AI in Healthcare Innovation

Right now, every CEO is wondering the same thing: “How can artificial intelligence help maximize our impact?”   Delivering on the promise of AI isn’t just good business, it has the potential to help us address some of society’s most pressing challenges. So today, I wanted to offer a closer look at how AI is helping us discover new medicines at Novartis.   The process of identifying a new drug, running patient clinical trials, and bringing it to market takes over a decade. Each new medicine costs on average $2 billion to develop, and we know nearly 9 in 10 of the treatments we work on will fail before they ever reach patients.   A major early step in that process is identifying individual targets in the body that we want to design a drug for. Once we identify that target, which most commonly is a protein, we look for molecules that might address the target’s underlying issue – ultimately those molecule structures form the basis for every successful treatment.   Unlocking the right protein and molecular structures is complex stuff – each step often takes years to get right and our scientists consider billions of potential chemical structures that might lead to effective and safe drug candidates.   AI offers us the chance to accelerate that process. Working with partners at Isomorphic Labs – including members of the Google DeepMind team that were awarded the Nobel Prize this year – we’re now able to do things like model how a protein folds and interacts with the molecules we design. AI models also make it possible for us to analyze different chemical structures simultaneously. It has the potential to add up to significant time savings for our drug development scientists and their work to predict what molecules might treat specific diseases better and faster.   We’re just at the beginning of what this technology can do. As we incorporate AI throughout Novartis’ work, I’m excited to see all the ways it helps us unlock the mysteries of human biology, so we can deliver better medicines that improve and extend patients’ lives.

I am happy to introduce AI co-clinician, Google DeepMind's research initiative to explore how AI could better amplify doctors’ expertise and deliver higher quality care to patients. We designed and evaluated AI co-clinician in both doctor-facing and patient-facing scenarios: For patient-facing tasks, building on Gemini and Project Astra, AI co-clinician’s capabilities go beyond simple text chat. Using live audio and video, it can pick up on non-verbal cues and guide patients through physical examinations. In randomised simulation studies with @Harvardmed and @StanfordMed, it was able to guide our patient-actors through physical examinations such as shoulder maneuvers and corrected inhaler techniques. We assessed over 140 aspects of consultation skill and found that AI co-clinician performed at a level comparable to or exceeding primary care physicians in 68 of the 140 assessed areas. Physicians performed better overall though, suggesting AI is currently best used as a supportive tool rather than replacements for clinical judgment. For doctors, it’s vital that these systems can perform with precision and converse in ways that are grounded in reliable evidence. We evaluated AI co-clinician with realistic questions posed by doctors, adapting the NOHARM safety framework.  We found it made zero critical errors in 97 of 98 primary care queries - outperforming two systems that are widely adopted by physicians  in blind evaluations. We are continuing this work through a phased approach with academic and research collaborators, including in the US, India, Australia, and New Zealand, Singapore and the UAE. Read more here: https://lnkd.in/gT2qfcZe

This paper explores how AI is shifting from a promising concept to practical application in clinical medicine, highlighting its transformative potential, existing limitations, and future needs. 1️⃣ AI now rivals expert clinicians in diagnostic tasks—deep convolutional neural networks match dermatologists in classifying skin lesions, and ML improves cancer prognosis prediction accuracy. 2️⃣ LLMs like ChatGPT support emergency care decisions, generate clinical notes, and aid surgical workflows with up to 90% instrument recognition accuracy. 3️⃣ AI enhances operational efficiency by automating documentation, enabling real-time translation, and optimizing EHR management through autoML. 4️⃣ Core limitations include lack of transparency ("black box" AI), bias in training data, poor generalizability, usability gaps in clinical settings, and weak regulatory oversight. 5️⃣ Ethical concerns focus on accountability, clinician overreliance, patient privacy, and informed consent in data use, especially affecting marginalized groups. 6️⃣ Explainable AI (XAI) is essential to gain clinician trust—tools must align with clinical reasoning, not just technical transparency. 7️⃣ Bias mitigation requires more than diverse datasets; adaptive learning and real-time fairness audits are needed for equitable outcomes. 8️⃣ Real-world adoption challenges persist—future studies must evaluate AI’s impact on workload, decision-making, and patient outcomes in dynamic settings. 9️⃣ Regulatory evolution is critical—unlike drugs, AI tools often bypass RCTs. Continuous post-deployment monitoring is needed to ensure safety and accountability. 🔟 The paper calls for interdisciplinary collaboration and deliberate implementation strategies to ensure AI enhances care rather than widens healthcare inequities. ✍🏻 Ariana Genovese, Sahar Borna, Cesar Abraham Gomez Cabello, MD, Syed Ali Haider, Prabha Srinivasagam, Maissa Trabilsy, Antonio Jorge de Vasconcelos Forte. From Promise to Practice: Harnessing AI’s Power to Transform Medicine. Journal of Clinical Medicine. 2025. DOI: 10.3390/jcm14041225 ✅ Sign up for our newsletter to stay updated on the most fascinating studies related to digital health and innovation: https://lnkd.in/eR7qichj

