Bob Klein

What is misleading in this post is in the context of LLM as AI. To make good decisions, it requires understanding with reasoning. That is missing in LLM, not a single line of code in any system that relies on LLM is attributed to understanding with reasoning. Such algorithm is not easy to code and does not exists in the public domain. Autonomous logical understanding requires a kind of AI that is equipped with logical deduction that works in conjunction with logic. ELAINE is one of this example; a system built for analysis of written narrative in English. It takes a narrative and breaks it down into related stories with abstraction, and then discovers reasons presented by the narrator.

10

2 Comments

In case you missed it.... We published a great H:F guest episode with Joel Portice, CEO of egnite - delving into the transformative role of data and AI in revolutionizing cardiovascular care. Joel shares his journey of leading five successful healthcare data companies, including the genesis and rapid growth of egnite, which was spun out from Edwards Life Sciences. egnite's flagship platform, Cardio Care, uses advanced analytics, machine learning, and AI to identify and manage cardiovascular patients, saving lives daily. This episode is a strong addition to our Heart Health Series on Health:Further. Please check it out: https://bit.ly/4fKmzCH

59

5 Comments

Instead of using a third-party app, consider a license for the validated MMAS scales directly into your EHR. Here’s why: ✅ Simplified Licensing – Fewer terms, less hassle. 💸 Cost-Effective – No hidden software fees. 🔒 Better Data Security – Keep patient data in-house. 💻 Customizable Workflows – use internal Large Language Learning models that you control. Tailored for your clinicians. 🕒 Independence from Vendor Support – Manage updates on your schedule. Think about the benefits of keeping validated tools close to home. Better efficiency for your team, better outcomes for your patients! www.moriskyscale.com #HealthcareInnovation #EHR #PatientCare #HealthcareTech #ClinicalWorkflows #MedicationAdherence

1

How is AI being used to improve the health of every American? This week, the U.S. Department of Health and Human Services (HHS) released the FY24 AI use case inventory, which highlighted 271 use cases from across the entire Department. This inventory highlights the breadth of topic areas where AI is being applied to augment and accelerate the work of HHS. https://lnkd.in/exEjeym8 As a public servant, it was a true honor to be a part of this effort, working to improve the health and well-being of all Americans. #AI #publichealth #innovation

12

1 Comment

This would have been the talk I would have most enjoyed had I gone to HLTH ... and had Benjamin Schwartz, MD, MBA gotten the speaking slot 😁. I appreciate the nuance commentary on VBC. The concept has resonated ever since I started working with health systems and practices on population health, risk models, and VBC programs back in the early 2010s (back before VBC was cool). I do think there is truth to the fact that Value Based Care should be more of a cultural and philosphical approach than a technology approach. In fairness to all the policy wonks, I always felt VBC programs were a constant push against the contrarian incentives built into healthcare over decades of policy and financing decisions. Said another way, VBC is a patch when really the smart idea would be to restart the whole system. Which is the best path (or most likely to succeed)? No idea. I do know that reading people like Benjamin Schwartz, MD, MBA who are much smarter than I forces me to think more deeply about these problems than I would on my own accord. Enjoy Ben's HLTH speach. Well worth the read.

6

NEWS - Validation and Evaluation for Trustworthy Artificial Intelligence (VET AI) Act Introduced U.S. Senators John Hickenlooper and Shelley Moore Capito introduced the bipartisan Validation and Evaluation for Trustworthy Artificial Intelligence (VET AI) Act (details below). I'm excited about this bill because it builds on recommendations we--as the IEEE USA AI Policy committee--made in a position statement released last February (https://lnkd.in/e9sETwNM). It was a pleasure to co-lead this effort with Dr. Cari Miller. I'd also like to thank Sen. Hickenlooper's staff for being receptive to feedback that I, and colleagues, provided on this bill. Details: ➡ Directs NIST, in coordination with the Department of Energy and National Science Foundation, to develop voluntary specifications and guidelines for developers and deployers of AI systems to conduct internal assurance and work with third parties on external assurance regarding the verification and red-teaming of AI systems. ➡ Such specifications require considerations for data privacy protections, mitigations against potential harms to individuals from an AI system, dataset quality, and governance and communications processes of a developer or deployer throughout the AI systems’ development lifecycles. ➡ Establish a collaborative Advisory Committee to review and recommend criteria for individuals or organizations seeking to obtain certification of their ability to conduct internal or external assurance for AI systems. ➡ Require NIST to conduct a study examining various aspects of the ecosystem of AI assurance, including the current capabilities and methodologies used, facilities or resources needed, and overall market demand for internal and external AI assurance.

