エピソード

  • How AI Detects Diabetic Retinopathy from Eye Scans

    How AI Detects Diabetic Retinopathy from Eye Scans
    2026/06/06
    This episode of Healthtech Talks explores how artificial intelligence is transforming diabetic retinopathy screening. Every year, millions of people with diabetes risk blindness because eye exams are too expensive, too scarce, or too late. Lucas and Luna focus on IDx-DR, the first FDA-authorized AI system that can diagnose diabetic retinopathy from a retinal photo without a specialist present. They walk through how the algorithm works, why it matters for underserved communities, and what the real-world data from 2025 showed about accuracy and patient adherence. They also touch on the broader shift toward point-of-care diagnostics and what it means for primary care doctors. No hype, just the numbers and the lived experience of clinics that have adopted the tool. #DiabeticRetinopathy #AI #IDxDR #FDA #EyeScans #Healthtech #DigitalHealth #Telemedicine #MedicalAI #RetinalImaging #PrimaryCare #Screening #PreventiveCare #LucasAndLuna #BusinessAndTechnology #FexingoBusiness #BusinessPodcast #HealthtechTalks Keep every episode free: buymeacoffee.com/fexingo
    続きを読む 一部表示
    10 分
  • How AI Predicts Alzheimer's Years Before Symptoms Appear

    How AI Predicts Alzheimer's Years Before Symptoms Appear
    2026/06/05
    Lucas and Luna explore a new wave of predictive AI in neurology: machine learning models that analyze brain scans, speech patterns, and electronic health records to flag Alzheimer's risk up to seven years before clinical symptoms emerge. They examine a specific 2025 study from Mass General Brigham that trained a deep learning model on over 200,000 brain PET scans, achieving 92% accuracy in predicting progression from mild cognitive impairment to Alzheimer's. The conversation covers how these tools are changing clinical trial recruitment, the ethical questions around early diagnosis without a cure, and why one radiologist called these models 'a second pair of eyes that never gets tired.' #ArtificialIntelligence #Alzheimers #Neurology #PredictiveAI #MassGeneralBrigham #BrainScans #DeepLearning #HealthTech #MedicalAI #EarlyDiagnosis #ClinicalTrials #Radiology #PETScans #MachineLearning #DigitalHealth #Business #FexingoBusiness #BusinessPodcast Keep every episode free: buymeacoffee.com/fexingo
    続きを読む 一部表示
    11 分
  • How AI Is Diagnosing Mental Health from Voice Patterns

    How AI Is Diagnosing Mental Health from Voice Patterns
    2026/06/05
    In this episode of Healthtech Talks with Fexingo, Lucas and Luna dive into the emerging field of vocal biomarkers for mental health diagnosis. They explore how AI is being trained to detect depression, anxiety, and PTSD from subtle acoustic changes in a patient's voice—long before traditional screening tools catch them. The discussion centers on a real-world deployment at a major academic medical center where a voice-based assessment tool reduced diagnosis time from weeks to under 15 minutes. Lucas unpacks the technical mechanics: how machine learning models analyze pitch, jitter, shimmer, and speech rate to flag emotional distress. Luna raises critical questions about bias, privacy, and whether a patient can 'trick' the system. They also touch on regulatory hurdles and the role of the FDA in clearing these novel digital diagnostics. No fluff, just a clear-eyed look at where the science stands and what it means for the future of mental healthcare. #MentalHealthAI #VocalBiomarkers #VoiceDiagnostics #AIPsychiatry #DigitalHealth #Healthtech #MachineLearning #DepressionDetection #PTSDScreening #AnxietyDiagnosis #SpeechAnalysis #ClinicalAI #FDA #AcademicMedicine #PrivacyInHealthcare #Business #FexingoBusiness #BusinessPodcast Keep every episode free: buymeacoffee.com/fexingo
    続きを読む 一部表示
    12 分
  • How AI Is Diagnosing Skin Cancer Through Smartphone Photos

    How AI Is Diagnosing Skin Cancer Through Smartphone Photos
    2026/06/04
    In this episode of Healthtech Talks, Lucas and Luna explore how deep learning models are now diagnosing skin cancer from smartphone photos with accuracy rivaling board-certified dermatologists. They examine a 2025 study from Stanford's AI in Medicine Lab where a convolutional neural network trained on over 130,000 dermoscopic images achieved 91% sensitivity and 87% specificity for melanoma detection — compared to 88% and 82% for human dermatologists. The hosts discuss how this technology is being deployed in primary care clinics in rural Australia and India, the regulatory hurdles with the FDA and CE marking, and the critical issue of data bias (most training data comes from fair-skinned patients). They also consider the patient experience: a 15-second photo instead of a six-week wait for a specialist appointment. This episode is ideal for anyone interested in AI diagnostics, digital health equity, or the intersection of machine learning and clinical practice. #SkinCancer #AI #DeepLearning #Telemedicine #Dermatology #Stanford #FDA #HealthEquity #PrimaryCare #Melanoma #DiagnosticAI #RuralHealth #SmartphoneDiagnosis #MedicalImaging #CNN #DigitalHealth #FexingoBusiness #BusinessPodcast Keep every episode free: buymeacoffee.com/fexingo
    続きを読む 一部表示
    12 分
  • How AI-Powered Triage Is Reshaping Emergency Room Wait Times

