Faces of Digital Health

AI couldn't cure his mother's stage 4 cancer. It caught three near-fatal errors, found a same-day appointment, and helped her leave on her own terms. When Pratik Desai's mother was diagnosed with stage four duodenal adenocarcinoma — a rare cancer with roughly 3,000 US cases a year — she was nearly discharged without an oncology appointment. Over the next 76 days, Desai used AI at her bedside, from 5am to 10pm, to understand each report, prepare for every appointment, and push a stretched health system to move at the pace her diagnosis demanded. This is a frank account of where AI helped, where it didn't, and the line he refuses to cross. This is a 1:1 interview in The Agentic Patient — a Faces of Digital Health series on how patients and caregivers actually use AI: which tools, which prompts, and which guardrails. GUEST Pratik Desai — New Jersey-based AI practitioner; caregiver and builder of a free, local AI tool for patients HOST Tjaša Zajc — Founder & host, Faces of Digital Health / The Agentic Patient WHAT THE CONVERSATION COVERS - Using AI to interpret a biopsy report and push for a same-day "stat" CT scan - Why AI and the doctors agreed on the care — and clashed on the speed - Finding a same-day oncology appointment through an AI-assisted network search - An error-riddled CT report the AI refused to read — and what it did to trust - Running three Claude "personas" as built-in second and third opinions - A local, open-source AI tool that keeps medical data off the cloud - How to prompt as a patient or caregiver: awareness, knowledge, advocacy — not diagnosis - Where AI failed him: prognosis, and the rule he broke under pressure - Defining quality of life when the outcome is already known CHAPTERS 0:00 How patients use AI — and the guardrails 1:20 Day one: a healthy mother, a diagnosis no one would name 3:34 The first prompt, and pushing for a stat CT scan 7:43 Using AI in the open: agreement on care, friction on speed 9:35 The counterfactual: 76 days with AI at the bedside 12:40 Finding a same-day appointment through a network search 13:40 The CT report the AI refused to read 15:50 When trust erodes: good faith, not competence 18:41 Why switching hospitals wasn't an option 21:54 Defining quality of life: her three goals 28:27 Three Claude personas, and a local private tool 35:12 How to prompt: awareness, knowledge, advocacy — not diagnosis 37:54 Where AI fell short, and the closing asks THE AGENTIC PATIENT SERIES New to the series? Start here → [PASTE PREVIOUS AGENTIC PATIENT EPISODE LINK] All episodes → https://www.facesofdigitalhealth.com/agentic-patient-blog MORE FROM FACES OF DIGITAL HEALTH 🌐 Website: https://www.facesofdigitalhealth.com 📨 Newsletter: https://fodh.substack.com 🎙 Podcast (Apple): https://podcasts.apple.com/gb/podcast/faces-of-digital-health/id1194284040 💼 LinkedIn: https://www.linkedin.com/company/faces-of-digital-health Pratik's tool Regana: https://github.com/RaganaCorp/openhealth-prototype-1 #DigitalHealth #HealthAI #AgenticPatient #PatientAdvocacy #AIinHealthcare #CancerCare #Caregiving #FacesOfDigitalHealth

In this episode of Faces of Digital Health, host Tjasa Zajc sits down with Bart de Witte for a candid conversation on what agent-based AI really means for healthcare. Recorded during a car ride in Ljubljana, the discussion explores why healthcare needs an operating system for AI agents, the risks of agent autonomy, privacy-by-design through on-device AI, and why monolithic EHRs struggle with the next generation of clinical workflows. Bart also shares his vision for open, decentralized AI ecosystems, certified clinical agents, and swarm intelligence and explains why Europe may be uniquely positioned to lead this shift. A practical, forward-looking episode for anyone working at the intersection of healthcare, AI, and digital infrastructure. Youtube video version: https://youtu.be/F_GRfIbqJJM?si=qheSsKvcg6WXUqTU

