Faces of Digital Health

When prevention works, nothing happens. That's why it gets cut. Joanna Lane on financing prevention as long-term investment.

Health systems in Europe spend a small fraction of their budgets on prevention, and that fraction is renewed annually and cut first when treatment costs rise. Joanna Lane has spent decades working on health and regional development across Europe, and argues that prevention should be reclassified from expenditure to capital investment, financed the way we finance energy grids, transport and pensions. In this conversation she explains what the Horizon Europe project Invest4Health learned from eight regional testbeds, why Public Health Wales disinvested around £10m to reinvest it, how health systems' own procurement and hiring can improve the determinants of health, and what finance ministers need to hear before they will treat prevention as an investment. We also discuss wearables and the longevity trend, where personal responsibility ends and public policy begins, and a three-layer model for funding prevention at scale.

GUEST

Joanna Lane — Managing Director, Stichting Health ClusterNET; Chair, FH Europe Foundation; Invest4Health (Horizon Europe)

Host: Tjaša Zajc

WHAT THE CONVERSATION COVERS

- Why the benefits of prevention stay invisible in health budgets

- Why local prevention and health promotion budgets are the first to be cut

- Invest4Health: eight regional testbeds for financing health promotion and disease prevention

- Smart capacitating investment and the EU high-level task force on social infrastructure

- Why a single financing template didn't work across regions

- Financing toolkits and competence building for regional health authorities

- Reclassifying prevention as capital investment and soft infrastructure

- Public Health Wales: disinvesting from health promotion to reinvest around £10m

- Prevention spending during and after COVID, and the pull back to acute care

- Defence, energy and climate competing for health budgets

- The ripple effect of health spending: local procurement, inclusive employment and NEETs

- How to make the investment case for prevention to a finance minister

- Local Multiplier 3 (LM3) and the return on local procurement

- 20-minute neighbourhoods and multi-use community hubs for older people

- Remote monitoring for independent living: the SMILE project

- Wearables, longevity and the retention problem in consumer digital health

- Personal responsibility vs public policy: school food, screening and familial hypercholesterolaemia

- Three layers of prevention finance: protected budgets, social outcome funds and patient capital

- Metrics for treating prevention as investment in national statistics

CHAPTERS

2:00 The Finance Minister Anecdote

MENTIONED

Invest4Health — Horizon Europe project on financing health promotion and disease prevention

High-Level Task Force on Investing in Social Infrastructure in Europe (report, January 2018)

Syreon Research Institute, Budapest • Public Health Wales • Local Multiplier 3 (LM3)

SMILE project (Innlandet Hospital Trust, Norway) • FH Europe Foundation • EU Safe Hearts Plan

Earlier FoDH episodes: Herko Coomans on interoperability as an infrastructure funding problem [LINK] • Henrique Martins on twenty years of EU health data projects [LINK]

FACES OF DIGITAL HEALTH

Website: https://www.facesofdigitalhealth.com

Newsletter: https://fodh.substack.com

LinkedIn: https://www.linkedin.com/company/faces-of-digital-health

Spotify: https://open.spotify.com/show/4cElKJHrauyP6QJQaCkvdY

Apple Podcasts: https://podcasts.apple.com/gb/podcast/faces-of-digital-health/id1194284040

#prevention #publichealth #healthfinancing #healthpromotion #diseaseprevention #Invest4Health #healthpolicy #healtheconomics #socialinfrastructure #healthinvestment #digitalhealth #FacesOfDigitalHealth

Nearly €200m, twenty years of EU health data projects. Henrique Martins on what they built and what EHDS needs next.

One ultrasound tech, four scanning rooms. NVIDIA's David Niewolny on what "augmented autonomy" asks of clinicians.

