Dans cet épisode, découvrez MedGPT, l’assistant médical développé par Synapse Medicine. Louis Létinier, directeur médical et cofondateur, partage les coulisses de ce projet ambitieux : de la genèse à la technologie utilisée, en passant par les enjeux de souveraineté, les sources médicales de référence et les premiers retours terrain.

Au programme :

• Pourquoi une IA 100 % française pour les soignants

• Comment MedGPT sélectionne et structure ses sources médicales

• Les différences clés avec ChatGPT et Open Evidence

• Les cas d’usage les plus pertinents aujourd’hui

• L’avenir de l’IA générative en santé en Europe

Un épisode passionnant pour comprendre les enjeux concrets de l’IA dans la pratique médicale.

Crédits :
Production : MedShake Studio
Host : Anca Petre

Hosted on Ausha. See ausha.co/privacy-policy for more information.

In this episode of Tech Anatomy, we bring machine learning out of theory and into practice with three real-world use cases that are already improving healthcare today.

Topics Covered:

• Classifying tumors as benign or malignant using logistic regression

• Predicting hospital readmissions to improve patient outcomes

• Estimating length of hospital stay with linear regression

• How hospitals are using ML models to support faster, smarter decisions

• Breaking down complex AI concepts into simple, step-by-step processes

Credits:
Production: MedShake Studio
Host: Anca Petre

✔ Stay connected and learn more:

• LinkedIn: linkedin.com/in/ancapetre

• Website: www.ancapetre.com & www.medshake-studio.com

• Email: anca@medshakestudio.com

More about the podcast:
Tech Anatomy is where health meets technology — and complexity becomes clarity.

Hosted by Anca Petre, pharmacist, entrepreneur, and health tech expert, this bilingual (English & French) podcast explores the innovations shaping the future of healthcare: AI, digital twins, data, virtual care, and more.

Each episode is a guided exploration: we dissect technology like in a lab, highlight what truly matters, and debunk common myths — health industry leaders the clarity they need to understand, anticipate, and act on technological change.

With a clear, professional, and approachable tone, Tech Anatomy makes health tech massively simplified — without ever compromising on depth or rigor.

Hosted on Ausha. See ausha.co/privacy-policy for more information.

What Is Machine Learning, Really?

In this episode, we break down one of the most hyped—and often misunderstood—technologies of our time: machine learning. Forget the buzzwords and sci-fi fantasies—this is your no-nonsense guide to what machine learning actually is, how it works, and why it matters.

Whether you’ve nodded through countless meetings or podcasts without really getting it, or you just want a clearer understanding of the basics, this episode is for you.

Topics Covered:

• What machine learning actually is (and what it isn’t)

• The difference between learning by instruction vs. learning by example

• Step-by-step breakdown: how a model is built, trained, and tested

• Real-world example: predicting diabetes using patient data

• Common challenges like messy data and overfitting

• Why understanding machine learning is empowering—not intimidating

Credits:
Production: MedShake Studio
Host: Anca Petre

✔ Stay connected and learn more:

• LinkedIn: linkedin.com/in/ancapetre

• Website: www.ancapetre.com & www.medshake-studio.com

• Email: anca@medshakestudio.com

More about the podcast:
Tech Anatomy is where health meets technology — and complexity becomes clarity.

Hosted by Anca Petre, pharmacist, entrepreneur, and health tech expert, this bilingual (English & French) podcast explores the innovations shaping the future of healthcare: AI, digital twins, data, virtual care, and more.

Each episode is a guided exploration: we dissect technology like in a lab, highlight what truly matters, and debunk common myths — health industry leaders the clarity they need to understand, anticipate, and act on technological change.

With a clear, professional, and approachable tone, Tech Anatomy makes health tech massively simplified — without ever compromising on depth or rigor.

Hosted on Ausha. See ausha.co/privacy-policy for more information.

In this episode, Anca tells the fascinating story of AlphaFold, the AI system that cracked one of biology’s greatest mysteries: how proteins fold. What started as an ambitious challenge at DeepMind has become a transformative breakthrough with far-reaching implications for medicine, drug discovery, and beyond.

Topics Covered:

• Why protein folding matters and why it was such a hard problem

• How DeepMind shifted from game-playing AIs like AlphaGo to solving core problems in biology

• The evolution from AlphaFold 1 to AlphaFold 2 and what made the second version revolutionary

• How AlphaFold uses deep learning and transformer models to predict 3D protein structures

• The impact of AlphaFold across global research, healthcare, and biotech

• What’s next in AI-powered biology—from protein complexes to drug discovery

Credits:
Production: MedShake Studio
Host: Anca Petre

✔ Stay connected and learn more:

• LinkedIn: linkedin.com/in/ancapetre

• Website: www.ancapetre.com & www.medshake-studio.com

• Email: anca@medshakestudio.com

More about the podcast:
Every week, I dive into the most transformative trends at the intersection of technology and healthcare. From AI-driven breakthroughs in diagnostics to the role of blockchain in securing health data, from decentralized science (DeSci) to NFT-powered health innovation, and from gamified fitness to the potential of digital twins, I’m here to make complex topics simple, accessible, and exciting.