Ambient documentation is becoming normal in clinics. But the most interesting “voice” capability may not be transcription at all.In the latest episode of Faces of Digital Health, Henry O'Connell (Canary Speech) explains why voice biomarkers stalled for decades: the field analyzed words, not the neurological signal behind speech production.Canary’s approach focuses on the “primary data layer”—how the central nervous system drives respiration, vocal cord vibration, and articulation in real conversational speech. A few details that stood out: ⏱️ ~45 seconds of conversation can be enough for assessment 🎛️ 2,590 voice features analyzed every 10ms (millions of data points) 🎯 Reported accuracy: 98%+ for progressive neurological conditions (e.g., Parkinson’s/Huntington’s/Alzheimer’s), while behavioral health tends to be lower (often in the 80s) 🌍 Validation is repeated per language/culture—no “deploy and hope” model 🧭 Use cases go beyond diagnosis: screening in primary care, clinical trials outcome tracking, and even in-room aggression risk signals to help protect staff One line that captures the idea: it’s about measuring what’s present in the moment—objective signals that complement clinical judgment. Time stamps: 00:00 Introduction to Voice Biomarkers in Digital Health 01:48 Historical Context and Evolution of Voice Analysis 06:52 Innovative Approaches to Voice Data Analysis 08:54 Technical Insights into Voice Analysis 16:07 Accuracy and Efficacy of Voice Biomarkers 28:27 Challenges and Acceptance in Clinical Practice 35:04 Ethical Dilemmas in Genetic Testing 36:32 Understanding Genetic Information and Its Implications 37:58 Objective vs. Subjective Assessments in Mental Health 39:59 Proactive Care and Early Detection of Cognitive Decline 42:43 Technology in Wellness and Employee Mental Health 45:18 Data Privacy and Ethical Considerations in Health Tech 49:06 Remote Monitoring and Clinical Trials 01:00:57 Future of Health Technology and Global Expansion Youtube: https://youtu.be/662VfHhdSFQ?si=t80_PblCf1L6dv4V Website: www.facesofdigitalhealth.com Newsletter: https://fodh.substack.com/

In this episode, Dipak Kalra, President of the European Institute for Innovation through Health Data, joins Faces of Digital Health to break down the real progress (and real gaps) in European health data, from legacy “hybrid” paper/digital workflows to the underused potential of clinical decision support that depends on structured data. We explore what EHDS changes—especially the promise of a standardized, downloadable patient dataset—and what it could unlock for patient-facing apps, analytics, and more active self-management. We also tackle the hard questions: how to protect citizens from misuse and scams, how opt-out choices might create bias in research and AI, why “beating clinicians with a stick” won’t fix data quality, and why delays aren’t just bureaucratic—they can translate into avoidable harm. 02:00 The State of Healthcare Data in Europe 07:59 Challenges in Data Interoperability 12:31 The Role of Patients in Data Management 16:37 AI and Data Privacy Concerns 22:01 Patient Consent and Data Usage 28:00 Optimism for the Future of Health Data 31:03 Optimistic Futures for EAGDS 33:02 Preparing for EHDs: Readiness and Challenges 35:48 Data Quality and Workforce Challenges 37:58 Delays and Future Discussions on EHDs 39:53 The Urgency of Health Data Readiness 42:38 The Evolving Role of Patients in Healthcare 50:19 Building Trust Among Healthcare Stakeholders 57:58 The Future of Healthcare Data Discussions

In this episode of Faces of Digital Health, we sit down with Anne Forsyth, Hospital leader in clinical applications from Women's College Hospital in Canada, to explore how AI — especially generative AI — is reshaping daily clinical practice. Over the past two years, enthusiasm for AI has skyrocketed inside hospitals, with clinicians themselves requesting new tools rather than resisting them. We discuss the cautious but deliberate rollout of AI scribes, the still-emerging trust in decision-support AI, and the safety and change-management considerations that mirror (and sometimes exceed) traditional IT implementations. Anne offers an honest look at the financial challenges of sustaining AI tools in publicly funded health systems and shares practical advice for hospitals navigating funding models, clinical buy-in, and responsible innovation. Show notes: 01:50 – Current AI Implementations 03:21 – Safety and Risk Considerations 04:00 – Comparing AI Rollouts to Traditional IT Tools 05:10 – The Business Equation: Funding AI in Public Healthcare 06:20 – Advice for Hospitals on Sustainable AI Adoption 06:40 – Looking Ahead: The Future of AI in Clinical Applications www.facesofdigitalhealth.com https://fodh.substack.com/