Healthtech Talks with Fexingo: Digital Health, Telemedicine, and Medical Software

For episode 100 of Healthtech Talks, Lucas and Luna explore how machine learning models are detecting early biomarkers of Alzheimer's disease from routine eye scans and speech patterns — years before cognitive symptoms appear. They dive into a 2025 study from the University of Edinburgh that used retinal imaging to predict disease onset with 85 percent accuracy, and discuss how the technology could transform screening in primary care. The hosts also touch on the ethical questions around early diagnosis without effective treatments, and what this means for drug trials and patient planning. A focused look at the intersection of ophthalmology, AI, and neurodegeneration.

#Alzheimers #AI #Healthtech #DigitalHealth #MedicalAI #RetinalImaging #EarlyDiagnosis #MachineLearning #Ophthalmology #Neurodegeneration #UniversityOfEdinburgh #PredictiveAnalytics #HealthcareInnovation #Biomarkers #BusinessAndTechnology #FexingoBusiness #BusinessPodcast #HealthtechTalks

Keep every episode free: buymeacoffee.com/fexingo

In this episode, Lucas and Luna explore how machine learning models are predicting surgical site infections before the patient even leaves the operating room. They dive into a specific model developed by a team at the University of Michigan Health System, which analyzes pre-op labs, patient history, and intra-operative data to assign a real-time infection risk score. The hosts discuss the 2024 study showing a 40 percent reduction in SSI rates when surgeons used the model to adjust antibiotic timing and wound care protocols. They also touch on the challenges of integrating predictive tools into surgical workflows and the potential for these models to become standard in pre-surgical briefings. A concrete look at how AI is moving from prediction to prevention in one of the most common hospital-acquired conditions.

#SurgicalSiteInfection #AIinSurgery #PredictiveAnalytics #UniversityOfMichigan #MachineLearning #PatientSafety #InfectionPrevention #HealthcareAI #ClinicalDecisionSupport #PreOpRisk #SurgicalOutcomes #HealthTech #BusinessAndTechnology #FexingoBusiness #BusinessPodcast #DigitalHealth #MedTech #AntibioticStewardship

Keep every episode free: buymeacoffee.com/fexingo

In this episode of Healthtech Talks, Lucas and Luna explore how AI is turning smartphone photos into skin cancer screening tools. They focus on a specific 2025 study from Stanford where a deep learning model matched dermatologists' accuracy in detecting melanoma from patient-taken selfies. The discussion covers the model's 96% sensitivity rate, the challenge of differentiating benign moles from malignancies, the FDA's cautious stance on direct-to-consumer skin AI apps, and what this means for access in rural areas. Lucas explains the technical difference between dermoscopic and clinical image models, while Luna questions whether convenience could lead to overdiagnosis or missed cases. They also touch on how companies like SkinVision and Google Health are commercializing this technology. The episode closes with a reflection on whether AI-assisted diagnosis will change the role of the dermatologist or simply expand the screening funnel.

#SkinCancerAI #MelanomaDetection #StanfordStudy #DeepLearning #Healthtech #Telemedicine #DermatologyAI #MedicalImaging #AIInHealthcare #SmartphoneDiagnosis #CancerScreening #DigitalHealth #BusinessAndTechnology #LucasAndLuna #FexingoBusiness #BusinessPodcast #HealthtechTalks #SkinVision

Keep every episode free: buymeacoffee.com/fexingo

Sepsis kills roughly 350,000 adults in US hospitals each year — often because clinicians catch it too late. In this episode, Lucas and Luna drill into a single case: how the University of Florida Health system deployed a real-time AI sepsis alert that flags patients 4 to 6 hours before they meet standard diagnostic criteria. They discuss the model's design — it analyzes 35 vitals and lab values every 5 minutes — and the hard part: getting doctors and nurses to trust and act on the predictions. Lucas shares a specific stat from a 2024 study: the algorithm reduced sepsis mortality by nearly 19 percent in the intervention group. Luna pushes back on alarm fatigue and false-positive rates. A focused episode on one of the most promising — and trickiest — applications of machine learning in critical care.

#SepsisPrediction #ClinicalAI #PatientSafety #RealTimeMonitoring #CriticalCare #UFHealth #MachineLearning #HealthcareAI #MedicalSoftware #DigitalHealth #HospitalTech #PredictiveAnalytics #Business #Technology #HealthtechTalks #FexingoBusiness #BusinessPodcast #PedroFonseca

Keep every episode free: buymeacoffee.com/fexingo

Episode 96 of Healthtech Talks dives into the fast-emerging field of AI-driven post-surgical monitoring. Lucas and Luna examine how hospitals are deploying machine learning models that analyze continuous vital-sign streams—heart rate, respiratory rate, oxygen saturation—from bedside monitors to detect complications like sepsis, internal bleeding, and anastomotic leaks hours before a patient feels anything. The episode centers on a specific deployment at the University of California San Francisco, where an algorithm trained on 50,000 surgical cases reduced unplanned ICU transfers by 29 percent. The hosts walk through how the model works, why current nurse-to-patient ratios make continuous human monitoring impossible, and the tricky implementation challenge: false alarms that exhaust clinicians. They also discuss the economic calculus—a single preventable ICU readmission can cost over $50,000—and whether smaller community hospitals can afford the compute infrastructure. The conversation closes with a look at where this technology is heading: predictive models that recommend specific interventions, not just flag risk.

