Healthcare AI Guy Weekly Newsletter | 1/9

Our Picks ✨

Highlights if you’ve only got 2 minutes…

1/

2024 Predictions

2023 was the year AI went from niche to mainstream about as fast as anything ever has. 2024 will be the year when the hype hopefully turns into reality while also exposing the use cases and limitations of AI at large. Here are a few ways we think that’s going to play out in healthcare:

• The year AI gets real: In 2024, the AI industry aims to deliver practical results beyond the hype, focusing on applications like early disease detection and personalized medical treatment plans. Neurosurgeon Omar Arnaout anticipates AI analyzing patient data, improving surgical precision and enhancing post-operative monitoring. Radiologist Marc Succi predicts a rise in AI chatbots for medical triage advice, prompting major tech companies to pursue health-tech acquisitions. (link)

• The rise of AI agents: We think agent-based models will show their stuff a little more convincingly than they did last year, and a few clutch use cases will show up for famously tedious processes like submitting insurance claims. (ex. AI-agent startup Amperos Health just launched out of stealth and provides AI agents for back-office healthcare ops)

• Content wars/licensing deals: The New York Times sued OpenAI for allegedly violating copyright law by training AI models on Times’ content. It’s the latest in a string of IP-related lawsuits over GenAI tech. Expect more licensing deals like OpenAI just signed with Axel Springer and changing business models in 2024. Healthcare-specific AI startups that are using healthcare data from WebMD or other sources to train their models might now need to pay for it. (link)

• Limits of monolithic LLMs become clearer: A recent study revealed ChatGPT diagnosed over 80% of pediatric case studies incorrectly. Using 100 cases, ChatGPT was tested for diagnostic capabilities in pediatric medicine & failed more than 80% of the time. Despite its high error rate, the study sees potential in AI for administrative purposes in healthcare but suggests more selective training for improved accuracy. The seemingly magical “emergent” capabilities of LLMs will be better studied and understood in 2024. We think a collection of smaller, more specific models and likely multimodal ones may prove more effective while being much easier to update piecemeal. (link)

• More rules: The year ahead “promises to introduce new layers of technological and regulatory nuance as policymakers and health care organizations grow more sophisticated in their understanding and use of AI, and as they acclimate to the ‘new normal,’ says Ronen Lavi, CEO and co-founder of Navina, a developer of AI technology that works with primary care data. (link)

• More incredible use cases: In 2023 we saw mind-blowing advancements like Swiss scientists rebuilding a spinal cord with AI. We think 2024 will only supercharge innovation. Bill Gates agrees, highlighting ambitious medical AI projects underway, from combatting antibiotic resistance to treating high-risk pregnancies. Just recently, AI was used to discover a new class of antibiotics, translate thoughts from text to brainwaves, and diagnose autism in children.

• Further dominance by large players: Not only do large health systems have higher AI adoption rates than smaller organizations, but market leaders including Epic, Google, Microsoft, and others will continue extending their lead in 2024. Recently, Epic Systems stated they will unleash a wave of generative AI innovations in the new year, with more than 60 AI-related development projects in progress. (link)

It’s hard not to be excited for all the innovation expected in the new year. As reported in our last issue, a recent KLAS report stated that 58% percent of healthcare executives said their organization will implement or acquire a generative AI solution in 2024. Expect the healthcare AI wave to only grow in 2024!

Tool Box 🧰

Latest on business, consumer, and clinical healthcare AI tools and partnerships…

🔧 UpDoc: Startup UpDoc has created a voice-based AI app to manage insulin and other prescriptions and has backing from powerhouses including Mayo Clinic and Eli Lilly. UpDoc pairs an AI assistant and a smartphone app to not only monitor patients with chronic conditions in their homes, but also to intervene with tasks like medication adjustment and lab ordering as needed. The AI is “clinician directed,” meaning it adheres to physician-created plans and loops in the care team when necessary. (link)