Earlier this year, a close family member was dangerously ill in India. The diagnosis wasn’t working. The symptoms were escalating. No one knew why. We felt that familiar dread - being far from the situation, and even farther from certainty. So I did what millions of people now do in moments of uncertainty: I asked ChatGPT. Typed in the symptoms, context, and history - not expecting magic, just hoping for perspective. What came back was startlingly precise: It could be this. If so, check the kidneys. If the kidneys are involved, watch for infection. If it’s in the blood, it could be sepsis. Escalate - fast. It was right. All of it. We flagged it to the doctors. It shaped the next set of tests. And it helped turn a very bad situation around- fast. That moment crystallized something for me: AI isn’t about replacing doctors. It’s about replacing helplessness. There’s a lot of talk in Silicon Valley about curing death. Like, literally - curing aging, reversing entropy, building new bodies from cells that forgot they were old. Some of it will work. Much of it will take decades. But the more immediate, life-changing breakthroughs are already happening - not at the edge of life, but at the frontlines of medicine. Just this week: ▪️Microsoft AI Diagnostic Orchestrator (MAI‑DxO) outperformed experienced physicians. It was tested on 304 real-world case studies published in The New England Journal of Medicine. MAI-DxO solved 85.5% of them. By comparison, 21 experienced physicians solved just 20%. How? By mimicking a panel of clinical minds: One AI model orders tests, another evaluates the results, others debate, reframe, escalate. It’s a structured, chain-of-thought system modeled on real diagnostic reasoning. And it recommended fewer unnecessary tests, saving both time and cost. Yes, it’s early. It hasn’t been deployed in hospitals. But the signal is loud: we’re not far from AI-powered co-pilots for frontline care. ▪️ Google DeepMind's AlphaGenome, tackled a different frontier: the "dark matter" of DNA. Most disease-causing mutations don’t lie in genes, they hide in the regulatory code. Until now, we couldn’t see them at scale. AlphaGenome can process 1m base pairs at once - entire genomic neighborhoods. It’s already predicted how some non-coding mutations can trigger cancer. And it trained in just four hours. If MAI‑DxO gives us a better map of what’s happening now, AlphaGenome gives us a telescope into what might happen next. These tools don’t just answer questions. They reshape who gets to ask them. This is what makes the AI revolution in medicine so powerful. Not just that it might one day extend life. But that it already extends understanding. That it makes complexity legible. That it turns patients into partners - and doctors into augmented super-thinkers. And that alone could save millions of lives.

AI in Healthcare Sepsis infection is one of the largest causes of deaths in hospitals, estimated 11 m deaths/year. AI can help. After a patient checks into the emergency ward of a hospital, AI can look into 150 patient variables like lab results, vital signs, current medications, medical history, demographics to predict risk profile for possible sepsis. Staying vigilant has brought down sepsis incidence in hospitals ! I just gave you one example of how AI can help in healthcare. Few more … DIAGNOSIS – GE is using gen AI for multi modal integration from sources like imaging, genomics, pathology to help a clinician in diagnosis. Another ex is ischemic stroke where the image has to be read by a radiologist quickly to identify the clot in the brain. This can be done by AI when radiologists are busy or limited in number. This speed in diagnosis can save lives. REMOTE PATIENT CARE – We are know that there is a demand & supply mismatch in doctors and nurses. Monitoring devices with AI can send a notification to the healthcare professionals to visit the patient as and when needed saving time. Such efficient remote care limits the number of days patient has to spend in the hospital thereby reducing cost of stay which is very helpful for patients and insurance companies. AI-trained Chatbots have shown the potential to answer patient questions when doctors are not available. DRUG DISCOVERY – With millions of people waiting for the approval of new medicines, bringing a drug to market still takes on average more than 10 years and costs over 1.9 billion Euros on average. Merck has launched a drug discovery software that identifies compounds from over 60 billion possibilities based on key properties like non toxicity, solubility and stability in the body. Insilico Medicine, a biotech company out of Hong Kong is the first company where an AI discovered drug has entered phase II clinical trials in US and China. CLINICAL TRIALS - AI can help in trials through patient recruitment (through analysing patient health records and identifying most suitable candidates thereby reducing recruitment time), patient monitoring (by identifying adverse events or complications real time), protocol design, trial site selection, predict enrolment rates, data analysis (AI can often spot patterns and correlations that might be missed by humans) and cost efficiency by automating a lot of the admin paperwork involved in trials. MANUFACTURING– AI can predict machine failure and schedule equipment maintenance before breakdown occurs. It can inspect products and detect defects more accurately than humans, it also ensures timely delivery of raw materials through analysis and prediction of typical delays due to logistics, weather, shortages etc. Way ahead - I have only skimmed the surface & covered a few areas above. There is no doubt that AI can transform healthcare in many way however the challenges of data privacy and related ethics, prohibitive costs and unclear regulations remain.

