Study Authors: Andrew J Del Gaizo, Jackson K Del Gaizo, Troy A Shahoumian Journal: JMIR Human Factors Published: 2 June 2026 DOI: https://doi.org/10.2196/92145 Article: https://humanfactors.jmir.org/2026/1/e92145 PDF: https://www.jmir.org/2026/1/e92145/PDF PubMed PMID: 42228936 Key reported results: 65 radiologists surveyed; only 18.5% found false-positive alerts acceptable; 50.8% trust AI when it agrees vs 3.1% when it conflicts; 32.3% agreed AI correctly identifies most ICH cases; 33.8% said FP review time outweighs benefits (primary endpoint); only 10.8% reported reduced interpretation time; neuroradiologists more critical (52.2% vs 23.8%); 32.3% worried about legal vulnerability when overriding AI. This video is educational commentary, not medical advice. Source screenshots and figures are used for attributed research discussion. Synthetic voice disclosure: narration generated with edge-tts en-US-AndrewNeural and post-processed for clarity. Watch this episode on YouTube: https://youtu.be/fgWL0JxrCnU 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.

Raphael T. Malikian, MBBS, BSc (Hons) translates healthcare AI research into practical, clinically grounded questions for builders, clinicians, researchers, and governance teams. GitHub: https://github.com/rtmalikian LinkedIn: http://www.linkedin.com/in/raphael-t-malikian-mbbs-bsc-hons-71075436a --- Healthcare AI Daily translates one recent peer-reviewed healthcare AI article into a short practical briefing. Today: why newer reasoning large language models can still generate stereotyped medical vignettes — and why healthcare teams should monitor representational bias before using synthetic cases in education, evaluation, or workflow design. Source article Title: Evaluating the Potential of Reasoning Large Language Models to Perpetuate Racial and Gender Disease Stereotypes in Health Care Authors: Joshua J Docking, Lee X Li, Bradley D Menz, Stephen Bacchi, Ashley M Hopkins, Michael J Sorich Journal: Journal of Medical Internet Research Published: 28 May 2026 DOI: https://doi.org/10.2196/82256 Article: https://www.jmir.org/2026/1/e82256 PDF: https://www.jmir.org/2026/1/e82256/PDF PubMed PMID: 42208042; PMCID: PMC13218561 Key reported results: 36,000 generated clinical vignettes; o3-mini had over 20% racial misrepresentation in 14/18 conditions; DeepSeek-R1 in 16/18; gender misrepresentation crossed the threshold in 10/18 and 12/18 conditions respectively; 16/20 sampled DeepSeek-R1 reasoning traces invoked disease-demographic associations. This video is educational commentary, not medical advice. Source screenshots and figures are used for attributed research discussion. Synthetic voice disclosure: narration generated with edge-tts en-US-AndrewNeural and post-processed for clarity. Watch this episode on YouTube: https://youtu.be/O9vRJFfWJGQ 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.

Raphael T. Malikian, MBBS, BSc (Hons) translates healthcare AI research into practical, clinically grounded questions for builders, clinicians, researchers, and governance teams. GitHub: https://github.com/rtmalikian LinkedIn: http://www.linkedin.com/in/raphael-t-malikian-mbbs-bsc-hons-71075436a --- Healthcare AI Daily translates one healthcare AI paper into a short practical briefing for builders, clinicians, researchers, and governance teams. Today: why medical AI models can look accurate while learning the wrong shortcut — and how dataset bias audits can help teams find risks before deployment. Source article Title: Detecting dataset bias in medical AI using a generalized and modality agnostic auditing approach Authors: Nathan Drenkow, Mitchell Pavlak, Keith Harrigian, Ayah Zirikly, Adarsh Subbaswamy, Mohammad Mehdi Farhangi, Nicholas Petrick, Mathias Unberath Journal: npj Digital Medicine Published: 29 May 2026 DOI: https://doi.org/10.1038/s41746-026-02807-y Article: https://www.nature.com/articles/s41746-026-02807-y Publisher supplementary PDF: https://static-content.springer.com/esm/art%3A10.1038%2Fs41746-026-02807-y/MediaObjects/41746_2026_2807_MOESM1_ESM.pdf Keywords: healthcare AI, medical AI, dataset bias, shortcut learning, AI governance, model validation, clinical artificial intelligence, machine learning bias, G-AUDIT, npj Digital Medicine, FDA, external validation, subgroup analysis, drift monitoring. This video is educational commentary, not medical advice. Source screenshots and supplementary figures are used for attributed research discussion. Like, subscribe, and enable notifications for daily Healthcare AI episodes. Healthcare AI Weekly releases every Friday at 9 AM. Watch this episode on YouTube: https://youtu.be/ejxWoBr2upk 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.

Mental-R1: Can we teach AI to reason like a clinician for mental health assessment? Researchers at Oxford developed CRPO (Cognitive Relative Policy Optimization), a reinforcement learning framework that aligns LLM reasoning with human cognitive processes for mental health assessment. Mental-R1 outperforms GPT-5, DeepSeek-R1, and domain-specific models across 8 mental health datasets covering depression, anxiety, suicide risk, stress, and loneliness. Key findings: - 80.8% accuracy on depression severity classification - 10.4 percentage point average improvement in weighted F1-score - Human-like reasoning pattern: broad exploration followed by confident decisions - Outperforms GPT-5, DeepSeek-R1, GPT-4o, and all domain-specific models Paper: arxiv.org/abs/2606.13176 Authors: Xin Wang, Boyan Gao, Yibo Yang, David A. Clifton #HealthcareAI #MentalHealth #ClinicalAI #MachineLearning #ReinforcementLearning #DigitalHealth #AIResearch #LargeLanguageModels #MentalHealthAI #AIinMedicine #HealthTech MEDICAL DISCLAIMER: This podcast is for educational and informational purposes only. It is not intended to diagnose, treat, cure, or prevent any medical condition, and should not be relied upon as medical advice. The content discussed reflects published research and does not constitute clinical recommendations. If you have any health concerns or medical questions, please consult a licensed healthcare professional immediately. 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. Created by Raphael T. Malikian, MBBS, BSc (Hons) For comments, questions, or concerns: [email protected] YouTube: https://www.youtube.com/@RaphaelMalikian-g4h 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.

New AI approaches to HER2 scoring in breast cancer pathology slides, improving consistency and speed of this critical biomarker assessment. Watch this episode on YouTube: https://youtu.be/ydhDqkzKOgY 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.

Examining how well LLMs calibrate their confidence when interpreting ultrasound images, and what this means for clinical decision support systems. Watch this episode on YouTube: https://youtu.be/NuRwJ8dPiJ4 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.

Exploring how LLMs are being used in professional development across healthcare, with implications for training, continuing education, and clinical competency assessment. Watch this episode on YouTube: https://youtu.be/QVpKbIyXdbk 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.