The Digital Health Podcast with Fexingo: Wearables, Diagnostics, and Health Software

On this episode of The Digital Health Podcast, Lucas and Luna explore how artificial intelligence is turning a simple breath sample into a diagnostic tool for lung disease. They focus on a specific case: a startup called BreathDX that has trained a machine learning model on over 50,000 breath volatile organic compound profiles to detect early-stage chronic obstructive pulmonary disease (COPD) and...

Lucas and Luna examine the quiet revolution in radiology reporting: AI that drafts structured medical reports from imaging data, not just detecting anomalies. They unpack how one health system reduced radiologist report turnaround from 24 hours to under 20 minutes using a natural language model trained on 10 million de-identified reports. The hosts discuss accuracy trade-offs — the model catches p...

Hearing aids are no longer just amplifiers. In this episode, Lucas and Luna explore how companies like Sonova and Starkey are embedding AI, fall detection, and even heart rate monitoring into hearing devices. They discuss the 2024 FDA rule that created an over-the-counter hearing aid category, how sensor fusion turns ear-level data into health insights, and why audiologists are both excited and wa...

Episode 101 of The Digital Health Podcast explores the emerging technology of digital twins in healthcare. Lucas and Luna dive into how these virtual models of individual patients are being used to predict disease onset, personalise treatment plans, and reduce clinical trial costs. With real-world examples from a recent study at a major UK hospital where digital twins helped predict sepsis six hou...

In this episode of The Digital Health Podcast, Lucas and Luna explore the emerging field of digital twins in healthcare — personalized AI simulations of your body that can predict disease, test treatments, and guide preventive care. They focus on a specific case: how researchers at the University of Nottingham are using digital twins to model metabolic syndrome risk for individual patients, combin...

In this episode, Lucas and Luna explore the emerging field of vocal biomarker analysis for mental health. They dive into how AI models can detect depression, anxiety, and even PTSD from subtle changes in speech patterns like pitch, tempo, and tone. The discussion centers on real-world studies, including a 2025 trial from the University of Vermont that achieved 85% accuracy in screening for major d...

Lucas and Luna explore the latest generation of consumer wearables that use AI and optical sensors to estimate blood pressure without a cuff. They focus on the technical challenge—how photoplethysmography and pulse transit time work—and the regulatory hurdles from the FDA. Specific numbers include the 1.28 billion people worldwide with hypertension, the 46 percent who are undiagnosed, and the 5 mm...

Lucas and Luna explore the rise of continuous glucose monitors as a consumer wellness subscription, not just a diabetes tool. They examine how companies like Levels, Nutrisense, and Signos are marketing CGMs to biohackers and healthy people, what the data actually shows about non-diabetic glucose spikes, and whether the $200–$400 monthly subscription is backed by solid science. The episode also to...

Lucas and Luna explore how artificial intelligence is transforming health insurance from a reactive claims processor into a proactive health partner. They focus on one specific case: how a major insurer used AI to analyze members' smartwatch data and predict hospitalization risk 30 days in advance, cutting preventable admissions by 18 percent. The episode drills into the numbers, the algorithm's d...

Episode 95 of The Digital Health Podcast explores the cutting-edge link between sleep architecture and mental health diagnostics. Lucas and Luna dive into how researchers at the University of Colorado Boulder have trained AI models to detect major depressive disorder with 80 percent accuracy using only overnight heart rate and movement data from wearables like the Oura Ring and Fitbit. They unpack...

Lucas and Luna dive into the emerging world of AI-driven precision oncology. They focus on a concrete case: how a startup called PathAI is using deep learning to analyze tumor biopsy slides and genomic data in hours instead of weeks, matching patients to targeted therapies with higher accuracy. Lucas explains the specific technology behind digital pathology — convolutional neural networks trained...

Lucas and Luna explore how researchers are using machine learning on voice recordings to detect early signs of postpartum depression. They discuss a 2025 study from the University of Vermont that achieved 87% accuracy by analyzing pitch variability, speech rate, and vocal jitter in just 30-second voice samples. The episode covers how this could enable routine screening in pediatrician offices and...