#AISurgery #PostSurgeryMonitoring #UCSF #SepsisPrediction #CriticalCare #MachineLearning #Healthtech #DigitalHealth #Telemedicine #HospitalOperations #PatientSafety #ClinicalAI #PredictiveAnalytics #ICUTransfer #AnastomoticLeak #VitalSigns #FexingoBusiness #BusinessPodcast

Keep every episode free: buymeacoffee.com/fexingo

Emergency rooms are chaotic by design, but AI is starting to bring order. In this episode, Lucas and Luna explore how hospitals are using machine learning to predict how long you'll wait before seeing a doctor — and why that matters for both patient satisfaction and clinical outcomes. They break down a real deployment at a major health system: how the model ingests real-time data from triage notes, bed availability, staffing schedules, and historical patterns to generate wait-time estimates that are actually useful. They also discuss the ethical tension between transparency and anxiety — is it better to tell someone they'll wait four hours if it's accurate, or does that just drive patients to leave? Featuring specific examples from hospitals in Ohio and California, including the surprising finding that AI-driven wait-time predictions reduced left-without-being-seen rates by 17 percent in a six-month pilot. This is Episode 95 of Healthtech Talks with Fexingo.

#AI #MachineLearning #EmergencyMedicine #PatientExperience #WaitTimes #PredictiveAnalytics #Healthtech #HospitalOperations #Triage #ClinicalDecisionSupport #PatientSatisfaction #RealTimeData #OperationalEfficiency #LeftWithoutBeingSeen #OhioHealth #KaiserPermanente #Business #FexingoBusiness

Keep every episode free: buymeacoffee.com/fexingo

Episode 94 of Healthtech Talks with Fexingo explores how artificial intelligence is moving beyond diagnosis to actually design personalized cancer treatment plans. Lucas and Luna examine a real-world case at Memorial Sloan Kettering, where an AI system called CURATE.AI helped a patient with advanced prostate cancer find the optimal drug combination after standard therapies failed. The algorithm analyzed the patient's own tumor data and drug response in real time, adjusting doses weekly. The result: the patient's PSA levels dropped by 80 percent within two months. Lucas walks through the technical architecture behind CURATE.AI, the ethical questions around AI-driven treatment decisions, and how this approach is scaling to other cancer types. Luna challenges the readiness of regulatory frameworks and brings data on similar trials at MD Anderson and the Mayo Clinic. They also discuss the role of patient data privacy and the need for diverse training datasets. This episode is for clinicians, healthtech builders, and anyone curious about the next frontier in precision oncology.

#CURATEAI #PersonalizedTreatment #CancerAI #PrecisionOncology #MemorialSloanKettering #MDAnderson #MayoClinic #ProstateCancer #DrugCombination #AIDosing #HealthtechAI #ClinicalDecisionSupport #FDAApproval #PatientDataPrivacy #BusinessAndTechnology #FexingoHealthtech #FexingoBusiness #BusinessPodcast

Keep every episode free: buymeacoffee.com/fexingo

Episode 93 of Healthtech Talks explores how hospitals are using generative AI to create personalized, plain-language discharge instructions for patients. Lucas and Luna examine a pilot at Johns Hopkins where an AI model tailors aftercare summaries to a patient's reading level, language, and condition. The system reduced 30-day readmissions by 18% in a six-month trial. They discuss the challenges of medical accuracy, liability, and patient comprehension, plus how this technology could scale beyond discharge to pre-operative instructions and chronic disease management. The hosts also make their case for keeping the show ad-free and invite listener support at buy me a coffee dot com slash fexingo.

#AI #PatientEducation #DischargeInstructions #HealthTech #JohnsHopkins #GenerativeAI #ReadmissionReduction #HealthEquity #NaturalLanguageProcessing #ClinicalAI #HospitalTech #PatientSafety #DigitalHealth #Telemedicine #MedicalSoftware #Business #FexingoBusiness #BusinessPodcast

Keep every episode free: buymeacoffee.com/fexingo

In this episode, Lucas and Luna explore how artificial intelligence is being used to personalize mental health therapy for millions. They focus on the case of Woebot, a chatbot therapist that uses cognitive behavioral therapy techniques and natural language processing to adapt to each user's emotional state in real time. The hosts discuss how Woebot's algorithms analyze user responses to tailor exercises and interventions, the evidence base behind its clinical trials—including a 2021 study showing significant reductions in depression symptoms—and the challenges of scaling AI therapy while maintaining privacy and efficacy. They also touch on the broader landscape of AI-powered mental health platforms, including Talkspace and Ginger, and the regulatory hurdles these services face. The episode raises the question: can an algorithm truly understand human emotion, or is AI therapy best used as a complement to human therapists? Lucas brings data on user engagement and dropout rates, while Luna challenges whether personalization can replace the therapeutic alliance.

#AI #MentalHealth #Therapy #Personalization #Woebot #CognitiveBehavioralTherapy #NaturalLanguageProcessing #Talkspace #Ginger #Healthtech #DigitalTherapy #ClinicalTrials #Depression #Anxiety #Business #Technology #FexingoBusiness #BusinessPodcast

Keep every episode free: buymeacoffee.com/fexingo

Lucas and Luna explore how artificial intelligence is being used to detect signs of anxiety and depression from voice patterns alone. They dive into a specific study from the University of Maryland where an AI model achieved 85% accuracy in identifying depression from just 30 seconds of speech. The episode covers the underlying technology—acoustic features like pitch, rhythm, and energy—as well as the practical applications, from therapist support tools to employee wellness programs. They also discuss the ethical concerns around privacy and consent, and how this technology might change mental health screening in the coming years.

#AI #MentalHealth #VoiceAnalysis #DepressionDetection #Anxiety #DigitalHealth #Healthtech #SpeechAnalysis #UniversityOfMaryland #ClinicalAI #WellnessTech #BehavioralHealth #MachineLearning #Telemedicine #Business #Technology #FexingoBusiness #BusinessPodcast

Keep every episode free: buymeacoffee.com/fexingo