82

8 Comments

This truly highlights the tension between innovation and market dominance in health tech. Particle Health's challenge against Epic could set a significant precedent for data interoperability and patient rights. As regulatory scrutiny on monopolistic practices intensifies, it will be interesting to see how this unfolds.

2

🗞️ This Week’s News in AI Applied to Healthcare ⏵ The Sequoia Project, a nonprofit advocate for nationwide health IT interoperability, to release Version 2 of its Data Usability Implementation Guide, addressing key areas like data provenance, reducing duplication, and laboratory data interoperability. The guide aims to improve health information usability across platforms and enhance clinical workflows. ⏵ Mayo Clinic to leverage real-world data for precision medicine. In partnership with Atropos Health, Mayo Clinic uses real-time patient data to generate automated reports called Prognostograms, offering clinicians evidence-based insights for better decision-making and personalized care. ⏵ AI to evolve into a critical resource in healthcare. Experts highlight AI’s growing role in embedding whole-person care into clinical workflows. By 2025, AI is expected to enhance patient interactions and support a more holistic view of care delivery. ⏵ Hyro to raise $35M in Series B funding. Hyro’s no-code conversational AI platform automates healthcare workflows like patient scheduling and prescription refills. The funding will expand its capabilities and integrate new technologies such as large language models. ⏵ Corti to launch an AI assistant for real-time clinical support. It offers real-time insights and automated documentation during patient interactions, aiming to reduce clinician errors and improve decision-making. ⏵ Massachusetts Institute of Technology researchers to introduce a laser-based imaging technique that penetrates deeper into living tissue, offering sharper visuals of cells and aiding in disease tracking and treatment assessment. ⏵ Researchers at MIT's Computer Science and Artificial Intelligence Laboratory to teach robots their limits for safer open-ended tasks. The PRoC3S method combines language and vision models to help robots safely complete complex tasks by refining plans in a simulated environment. This ensures adaptability and safety in dynamic settings. ⏵ A new technique to reduce AI bias while preserving accuracy. It removes specific data points causing bias in training datasets, improving fairness and maintaining or enhancing model accuracy. This innovation addresses equity issues in AI applications like healthcare. ⏵ Massachusetts Institute of Technology researchers to enable AI to explain predictions in plain language. They developed a system that translates complex machine-learning explanations into clear narratives, enhancing user understanding and trust in AI predictions. ⏵ ContextCite to enhance trust in AI-generated content. This tool identifies the exact sources used by language models to generate responses, improving transparency and reliability in fields like healthcare and education. 👋 See you next week! #AI #healthcare #innovation #healthtech #medtech #precisionmedicine

20

#ahrmm #idnsummit #healthcaresupplychain #supplychainanalytics #healthcareai #erp #emr #ehr Enterprise software systems (ERP, EMR, EHR) will undergo a significant transformation in the world of generative AI. If you are a decision-maker in the health system, are you leveraging this unique opportunity? Read our white paper https://lnkd.in/eedUiGG4

36

1 Comment

Oracle Health announced plans to launch a next-generation EHR platform in 2025, incorporating advanced cloud and AI capabilities. This new EHR will feature Oracle's clinical AI agent, voice-activated navigation, and Health Data Intelligence, which integrates patient data from over 300 sources to enhance clinical workflows and improve patient outcomes. The platform aims to replace traditional EHRs, which often dictate rigid workflows, with a more dynamic system that allows for personalized care plans. Oracle Health emphasizes that the new EHR is a complete reinvention rather than an upgrade of the previous Cerner system. Additionally, Oracle plans to become a Qualified Health Information Network to facilitate standardized data sharing across the healthcare ecosystem. Executives at Oracle also unveiled new cloud applications to improve patient management and streamline front-office operations, reinforcing the company's commitment to enhancing healthcare accessibility and data utility while addressing interoperability challenges in the industry.