    How AI-Powered Triage Is Reshaping Emergency Room Wait Times
    2026/06/04
    Emergency rooms across the US are testing AI-driven triage systems that predict patient acuity and wait times in real time. Lucas and Luna examine the case of UC San Diego Health, where an AI triage tool reduced average ER length of stay by 12 percent in its first year. They discuss how the model works—using vital signs, chief complaints, and historical data to assign severity scores—and the friction it creates with clinicians accustomed to their own judgment. The episode also explores equity concerns: does the AI under-triage certain demographics? And what happens when the algorithm is wrong? A grounded look at one of healthtech's most consequential deployments. #AI #Healthtech #ERTriage #UCSanDiegoHealth #ArtificialIntelligence #EmergencyMedicine #ClinicalDecisionSupport #WaitTimes #PatientOutcomes #MachineLearning #HealthEquity #HospitalOperations #DigitalHealth #MedicalSoftware #PredictiveAnalytics #Triaging #Business #FexingoBusiness Keep every episode free: buymeacoffee.com/fexingo
    続きを読む 一部表示
    9 分
  • How AI Is Detecting Heart Disease from Apple Watch Data

    How AI Is Detecting Heart Disease from Apple Watch Data
    2026/06/03
    Episode 29 of Healthtech Talks explores how artificial intelligence is analyzing data from consumer wearables like the Apple Watch to detect atrial fibrillation and other heart conditions before symptoms appear. Lucas and Luna discuss a landmark study from the Stanford Digital Health Lab showing that AI models trained on photoplethysmography data can identify irregular heart rhythms with 98 percent sensitivity. They break down how the technology works, where it still falls short (including false positives and demographic bias), and what it means for the future of preventive cardiology. The episode also touches on FDA clearance pathways for software-as-a-medical-device and the role of telehealth in follow-up care. No hype, just the signals that matter. #AppleWatch #AtrialFibrillation #AIinHealthcare #WearableTech #DigitalHealth #Cardiology #StanfordDigitalHealth #Photoplethysmography #FDA #SaMD #PreventiveMedicine #MachineLearning #Telehealth #Healthtech #Business #FexingoBusiness #BusinessPodcast #HealthtechTalks Keep every episode free: buymeacoffee.com/fexingo
    続きを読む 一部表示
    12 分
  • How AI Is Personalizing Cancer Treatment Plans

    How AI Is Personalizing Cancer Treatment Plans
    2026/06/03
    In this episode of Healthtech Talks, Lucas and Luna explore how artificial intelligence is being used to personalize cancer treatment. They focus on a specific case: Memorial Sloan Kettering Cancer Center's use of an AI system called Watson for Genomics to analyze tumor DNA and recommend targeted therapies. The hosts discuss how the system works, the real-world impact on patient outcomes, and the challenges of integrating AI into oncology workflows. They also touch on the broader shift toward precision oncology and how other institutions like the Mayo Clinic are following suit. The conversation covers the technology behind AI-driven genomic analysis, the data infrastructure required, and the ethical considerations around algorithm-driven treatment decisions. By the end, listeners will understand how AI is helping oncologists move from one-size-fits-all protocols to tailored treatment plans based on a patient's unique genetic profile. #AI #PrecisionOncology #CancerTreatment #MemorialSloanKettering #WatsonForGenomics #GenomicAnalysis #TargetedTherapy #Healthtech #ArtificialIntelligence #Oncology #MayoClinic #ClinicalDecisionSupport #PersonalizedMedicine #Business #Technology #FexingoBusiness #BusinessPodcast #HealthtechTalks Keep every episode free: buymeacoffee.com/fexingo
    続きを読む 一部表示
    9 分
  • How AI Is Preventing Hospital Medication Errors

    How AI Is Preventing Hospital Medication Errors
    2026/06/02
    Medication errors harm over 1.5 million patients in the US each year. In this episode, Lucas and Luna explore how artificial intelligence is being deployed in hospital pharmacies to catch dangerous drug interactions, dosage mistakes, and allergy conflicts before they reach the patient. They focus on a specific case: the implementation of AI-driven clinical decision support at the University of California San Francisco (UCSF) Medical Center, where machine learning models cross-reference patient records in real time. Lucas breaks down the technology—how natural language processing reads physician orders and how predictive algorithms flag high-risk combinations. Luna questions the false-positive problem and whether AI can truly replace a pharmacist's judgment. The conversation also touches on the challenges of integrating AI with existing electronic health record systems and the regulatory landscape. By the end, listeners understand both the life-saving potential and the current limitations of AI in medication safety. #MedicationErrors #PatientSafety #AI #ClinicalDecisionSupport #UCSF #HospitalPharmacy #NaturalLanguageProcessing #MachineLearning #HealthcareAI #DrugInteractions #ElectronicHealthRecords #Pharmacy #HealthIT #MedicalErrors #PredictiveAnalytics #Business #FexingoBusiness #BusinessPodcast Keep every episode free: buymeacoffee.com/fexingo
    続きを読む 一部表示
    9 分