NVIDIA is best known for GPUs, but its healthcare strategy is a full stack: compute to train models, compute to run them at the point of care, and simulation to generate the data that medical robotics lacks. In this interview, David Niewolny explains how that stack underpins autonomous X-ray and ultrasound with GE HealthCare, surgical robotics simulation with Johnson & Johnson MedTech, and the Open-H surgical robotics dataset. He argues that open models are a regulatory necessity in healthcare, that software-defined medical devices will shrink innovation cycles from years to months, and that the path to autonomy in surgery will follow the one taken by autonomous vehicles, one level at a time. We also press on the parts that remain unsettled: whether AI efficiency turns into clinician burnout, whether synthetic data reflects real patient populations, and what could still stall adoption.

GUEST

David Niewolny — Senior Director and Global Head of Business Development, Healthcare & Medical, NVIDIA

Host: Tjaša Zajc

WHAT THE CONVERSATION COVERS

- NVIDIA's healthcare strategy: training, simulation and edge deployment for medical AI

- Why open models and open datasets matter in regulated healthcare

- Autonomous X-ray and ultrasound with GE HealthCare: one technician, multiple rooms

- "Augmented autonomy": keeping a clinician in the loop

- AI efficiency, cognitive load and the risk of a new wave of clinician burnout

- Ambient clinical documentation as the clearest efficiency case

- Surgical robotics and the autonomous-vehicle model of stepwise autonomy

- Why robotics costs are falling, and what it means for hospital ROI

- Synthetic data for healthcare robotics: Cosmos-H, Isaac for Healthcare and the Open-H dataset

- Can simulated data reflect a local patient population?

- Model drift, verification and validation, and governance of AI agents in healthcare

- Is healthcare worried about superintelligence?

- Software-defined medical devices and the FDA's predetermined change control plan (PCCP)

- How regulators are adapting to AI-enabled devices

- Change management and ROI as the real barriers to adoption

CHAPTERS

02:00 NVIDIA in healthcare: more than GPUs

03:03 The full stack: training, simulation and edge deployment

08:46 Why open models matter in regulated healthcare

10:51 Autonomous X-ray and ultrasound: one technician, four rooms

14:49 AI augmentation and clinician burnout

18:16 What NVIDIA looks for in a partner, and the surgical robotics bet

23:38 Cheaper robots, more competition, clearer ROI

27:37 Synthetic data, Cosmos-H and the Open-H surgical dataset

31:43 Does simulated data reflect the real world?

34:37 Model drift and governing AI agents in healthcare

38:03 Superintelligence, change management and the autonomous-vehicle analogy

43:10 Software-defined medical devices, the FDA's PCCP and regulators

49:54 A golden age for medtech? What could still slow it down

MENTIONED

NVIDIA Isaac for Healthcare • Cosmos-H • Open-H-Embodiment dataset • NVIDIA Nemotron • BioNeMo Agent Toolkit

GE HealthCare • Johnson & Johnson MedTech (MONARCH platform) • Abridge • Aidoc • OpenEvidence • Sword Health

FDA Predetermined Change Control Plan (PCCP)

FACES OF DIGITAL HEALTH

Website: https://www.facesofdigitalhealth.com

Newsletter: https://fodh.substack.com

LinkedIn: https://www.linkedin.com/company/faces-of-digital-health

Spotify: https://open.spotify.com/show/4cElKJHrauyP6QJQaCkvdY

Apple Podcasts: https://podcasts.apple.com/gb/podcast/faces-of-digital-health/id1194284040

#NVIDIA #healthcareAI #medtech #surgicalrobotics #physicalAI #digitalhealth #medicalimaging #syntheticdata #openmodels #FDA #medicaldevices #healthcarerobotics

Most interoperability debates focus on exchange formats. In this discussion with Dr Nils Hellrung, medical informatician and CEO Strategy & Operations at vitagroup, Nils focuses on the need for shaping hospital IT with governance and financing.

This episode is supported by vitagroup.

She has 408 of her daughter's medical documents. She stopped letting AI read them — here's why.