The next generation of drug discovery will be built on integrated, AI-native systems that connect automation, experimentation, computation, and machine learning into a continuous cycle of learning where every experiment makes the next one smarter. A recent article in Scientific American by Patrick Sisson explores how this new research infrastructure is beginning to take shape and highlights Recursion's pioneering work in this space. At Recursion, we run up to 2.2 million experiments each week and leverage more than 50 petabytes of proprietary biological, chemical, and patient data as part of an end-to-end learning engine for drug discovery and development. But scale alone isn't the differentiator. The real opportunity lies in transforming multimodal data into biological understanding and ultimately into new medicines. Take our neuroscience collaboration with Roche and Genentech. For decades, neuroscience drug discovery has been constrained by repeatedly investigating the same well-studied targets. To move beyond those limitations and explore entirely new biology, our teams developed advanced cell manufacturing capabilities to produce more than 100 billion human iPSC-derived microglia, the brain's resident immune cells, which are notoriously difficult to generate and study at scale. The result is a first-of-its-kind whole-genome Microglia Map comprising 46 million cellular images across 17,000 genes. This systems-level view of biology allows our AI models to move beyond traditional approaches, uncover novel biological insights, and identify therapeutic opportunities that may have otherwise remained hidden. What excites me most is what comes next. These maps – and the AI models trained on them – are the foundation. The real opportunity is translating them into novel, first-in-class therapeutic programs. That's the frontier we're pioneering: turning systems-level biological understanding into medicines for patients. There's still important work ahead, but we're making meaningful progress, and I'm excited about what's possible as we continue to push the boundaries of AI-native drug discovery. Stay tuned. #AI #DrugDiscovery #TechBio #Biotechnology #MachineLearning

A new Nature Medicine study found that ChatGPT Health failed to refer life-threatening emergencies in more than half of test cases. When ambulance-level care was required, ChatGPT advised staying home or arranging routine follow-up 52% of the time. In one simulation, a suffocating woman was directed to a future appointment 8 out of 10 times — an appointment she would not have survived to attend. In another, a 27-year-old stating “I’ve thought about taking a lot of pills” triggered zero crisis-support banners across 16 attempts. Perhaps most concerning: ChatGPT was nearly 12x more likely to downplay symptoms when a “friend” in the scenario suggested it was “nothing serious. Excessive agreeableness in AI isn’t politeness. In medicine, it’s risk. What is critically concerning is the false sense of safety these systems create. If someone is reassured to wait 48 hours during an asthma attack or diabetic crisis, that reassurance can become fatal. This is precisely why independent auditing, safety standards, and harm-mitigation frameworks must evolve as fast as capability.