Lucas and Luna explore how deep learning models are turning smartphone photos into triage tools for melanoma and other skin cancers. They unpack the FDA clearance of a direct-to-consumer app from the company Skin Analytics in early 2026, walk through the study data behind its sensitivity rate of 96.3 percent for detecting malignant lesions, and debate where these tools fit alongside dermatologists...

In this episode, Lucas and Luna explore how AI models are using smartwatch sensor data—heart rate, skin temperature, sleep patterns—to detect early signs of COVID-19 days before symptoms appear. They dive into a specific study from Mount Sinai Health System that used the Apple Watch to catch infections with 90% sensitivity, and discuss the broader implications for wearable-based pandemic surveilla...

In this episode of The Digital Health Podcast, Lucas and Luna explore how artificial intelligence is turning smartphone cameras into clinical-grade ultrasound devices. They focus on Butterfly Network's latest handheld probe and the role of deep learning in image reconstruction, making diagnostic imaging accessible in rural clinics and ambulances. The hosts discuss the regulatory hurdles, the shift...

Lucas and Luna explore how synthetic control arms—using AI to match historical patient data instead of recruiting a separate placebo group—are accelerating clinical trials and cutting costs. They break down the real-world case of a mid-stage oncology trial that used a synthetic arm built from over 10,000 historical records, reducing enrollment time by 40% and saving an estimated $12 million. The h...

Finding the right clinical trial is notoriously slow — fewer than 5 percent of eligible cancer patients ever enroll, and matching often takes weeks of manual chart review. But a new wave of AI-powered matching platforms is changing that. In this episode, Lucas and Luna look at how natural language processing and graph-based reasoning are scanning electronic health records and trial protocols to fi...

Lucas and Luna explore how companies like DayTwo and Viome are using machine learning to analyze the gut microbiome and generate personalized food recommendations. They break down the science of metagenomic sequencing, the challenge of translating microbial data into actionable diet plans, and why the FDA is paying close attention. With specific numbers on trial outcomes and cost barriers, this ep...

In this episode of Digital Health Podcast, Lucas and Luna explore how machine learning models are mining electronic health records to detect sepsis hours before visible symptoms appear. They focus on the case of Epic's Sepsis Model, which analyzes over 200 variables in real time to flag patients at risk. Lucas explains why sepsis is the deadliest condition in hospitals, claiming 270,000 lives annu...

In this episode of The Digital Health Podcast, Lucas and Luna explore how artificial intelligence is being used to predict which patients will be readmitted to the hospital within 30 days of discharge — before they even leave the building. They focus on a specific case: a machine learning model deployed at Mount Sinai Health System that analyzes electronic health record data, social determinants o...

Lucas and Luna unpack the quiet revolution in health insurance underwriting: how AI models are analyzing everything from smartwatch step counts to grocery purchase data to calculate personalized risk scores. They walk through a real case study where one major insurer cut claims costs by 18 percent in a pilot program by using wearable data to incentivize preventive care. But they also dig into the...

Lucas and Luna dive into the emerging field of AI-powered retinal imaging, where a simple eye scan can detect signs of Alzheimer's, diabetes, and cardiovascular risk years before symptoms appear. They explore how Google Health's algorithm for diabetic retinopathy achieved FDA clearance in 2020, and how newer models are predicting age-related macular degeneration and systemic diseases from retinal...

In this episode, Lucas and Luna dive into how consumer smartwatches are becoming frontline tools for detecting atrial fibrillation, a heart condition that affects over 33 million people globally and is a leading cause of stroke. They explore a landmark 2025 study from the Stanford Center for Digital Health that analyzed data from over 400,000 Apple Watch users, showing that the device's photopleth...

In this episode, Lucas and Luna explore how smart rings — tiny wearable computers you wear on your finger — are quietly becoming the next frontier in continuous health monitoring. They break down the key players: Oura, the Finnish company that started the category, and newcomers like Samsung and Amazfit. Lucas shares how Oura's latest generation tracks everything from heart rate variability to ski...

Lucas and Luna explore how pharmaceutical companies and startups are using artificial intelligence to find new therapeutic uses for drugs already approved for other conditions. They focus on a specific case: a 2025 study where an AI model screened 4,000 existing compounds and identified a common blood pressure medication as a potential treatment for a rare genetic liver disease — cutting the typic...