Karlien Hollanders is a pharmacist who became her daughter's medical record. When her firstborn went into complete renal failure at two months old and was diagnosed with primary hyperoxaluria type 1, she spent three years moving between Belgian specialist hospitals that could not exchange data with each other. Eleven years later, 408 documents sit on a national platform that can be filtered by doctor, hospital and date — and nothing else. This episode of The Agentic Patient is about what happens when you actually try to put AI to work on a real, fragmented, non-anonymised medical history: what it organised, what it got wrong, and the reason she has stopped.

GUEST

Karlien Hollanders — Patient expert and caregiver; consultant to the Belgian federal health service; pharmacist by training

Host: Tjaša Zajc

THE AGENTIC PATIENT

A Faces of Digital Health series on how patients and caregivers actually use AI — which tools, which prompts, which guardrails — and what that does to the clinical relationship.

Series hub: https://www.facesofdigitalhealth.com/agentic-patient

WHAT THE CONVERSATION COVERS

- Being your child's "walking medical record" across five specialist hospitals

- Why case management exists for cancer and diabetes but not for rare disease

- 408 PDFs, three filters: the missing-metadata problem in national health data platforms

- Federated health data explained — and why you still download your records one click at a time

- What AI document tools did well: organising records by organ and producing specialty-specific summaries

- The timeline AI could not build, and why copy-pasted clinical letters break text-layer summarisation

- Why internists and orthopaedic surgeons need the same records and completely different queries

- The 15-minute consultation and the unprepared specialist

- Local bulk de-identification: the tool that does not exist, and the vendors who could not supply it

- A child's medical data on third-party AI platforms, twenty years forward

- Patient summaries and cross-border care: what an emergency room in another country actually needs

- EHDS versus AI extraction — structuring data at the point of documentation instead of after it

- SNOMED CT and structured entry that clinicians do not know they are doing

- ZAS Antwerp's AI-generated patient-friendly discharge letters, already in production

- Why a shared medication scheme can be five years out of date in a country with electronic prescribing

- Personal health vaults, EU wallets and itsme: identity is not the same as data portability

- Practical advice: how to assemble your own records before letting AI near them

CHAPTERS

03:00 A caregiver's view of patients using AI

04:28 Renal failure at two months: primary hyperoxaluria type 1

10:43 Ten years on: what has changed in Belgian health data sharing

13:04 408 documents, three filters: the metadata problem

15:03 Downloading a federated health record one PDF at a time

17:20 What AI document tools organised — and the timeline they couldn't build

22:27 Same records, different questions: internists vs orthopaedic surgeons

26:38 Not a technology problem: incentives and the missing business case

30:18 AI as a better Google — and where that gets dangerous

32:30 The anonymisation gap: why she stopped feeding AI real data

36:10 Patient summaries, EHDS and structuring data at the source

41:46 ZAS Antwerp's AI-written patient letters, already in production

47:49 Preparing for a specialist visit when your records are scattered

MENTIONED

European Health Data Space (EHDS) — Regulation (EU) 2025/327

ZAS (Ziekenhuis aan de Stroom), Antwerp — AI-generated patient-friendly letters

FACES OF DIGITAL HEALTH

Website: https://www.facesofdigitalhealth.com

Newsletter: https://fodh.substack.com

LinkedIn: https://www.linkedin.com/company/faces-of-digital-health

Spotify: https://open.spotify.com/show/4cElKJHrauyP6QJQaCkvdY

Apple Podcasts: https://podcasts.apple.com/gb/podcast/faces-of-digital-health/id1194284040

#TheAgenticPatient #digitalhealth #healthdata #EHDS #patientdata #caregiver #healthAI #rarediseases #interoperability #healthtech #patientempowerment #medicalrecords

35 views Aug 24, 2026 In-person video interviews"Interoperability isn't failing — it's underfunded." Herko Coomans on standards as public infrastructure.

"I am actually quite wildly uncomfortable with patients using LLMs." She built an AI companion for breast cancer patients — and she means it.