Hosted on Ausha. See ausha.co/privacy-policy for more information.

In this episode, we explore a major breakthrough in the world of artificial intelligence: Chain-of-Thought prompting. This new technique teaches AI to reason step by step—just like a human would—and it’s revolutionizing how machines tackle complex problems.

Topics Covered:

• What Chain-of-Thought prompting is and how it works

• Why traditional AI models struggled with multi-step reasoning

• How Chain-of-Thought led to massive performance gains on tasks like math word problems

• The importance of reasoning transparency and trust in AI systems

• Implications for healthcare, logic, and problem-solving across industries

• How this shift signals a deeper, more human-like kind of machine intelligence

Credits:
Production: MedShake Studio
Host: Anca Petre

✔ Stay connected and learn more:

• LinkedIn: linkedin.com/in/ancapetre

• Website: www.ancapetre.com & www.medshake-studio.com

• Email: anca@medshakestudio.com

More about the podcast:
Every week, I dive into the most transformative trends at the intersection of technology and healthcare. From AI-driven breakthroughs in diagnostics to the role of blockchain in securing health data, from decentralized science (DeSci) to NFT-powered health innovation, and from gamified fitness to the potential of digital twins, I’m here to make complex topics simple, accessible, and exciting.

Hosted on Ausha. See ausha.co/privacy-policy for more information.

In this episode, we unpack a groundbreaking study that could reshape how we think about mental health care. For the first time, a fully automated generative AI chatbot—Therabot—has been rigorously tested in a randomized controlled trial to treat depression, anxiety, and eating disorders. The results? Promising, and possibly game-changing.

Topics Covered:

• How Therabot was developed and clinically validated

• Key findings from the NEJM AI-published study

• The significance of small but meaningful improvements in mental health symptoms

• Understanding therapeutic alliance with AI: can bots build trust?

• Implications for access, scalability, and the future of digital mental health care

• Ethical considerations and what still needs to be addressed

Credits:
Production: MedShake Studio
Host: Anca Petre

✔ Stay connected and learn more:

• LinkedIn: linkedin.com/in/ancapetre

• Website: www.ancapetre.com & www.medshake-studio.com

• Email: anca@medshakestudio.com

More about the podcast:
Every week, I dive into the most transformative trends at the intersection of technology and healthcare. From AI-driven breakthroughs in diagnostics to the role of blockchain in securing health data, from decentralized science (DeSci) to NFT-powered health innovation, and from gamified fitness to the potential of digital twins, I’m here to make complex topics simple, accessible, and exciting.

Hosted on Ausha. See ausha.co/privacy-policy for more information.

In this episode, we dive into HealthBench, a groundbreaking benchmark released by OpenAI to evaluate how well large language models perform in real-world healthcare conversations.

Topics Covered:

• What makes HealthBench different from previous AI benchmarks

• How GPT-3.5, GPT-4o, and the unreleased o3 model scored

• Why a 60% success rate still falls short of clinical standards

• The future role of AI in healthcare—augmentation, not replacement

• Key takeaways about responsible AI deployment in medicine

Credits:
Production: MedShake Studio
Host: Anca Petre

✔ Stay connected and learn more:

• LinkedIn: linkedin.com/in/ancapetre

• Website: www.ancapetre.com & www.medshake-studio.com

• Email: anca@medshakestudio.com

More about the podcast:
Every week, I dive into the most transformative trends at the intersection of technology and healthcare. From AI-driven breakthroughs in diagnostics to the role of blockchain in securing health data, from decentralized science (DeSci) to NFT-powered health innovation, and from gamified fitness to the potential of digital twins, I’m here to make complex topics simple, accessible, and exciting.

Hosted on Ausha. See ausha.co/privacy-policy for more information.

In this episode, we uncover the surprising reality behind your medical records: even though your electronic health record (EHR) contains rich, personal data, most of it goes unused. We explore why that is — and how new technologies are beginning to unlock its potential.

Topics Covered:

• What’s actually inside your EHR — and why it often stays siloed

• The barriers to health data sharing, from privacy laws to unclear incentives

• How synthetic data is revolutionizing research and AI training

• The rise of digital twins: virtual versions of patients used to personalize care

• What it takes to safely and ethically use personal health data in medicine

Credits:
Production: MedShake Studio
Host: Anca Petre

✔ Stay connected and learn more:

• LinkedIn: linkedin.com/in/ancapetre

• Website: www.ancapetre.com & www.medshake-studio.com

• Email: anca@medshakestudio.com

More about the podcast:
Every week, I dive into the most transformative trends at the intersection of technology and healthcare. From AI-driven breakthroughs in diagnostics to the role of blockchain in securing health data, from decentralized science (DeSci) to NFT-powered health innovation, and from gamified fitness to the potential of digital twins, I’m here to make complex topics simple, accessible, and exciting.

Hosted on Ausha. See ausha.co/privacy-policy for more information.