🔧 Intuition Robotics: Intuition Robotics' AI companion ‘ElliQ' shows promise in reducing loneliness for seniors by providing personalized conversations, entertainment, and health reminders. (link)

🔧 Amperos Health: Amperos Health just came out of stealth and focuses on integrating AI into healthcare back-office operations. Their team works to build custom healthcare AI assistants for customers. (link)

🔧 AI system translates thoughts to text from brainwaves: Researchers just developed an AI system called ‘DeWave’ that can turn silent thoughts into text by decoding brain signals without the need for invasive implants. The system achieved over 40% accuracy in translating verbs directly from neural signals. (link)

🔧 AI to help with palliative, hospice care decision: AI software could begin helping physicians know when to bring up the subject of palliative or hospice care. This is “the idea behind Serious Illness Care Connect, a software tool that about 150 doctors are testing in a pilot program in New Jersey’s largest healthcare network, Hackensack Meridian Health. (link)

🤝 Option Care Health + Palantir: Palantir Technologies (NYSE: PLTR), a leading provider of AI systems, and Option Care Health (NASDAQ: OPCH), the nation’s largest independent national provider of home and alternate site infusion services, announced a multi-year commercial partnership for Palantir’s software to be leveraged across the company to help improve patient outcomes and increase efficiency. (link)

🤝 Mayo Clinic + Aiforia + SimBioSys: Mayo Clinic is partnering with AI software developers Aiforia and SimBioSys to enhance cancer care. The collaboration with Aiforia involves a developed AI model to analyze pathology slides and data from colorectal cancer patients, estimating the likelihood of cancer recurrence based on specific tissue characteristics. The “Clinic’s work with SimBioSys will specifically target breast cancer, with a plan to co-develop new software tools aimed at improving treatment for early-stage cases of the disease.” (link)

🤝 Mayo Clinic + K Health: AI-driven primary care company K Health entered into a know-how agreement with Mayo Clinic aimed at developing an AI solution for electrocardiography-based risk assessments, remote patient monitoring, and treatment personalization. (link)

Deal Desk 💸 

Spotlight on latest capital raises, M&A and investments…

📈 Unnatural Products: a computational biology startup based in Santa Cruz, CA raised $32M in Series A funding. Merck Global Health Innovation Fund and Artis Ventures co-led and were joined by First Spark Ventures, The Venture Collective, Humain Ventures, LongeVC, and Not Boring Capital. (link)

📈 Nabla: an AI-powered copilot that generates clinical notes based in Boston, MA raised $24M in Series B funding. The round was led by Cathay Innovation and was joined by investors including ZEBOX Ventures. (link)

✅ Isomorphic Labs + Eli Lilly + Novartis: Isomorphic Labs, a spin-out of Google DeepMind, has signed its first pharma partnerships and will work with Eli Lilly and Novartis in deals worth up to $3B. Isomorphic has quickly become a leading figure in the fast-growing AI biotech space. The startup is applying Google DeepMind’s biology research – particularly its protein-structure predicting model AlphaFold – to drug discovery. (link)

✅ Guidehealth + Arcadia’s VBC Division: Startup Guidehealth buys Arcadia's value-based care division, plans to build out predictive, gen AI tech. (link)

✅ OrbiMed + Nanotronics: OrbiMed invested in Nanotronics, a New York-based industrial AI company focused on quality control and precision manufacturing for health care and biotech. (link)

Other Relevant News 🔍

News, podcasts, blogs, tweets, resources, etc…

• A robust data-driven overview for Medical LLMs (link)

• 8 predictions for healthcare 2024: what the market is signaling (link)

• Three digital health hypotheses for 2024 (link)

• Survey: some physicians enthusiastic about AI as others remain wary (link)

• Health systems using AI to improve patient safety (link)

• Study: showing clinicians AI model explanations may not root out systematic bias (link)

• AI tools could make health insurance shopping easier (link)

Visuals of the Week 📸

Funny memes, cool pics, and interesting data from around the web…