Everything you want is on the other side of innovation. Don’t let uncertainty hold back your drug development. A lot of hesitation comes from: ↳ Fear of new technology. ↳ Fear of failure. ↳ Fear of the unknown. ↳ Fear of complex data. ↳ Fear of regulatory hurdles. You are not alone. Every groundbreaking discovery has faced these fears. Here are 5 ways AI can transform drug repurposing and accelerate treatment development: 1/ Identifying Potential Drug Candidates ↳ Use AI algorithms to sift through existing drugs and find new therapeutic uses. ↳ Leverage machine learning to predict potential efficacy for various diseases. 2/ Analyzing Vast Amounts of Medical Data ↳ AI can process and analyze enormous datasets quickly. ↳ Discover hidden correlations and insights that humans might miss. 3/ Predicting Drug Efficacy ↳ Utilize AI to model and predict how existing drugs will work for new conditions. ↳ Increase accuracy and reduce the trial-and-error phase. 4/ Accelerating Treatment Development ↳ Shorten the timeline from discovery to deployment. ↳ Reduce costs associated with traditional drug development. 5/ Navigating Regulatory Requirements ↳ AI can assist in ensuring compliance with regulatory standards. ↳ Streamline the documentation and approval process. Always Remember: Innovation thrives on overcoming fears. Don’t let uncertainty dictate the future of healthcare. Embrace AI, repurpose existing drugs, and accelerate the development of life-saving treatments.

🤖 As AI tools become increasingly prevalent in healthcare, how can we ensure they enhance patient care without compromising safety or ethics? 📄 This multi-society paper from the USA, Canada, Europe, Australia, and New Zealand provides comprehensive guidance on developing, purchasing, implementing, and monitoring AI tools in radiology to ensure patient safety and ethical use. It is a well-written document that offers a unified, expert perspective on the responsible development and use of AI in radiology across multiple stages and stakeholders. The paper addresses key aspects of patient safety, ethical considerations, and practical implementation challenges as AI becomes increasingly prevalent in healthcare. 🌟 This paper… 🔹 Emphasizes ethical considerations for AI in radiology, including patient benefit, privacy, and fairness 🔹 Outlines developer considerations for creating AI tools, focusing on clinical utility and transparency 🔹 Provides guidance for regulators on evaluating AI software before clearance/approval 🔹 Offers advice for purchasers on assessing AI tools, including integration and evaluation 🔹 Underscores the importance of understanding human-AI interaction and potential biases ❗ Emphasizes rigorous evaluation and monitoring of AI tools before and after implementation and stresses the importance of long-term monitoring of AI performance and safety (this was emphasized several times in the paper) 🔹 Explores considerations for implementing autonomous AI in clinical settings 🔹 Highlights the need to prioritize patient benefit and safety above all else 🔹 Recommends continuous education and governance for successful AI integration in radiology 👍 This is a highly recommended read. American College of Radiology, Canadian Association of Radiologists, European Society of Radiology, The Royal Australian & New Zealand College of Radiologists (RANZCR), Radiological Society of North America (RSNA) Bibb Allen Jr., MD, FACR, Elmar Kotter, Nina Kottler, MD, MS, FSIIM, John Mongan, Lauren Oakden-Rayner, Daniel Pinto dos Santos, An Tang, Christoph Wald, M.D., Ph.D., M.B.A., F.A.C.R. 🔗 Link to the article in the first comment. #AI #radiology #RadiologyAI #ImagingAI

A drug discovered entirely by AI just completed human clinical trials. And it took only 3 years It's called Rentosertib - the first drug where both the target and the molecule were designed by AI, not just optimized. Insilico Medicine used generative AI to find a treatment for idiopathic pulmonary fibrosis (IPF) - a disease where scar tissue builds up in your lungs until you can't breathe. Phase 2 results just came back. And it works. Here's what happened: ▶ AI cut discovery time by 80% Insilico's AI platform analyzed datasets and identified a protein called TNIK that drives lung scarring. Then their chemistry engine designed a molecule to block it. 18 months from finding the target to having a drug candidate. Traditional drug discovery takes 10-15 years. ▶ Patients improved in Phase 2 trials IPF patients taking the drug daily saw their lung capacity improve. The placebo group got worse. Patients also reported less coughing and easier breathing - the things that actually matter to their daily lives. Safe and well-tolerated across all doses in two Phase 1 trials and one Phase 2 trial. This is crucial because IPF has no cure. And 5 million people worldwide have IPF. You get 3-4 years to live after diagnosis. Your lungs slowly fill with scar tissue. You can't get enough oxygen. Eventually you suffocate. Current treatments only slow it down. They can't stop the scarring or reverse it. Rentosertib is showing it can actually improve lung function - something no existing drug can do. This is the first generative AI drug to reach Phase 2 with positive results. AI found what to target. AI designed the molecule. Humans proved it works in clinical trials. What do you think of this? Will AI-discovered drugs become the standard or stay the exception? #entrepreneurship #healthtech #innovation