Ellyn Winters-Robinson was diagnosed with breast cancer in March 2022, months before ChatGPT launched. She wrote a book about it on her iPhone during chemotherapy. That book became AskEllyn, an AI companion used across a hundred countries. In this episode of The Agentic Patient — a Faces of Digital Health series on how patients actually use AI, which prompts, which guardrails — she talks to Tjasa Zajc about what an AI companion can hold that a clinician cannot, and why she still worries about where patient data goes.

Guest: Ellyn Winters-Robinson, CEO of The Lyndall Project and AskEllyn, author of "Flat Please Hold the Shame"

What the conversation covers:

- Building an AI companion from a book written on an iPhone during chemotherapy

- Why she keeps AskEllyn strictly non-medical, and how that guardrail held up under health-insurer review

- Whether one woman's lived experience can support patients with different cancers, cultures and languages

- Why traditional cancer support groups can become "places of collective trauma"

- Scanxiety, and what happened when she used her own chatbot during a CT scare

- Why she is uncomfortable with patients uploading medical records to ChatGPT or Claude

- Patient data rights, desperation, and the risk of being "victimized again" by AI tools

- The Canadian Cancer Society-funded study now testing whether AI companions actually help

- "The patient is the workflow" — lived experience as an untapped resource in health system design

- How clinicians can coach patients to use AI safely instead of pretending they aren't

Chapters:

00:00 Intro: why The Agentic Patient series exists

04:00 Meeting Ellyn Winters-Robinson

05:26 Diagnosed in 2022, before ChatGPT existed

07:36 From a book written on an iPhone to an AI companion

08:53 Why nurses and social workers started recommending it

09:54 Can one woman's story support every patient?

12:21 Shame, language, and cultures where breast cancer isn't discussed

13:25 Scanxiety — and taking a pep talk from your own chatbot

15:57 How AskEllyn is built on top of the LLMs

18:19 The non-medical guardrail, and how it held up under insurer review

21:51 Why patient AI use is outpacing the system

29:25 "The patient is the workflow": lived experience as untapped data

34:05 Inside the Canadian Cancer Society study

41:03 Why she's uncomfortable with patients uploading records to LLMs

45:54 The trauma healthcare never sees

6 tips on using AI as a patient: https://youtu.be/DGGVXxB4ygI?si=7m7HqCLKow51KSlQ

Faces of Digital Health:

Website: https://www.facesofdigitalhealth.com

LinkedIn: https://www.linkedin.com/company/faces-of-digital-health

Spotify: https://open.spotify.com/show/4cElKJHrauyP6QJQaCkvdY

Apple Podcasts: https://podcasts.apple.com/gb/podcast/faces-of-digital-health/id1194284040

Newsletter: https://fodh.substack.com

The Agentic Patient series: https://www.facesofdigitalhealth.com/agentic-patient

AskEllyn: https://askellyn.ai

#DigitalHealth #AIinHealthcare #BreastCancer #PatientAdvocacy #CancerSurvivorship #HealthTech #TheAgenticPatient

In this episode of The Agentic Patient, a Faces of Digital Health series on how patients are using AI to find answers the healthcare system didn't give them, Tjasa Zajc talks to Elena Ikonomovska, CEO and Co-Founder of Diadia Health. She stopped trusting chatbots with her own health data — after building an AI company on the problem they create.

Elena talks about her two-and-a-half-year journey that followed her mother's death and her own dismissed symptoms — and the causal-reasoning AI she built in response.

