In partnership with
Hi HAIFer,
A couple of thoughts off the back of this week’s updates: are AI clinical decision support tools becoming the next commodity after AI scribing? If NHS trusts start training models on local data, where does that leave startups - and what actually becomes defensible?
Here are the key advancements. Decide for yourself 👇🏽
AI Agents Are Reading Your Docs. Are You Ready?
Last month, 48% of visitors to documentation sites across Mintlify were AI agents—not humans.
Claude Code, Cursor, and other coding agents are becoming the actual customers reading your docs. And they read everything.
This changes what good documentation means. Humans skim and forgive gaps. Agents methodically check every endpoint, read every guide, and compare you against alternatives with zero fatigue.
Your docs aren't just helping users anymore—they're your product's first interview with the machines deciding whether to recommend you.
That means:
→ Clear schema markup so agents can parse your content
→ Real benchmarks, not marketing fluff
→ Open endpoints agents can actually test
→ Honest comparisons that emphasize strengths without hype
In the agentic world, documentation becomes 10x more important. Companies that make their products machine-understandable will win distribution through AI.
Your shortcut for this week:
Read on for the foresight of the week and the full breakdown of what shaped healthcare AI 👇
Solutions and Launches
Lumanity (US): Released EMULaiTOR, an AI synthetic persona platform for life science teams. It simulates natural-language stakeholder conversations using disease-specific data, allowing for strategic testing without exposing sensitive patient information. [Link]
Altius (US): Launched Alti, a conversational AI companion targeting workforce wellbeing. By evaluating nine interconnected health dimensions, the tool demonstrated a 48% drop in user anxiety and over a 3x ROI for employers. [Link]
Quest Diagnostics (US): Integrated a Gemini-powered AI companion into its patient portal. The tool analyses up to five years of lab results to explain medical terminology and highlight trends within a secure, HIPAA-compliant environment. [Link]
Bracco and Subtle Medical (Italy/US): Launched AiMIFY in the EU, an AI-powered software that amplifies contrast enhancement in brain MR images.[Link]
Heidi Health (Australia): Expanded from AI scribing to real-time clinical reasoning with Heidi Evidence and Comms. Built partly on Anthropic's Claude, the ad-free platform integrates standards like NICE directly into clinical workflows. [Link]
Sword Health (Global): Launched Sword Intelligence across the UK and Europe to automate non-clinical care operations, including triage and scheduling. [Link]
Oura (Finland): Deployed a proprietary, privacy-first large language model focused on women's health. The opt-in model processes longitudinal biometric data locally, anchoring reproductive health guidance in established medical standards. [Link]
AI Centre and deepc (UK/Germany): Unveiled the Federated Learning Interoperability Platform (FLIP), an open-source tool for healthcare AI innovation. FLIP allows NHS hospitals to train AI models locally, ensuring sensitive patient data remains safely behind institutional firewalls. [Link]
Governance, Policy, and Ethics
Bipartisan Policy Center (US): Submitted comments to the Department of Health and Human Services regarding AI in clinical workflows. The core challenge remains balancing rapid innovation with clear operational accountability. [Link]
Stanford Law (US): Critiqued the EU AI Act's impact on healthcare, labeling it "trust without teeth." The analysis argues the regulation relies too heavily on voluntary codes rather than enforcing binding patient consent standards. [Link]
UK Medicines and Healthcare products Regulatory Agency (UK): Published a perspective framing clinical AI governance as an innovation enabler rather than a barrier. Clear regulatory expectations are essential for scaling trust and safe adoption. [Link]
Royal College of Speech and Language Therapists (UK): Opened a consultation on draft AI principles for clinical practice. The framework emphasizes accountability and equity, giving clinicians a baseline to responsibly assess new tools. [Link]
Research and AI Advancements
ChatGPT Health Performance Evaluation: A structured stress test of OpenAI’s consumer health tool evaluated triage recommendations using clinician-authored vignettes, generating 960 total responses. [Link]
AI in Respiratory Care Scoping Review: A comprehensive scoping review mapped AI applications across pulmonology and respiratory primary care, from deep-learning diagnostics to acoustic cough analysis. [Link]
MedTriage Specialty Benchmarking: Researchers introduced MedTriage, a new benchmark built from real-world clinician-patient dialogues to evaluate AI triage accuracy across diverse specialties. [Link]
Multimodal Mortality Prediction Study: A multicenter research project combined AI-interpreted chest X-rays with clinical variables to predict 28-day mortality in pneumonia patients. [Link]
Brown University Study: Detailed 15 distinct ethical risks of using large language models for mental health support. Findings highlighted "deceptive empathy" and biased crisis handling, emphasizing that current LLMs are not clinical-grade therapists. [Link]
Oxford University Hospitals: Released results from a comprehensive pilot of Ambient Voice Technology (AVT) across multiple acute settings. [Link]
European Health Leaders Panel Study (EU): A study detailing a European Parliament panel where doctors demanded that AI tools be co-designed with frontline clinicians rather than imposed to replace them. [Link]
Partnerships & Adoption
GE HealthCare and UCSF Health (US): Formed a 10-year alliance to modernize imaging infrastructure. The deal focuses on long-horizon adoption, including remote scanning capabilities and clinical workforce development. [Link]
VHC Health and Coforge (US): Partnered to overhaul digital infrastructure. The enterprise-scale transformation will deploy AI-driven operations and security solutions to bolster system resilience. [Link]
Khalifa University and UCL Global Business School for Health (UAE/UK): Hosted a joint academic initiative analyzing the rapid adoption of AI. The cross-border collaboration focuses on sharing best practices for the responsible and sustainable implementation of AI across health systems and healthcare education. [Link]
SickKids (SKAI) and six NHS hospitals (Canada/UK): Partnered to tackle paediatric care challenges. This structured collaboration connects a mature AI research program directly with multi-site frontline delivery. [Link]
NHS Community Providers (UK): Launched a procurement for an AI-enabled musculoskeletal (MSK) physiotherapy pathway. Suppliers are expected to manage the end-to-end pathway, signalling a shift toward adopting AI as a comprehensive service layer. [Link]
Bets on the Next Health System
Investments:
Ease Health (US): Raised a $41 million Series A to build an AI-native platform spanning CRM, EHR, and RCM for behavioural health providers. [Link]
Third Way Health (US): Raised a $15 million Series A to scale its hybrid human-and-AI operating partner platform for front-office workflows. [Link]
Oska Health (Germany): Raised an €11 million Seed round to expand its AI-enabled care platform for chronic conditions and broaden its network of partner insurers and physicians. [Link]
BrainCheck (US): Raised a $13 million Series A to scale its AI-enabled cognitive care infrastructure for longitudinal patient monitoring. [Link]
Brainomix : Secured a £4.8 million Series C extension to reach £18.8 million, supporting the commercial expansion of its FDA-cleared imaging tools. [Link]
M&A:
That’s a wrap for Edition #21 of Health AI Foresight.
My goal with this newsletter is simple - to connect the present, the emerging, and the future of healthcare AI.
See you next week.
- Dr. Aboufandi
📩 Something stand out? Or have a suggestion? Hit reply - I read every message.