1

🏥🚀 AI Innovation in Healthcare: The BCBSM Success Story Blue Cross Blue Shield of Michigan (BCBSM) AI journey offers valuable insights for organizations looking to innovate with AI in regulated industries. 🔍 The Challenge: $36 billion healthcare payor 5+ million members, 37,000+ providers Increasing complexity due to ACA and aging population Rigid legacy infrastructure Strict regulatory environment 🔑 7 Key Principles for AI Innovation: 1️⃣ Get the board on board 🎓 Regular AI education sessions for board members 🔗 Link incremental projects to overall business goals 📊 Frequent updates on AI efforts and results 2️⃣ Establish cross-functional collaboration 🤝 Weekly meetings with compliance, analytics, IT, legal, business, and cybersecurity teams ⚡ Enabled rapid innovation (e.g., COVID-19 analytics dashboard in just 2 weeks) 🔄 Ongoing collaboration for continuous improvement 3️⃣ Ensure scalable, secure access 🔒 Developed SecureGPT for role-based access to AI models 🛡️ Enhanced data privacy and HIPAA compliance 📝 Logged all AI interactions for auditing purposes 4️⃣ Realize architecture matters 🏗️ Created adaptable cloud data solution with API integration 🧠 Leveraged AI embeddings for better data relationships 🔗 Reduced need for multiple master data management systems 5️⃣ Provide ongoing employee education 📚 Continuous training on responsible AI practices 👥 Cross-functional team leads education efforts 🔄 Treated AI adoption as an evolving organizational capability 6️⃣ Beat back bias 🎯 Developed model bias testing algorithms 🧪 Implemented proper training data procedures 📊 Maintained AI system inventory for governance and learning 7️⃣ Provide a safe haven for innovation 🧪 Used NASCO subsidiary for testing and development 💡 Enabled compliant innovation without disrupting operations 💰 Created potential for cost savings through solution reselling 🚀 Impressive Results: Implemented 20+ production AI applications Launched innovative gen AI tools: BenefitsGPT: Enhances customer service with quick data queries ContractsGPT: Streamlines contract management and analysis Achieved most efficient technology use in peer group Positioned as a leader in innovative AI use within healthcare 💼 Key Takeaways: AI innovation is possible even in highly regulated industries Cross-functional collaboration is crucial for responsible AI implementation Ongoing education and board involvement drive successful AI adoption Secure, scalable architecture is fundamental for AI integration Bias mitigation and governance are essential for maintaining trust and compliance 🌟 BCBSM's journey demonstrates that with the right approach, organizations can harness the power of AI to drive efficiency, improve customer service, and stay ahead in a rapidly evolving landscape. https://lnkd.in/gzdA2VHv #HealthcareInnovation #AIStrategy

8

Now that we have watered down the best working example of privacy for sensitive health care data (42 CFR Part 2 for substance use data), I'm curious to see what this is going to look like?! #HealthIT #privacy #HIE #consent

3

Mentorship: Election Analysis The 2024 presidential election revealed a significant drop in Democratic votes compared to 2020, with over 10 million fewer ballots cast for the Democratic candidate. This unexpected decline sparked a deeper investigation into voter behavior trends and the accuracy of predictive models. Leveraging historical voting data from 1976 to 2020, we built regression models to forecast the 2024 voting totals for both major parties. While our models accurately predicted the Republican vote count, they fell short in capturing the dramatic decline in Democratic turnout. By analyzing actual results from both elections, we highlight the limitations of using linear regression models for forecasting election outcomes, emphasizing the need for more sophisticated, non-linear approaches that account for the dynamic socio-political landscape. Our findings demonstrate that voter turnout is influenced by complex factors beyond historical trends, requiring a more nuanced approach to election analytics for future predictions. #DataScience #ElectionAnalysis #PredictiveModeling #VotingTrends #PoliticalData #DataDrivenInsights #LinkedInAnalytics #Election2024 #RegressionAnalysis

2

1 Comment

Medicaid MCO's, their providers, their regulators, and of course their enrollees are fundamentally interdependent. The "inescapable network of mutuality" is real in the Medicaid world. Whether it works as designed or not is largely dependent on the data that passes between them. This explainer video is my attempt to try and map the Medicaid Ecosystem from end-to end and to share how data and funding connect the different kinds of organizations that collaborate to provide healthcare for the most marginalized Amercans. Its my first attempt to do anything like this. If you find it valuable please let me know and I'll put out part two. We have a webinar coming up on this topic with some of my brilliant colleagues so let me know if you'd like to learn more.

25

4 Comments

Sorry in advance for the long post - 85% of health care executives feel #AI isn't being adopted quickly enough, there's anxiety about its role in direct patient care (this needs to be decomposed!). @EY outlines industry's readiness for AI's administrative prowess versus its involvement in medical decisions. IMHO, there are massive efficiency and fidelity gains in the HC industry that I've seen first hand through med-tech providers and @Microsoft - there is low hanging value that is proven and available now!! Thoughts on how we can responsibly harness AI for better health outcomes? Do you agree with me?

20