Guest: Elena Ikonomovska, CEO, Co-Founder & Chief AI Officer, Diadia Health

What the conversation covers:

- Why Ikonomovska calls generative AI's confident wrong answers "faithful hallucinations"

- Building a causal-reasoning engine instead of using large language models for clinical decisions

- Why lab "normal" ranges differ by genetics — and what that means for your bloodwork

- Her own dismissed thyroid and pre-diabetes symptoms, and what a two-and-a-half-year diagnosis journey actually looks like

- The risk of self-diagnosing from ChatGPT-style tools, and what to ask any AI health platform about your data

- Why she believes AI should strengthen, not replace, the doctor-patient relationship

- Early clinical results: agreement rates with physician judgment and reductions in diagnostic trial-and-error

- The equity risk in AI-driven healthcare — who gets access to validated tools, and who doesn't

Chapters:

00:00 Intro: why The Agentic Patient series exists

02:30 Meet Elena Ikonomovska and the case for causal AI

03:35 From two decades in machine learning to health AI

07:07 What the model needs: blood panels, genetics, and interactions

09:45 Elena's own diagnosis journey — two and a half years to answers

11:32 Why she wouldn't trust chatbots with her health today

12:40 "Faithful hallucinations": the hidden risk in generative AI

15:04 Inside a causal-reasoning engine built without generative AI

17:32 What happens when clinicians outsource reasoning to chatbots

21:06 Redefining "normal": genetics and personalized lab ranges

27:08 Women's health data gaps and the DTC testing boom

28:13 Strengthening, not replacing, the doctor-patient relationship

31:42 Chatbot safety advice: what patients should never share

38:07 Clinical validation, agreement rates, and what's next

Faces of Digital Health:

Website: https://www.facesofdigitalhealth.com

LinkedIn: https://www.linkedin.com/company/faces-of-digital-health

Spotify: https://open.spotify.com/show/4cElKJHrauyP6QJQaCkvdY

Apple Podcasts: https://podcasts.apple.com/gb/podcast/faces-of-digital-health/id1194284040

Newsletter: https://fodh.substack.com

The Agentic Patient series hub: https://www.facesofdigitalhealth.com/agentic-patient-blog

#DigitalHealth #AIinHealthcare #PatientAdvocacy #WomensHealth #HealthTech #ClinicalAI #TheAgenticPatient

The viral robot videos are choreography, not capability. A historian who has lived with a humanoid robot for a year explains what they still can't do.

Emily Kate Genatowski is a historian and AI domestic robotics researcher who bought a humanoid robot, lived with it for over a year, and turned the experience into a TED talk. In this episode of Faces of Digital Health, she separates the demo-reel hype of 2026 from what home robots can actually deliver — and what that means for aging populations, caregiver shortages, and healthcare systems hoping robots will help people stay independent at home.

She explains why companion robots and chore robots are still entirely separate machines, why her robot cannot get into a car, why retirees are often more enthusiastic about robots than younger workers — and why she believes we are living through a "digital Engels pause," where productivity rises faster than the mechanisms that distribute its benefits.

Guest: Emily Kate Genatowski, historian and AI domestic robotics researcher

Her TED talk: https://www.youtube.com/watch?v=rIg-Zt7bFHY

What the conversation covers:

- Humanoid robot hype in 2026 vs real-world capability

- Living with a humanoid robot: logistics, frustration, and transport

- Robots for elderly care and aging in place

- Companion robots vs domestic chore robots — why they're separate devices

- The "digital Engels pause": AI productivity without shared gains

- US, China, and EU approaches to AI and robotics regulation

- Emotional attachment to robots and robot design

- Dual-use risks: humanoid robots and drones in warfare

- "One soul, many bodies": the future architecture of home robots

CHAPTERS

CHAPTERS:

00:00 Intro

02:25 Welcome: a historian who lives with a humanoid robot

03:40 2026 robot hype: why the viral videos are choreography

05:45 Why buy a robot? How the year-long experiment began

07:28 From board games to policy: what the project became

08:57 Robots for aging in place — and who's actually optimistic

09:53 Companion robots vs chore robots: two separate machines

13:55 The digital Engels pause: productivity without shared gains

20:09 US, China, Europe: three regulatory cultures for AI and robotics

25:55 Why the robot travels in a box — and can't get into a car

29:58 Robot guilt: emotional attachment without eyes

33:14 Policy, not technology, as the bottleneck for elderly care robots

35:02 Dual use: drones, warfare, and where robotics could go wrong

37:25 One soul, many bodies: the bifurcated future of home robots

FOLLOW FACES OF DIGITAL HEALTH

Website: https://www.facesofdigitalhealth.com

Newsletter: https://fodh.substack.com

LinkedIn: https://www.linkedin.com/company/faces-of-digital-health

Spotify: https://open.spotify.com/show/4cElKJHrauyP6QJQaCkvdY

Apple Podcasts: https://podcasts.apple.com/gb/podcast/faces-of-digital-health/id1194284040

#DigitalHealth #Robotics #HumanoidRobots #AgingInPlace #HealthTech #AI

Hallucination is not the biggest risk in clinical AI. Omission is — and it is far harder to detect.

John Laursen, SVP at IMO Health, has spent his career on the layer of healthcare AI that gets the least attention: clinical terminology and the semantic data infrastructure underneath every model deployed in a hospital. IMO Health's terminology has been built and curated since 1994 and now sits behind roughly 12 billion terminology search transactions a year across US provider organisations and every major EHR.

In this interview with Tjaša Zajc, Laursen makes the case that structured data was necessary but is no longer sufficient. AI reasoning across a thirty-year patient chart needs semantic continuity — an understanding that clinical language recorded in the 1990s and language recorded today can mean the same thing. Without it, health systems are investing in models that cannot reliably interpret their own records.

The conversation also covers what happens when ambient AI scribes get it wrong, why accumulated clinical data has become a computational cost rather than an asset, and why clinician trust is the constraint that determines how fast clinical AI can move.

Guest:

John Laursen — Senior Vice President, IMO Health (Chicago, US)

Host:

Tjaša Zajc — Faces of Digital Health

What the conversation covers:

- Why omissions, not hallucinations, are the underrated risk in clinical AI

- What a semantic layer does that structured data alone cannot

- How clinical terminology maps to SNOMED CT and ICD-10 — and why those code sets were built for different purposes

- Ambient AI scribes: what happens when a model mishears or over-infers a diagnosis

- The billing and clinical consequences of an error entering the patient record

- Why problem lists hundreds of entries long now cost money in token burn

- Patient-generated and AI-generated content entering the EHR, and why health systems resist it

- Translating lay language into clinical terminology without losing specificity

- Ambient documentation, billing intensity and friction with payers

- How data quality expectations differ between the US, the NHS, the Gulf states and Singapore

- Who governs clinical data as coding complexity increases

- Why AI performance breaks down on rare disease and the difficult 20% of cases

- Knowledge graphs as a grounding source for clinical AI models

- What health systems should require from AI vendors before clinical deployment

Chapters:

02:20 Why the data layer decides what clinical AI can do

03:27 Inside IMO Health: 12 billion terminology searches a year

05:36 Keeping terminology current: SNOMED, ICD-10 and clinical governance

07:35 The semantic bridge: why structured data alone is not enough

10:17 Patient language versus clinical language in the record

12:23 When an ambient scribe mishears: clinical and billing consequences

14:53 Omissions, bloated problem lists and unnecessary token burn

19:12 Outside the US: the NHS, the Gulf, Singapore and coding complexity

20:46 Who governs clinical data as complexity increases

23:35 Patient-side AI recorders and resistance to external data

26:08 Ambient documentation, billing intensity and payer friction

29:21 The last 20%: rare disease, model limits and AI governance

33:38 Grounding, clinician trust and the cost of misfiring

Faces of Digital Health:

Website: https://www.facesofdigitalhealth.com

Newsletter: https://fodh.substack.com

Spotify: https://open.spotify.com/show/4cElKJHrauyP6QJQaCkvdY

Apple Podcasts: https://podcasts.apple.com/gb/podcast/faces-of-digital-health/id1194284040

LinkedIn: https://www.linkedin.com/company/faces-of-digital-health

#digitalhealth #healthcareAI #clinicalinformatics #EHR #ambientAI #interoperability #healthdata #SNOMED #healthIT #medicalcoding