Eugenio Zuccarelli

Interesting that on the MIMIC IV readmission models, the LLMs get closer to the traditional ML models. Perhaps because LLMs are not great at representing numeric values (especially at the model sizes tested) that are informative of the prediction for acute outcomes, whereas readmission relies more on diagnoses and other more linguistically-coded inputs. (Though that doesn't explain the poor performance on MIMIC-III readmission)

3

Insightful read on the potential of Causal Machine Learning for estimating individualized treatment effects: Full access (free): http://rdcu.be/dFb9K --- ✨ OM1, Inc. Launches Three New Products Powered by Patented AI Platform to Expedite Delivery of Personalized Medicine ✨ 🚀 🚀 ⭐ OM1, Inc. Lyra provides insights for clinicians and patients toward the most effective treatment options by generating and stratifying predictions around diagnosis, treatment, and risk to #personalizemedicine at the point of care for providers with highly relevant and actionable insights. OM1 Lyra puts the patient's uniqueness at the forefront of their treatment by translating population-level patterns to the patient through disease and treatment-specific phenotypes. Examples of this impact include: ⭐ Precision calibration of outcomes tailored to individuals, drawing insights from similar patient profiles ⭐ Amplified visibility into patient-level heterogeneity, breaking down decisions reliant on using population averages ⭐ Laser-focused identification of patient unmet needs, pinpointing actionable opportunities for intervention https://lnkd.in/et9PywCM #preventativemedicine #medicalaffairs #patientjourney #patientoutcomes #evidence #marketaccess #treatmentinsights #risk #precisionmedicine #personalizedmedicine #access #commercial #outcomes #patientcentricity #heor #healtheconomics #rwe #rwd #healthcareit #realworldoutcomes #realworlddata

35

1 Comment

A solid #GenAI report from Alex Singla, Alexander Sukharevsky, Lareina Yee, Michael Chui and Bryce Hall at McKinsey & Company. Key highlights - AI adoption is surging - GenAI adoption is increasing in business functions like sales and marketing, R&D, professional services and HR - Investments in GenAI and Analytical AI are creating value - Inaccuracy is a big risk in GenAI use https://lnkd.in/epEWHHdF

36

1 Comment

🚀 Transforming Healthcare with AI: Precision and Progress 🚀 * Rapid Drug Development: AI reduced the drug discovery process to just 30 months with Insilico Medicine, compared to the traditional decade-long timeline. *Enhanced Diagnostic Accuracy: In diagnosing complex cases, AI accurately identified tricky diagnoses 39% of the time and provided the correct diagnosis as a secondary option 64% of the time, surpassing traditional methods. *Focused Patient Care: Ambient AI allows doctors to focus more on patients by handling administrative tasks, shown in a study where 3,442 physicians adopted this technology for over 303,266 visits. *Empathetic Patient Communication: AI proved to make patients feel more heard, offering emotional support effectively, though it showed a reduction in perceived empathy when users knew the communication was AI-generated. #AIinHealthcare #HealthTech #PatientCare #Innovation Summary from: https://lnkd.in/gyqbBf5R

20

👉🏼 Practical Applications of Large Language Models for Health Care Professionals and Scientists 🤓 Florian Reis 👇🏻 https://lnkd.in/e3Apu-2f 🔍 Focus on data insights: - 📊 The article identifies six practical use cases for LLMs that can enhance the efficiency of health care professionals and scientists. - 🛠️ Strategies for customizing translations and improving text clarity are outlined, offering insights into language processing capabilities. - 🔍 Emphasis is placed on extracting relevant information and generating specialized knowledge from vast datasets, showcasing LLMs' potential in data analysis. 💡 Main outcomes and implications: - ✅ Increased productivity in research and clinical settings is anticipated with the integration of LLMs into daily workflows. - ⚠️ The article highlights the importance of user experience and the potential pitfalls of initial usage, advocating for adequate training and support. - 💭 Encouraging interdisciplinary collaboration, the piece illustrates how LLMs can assist in brainstorming and narrative compilation for research proposals and presentations. 📚 Field significance: - 🌐 This viewpoint contributes to the ongoing discourse on artificial intelligence in biomedicine, illustrating its relevance in modern health care practices. - 🧠 By addressing implementation challenges, the article fosters a better understanding of how to harness AI tools effectively in scientific environments. - 📈 It sets a framework for future research on optimizing AI applications in clinical and educational contexts, paving the way for innovation in health care. 🗄️: [#LargeLanguageModels #AIinHealthcare #DataInsights #HealthResearch #ArtificialIntelligence #Productivity #Biomedicine #UserExperience]

3

Snowflake’s BUILD 2024 AI/ML Development Conference Highlight – AI, Agents, and Data by Andrew Ng (Founder of LandingAI) The session centered on the evolution of AI with a specific focus on AI agents and the emerging paradigm of agentic reasoning. This new approach emphasizes the ability of AI to engage in iterative, goal-directed workflows, fundamentally transforming how applications are developed and deployed. With the exponential growth in the volume and complexity of data, the role of unstructured data—text, images, audio, and video—is becoming a cornerstone in driving automation and innovation across industries. Ng's discussion explored how this convergence of AI agents, agentic reasoning, and unstructured data is poised to revolutionize automation and application development in a cost-effective and scalable way. Recent advancements in large language models (LLMs) and large multimodal models (LMMs) have drawn significant attention due to their broad applicability and power. Building on these foundational technologies, AI agents are emerging as the next significant breakthrough. These agents, designed to reason and interact iteratively, are proving transformative in industries that require complex decision-making and operational efficiencies. As the emphasis shifts from models to applications, the importance of effective data engineering and the integration of unstructured data in workflows will shape the next generation of AI-driven solutions. Video Link: https://lnkd.in/e2njUdFC #ArtificialIntelligence #AIAgents #AgenticReasoning #LLMs #LargeLanguageModels #Innovation #LargeMultimodalModels #LMMs

9

The article by Issa Dahabreh et al. , "Using Trial and Observational Data to Assess Effectiveness: Trial Emulation, Transportability, Benchmarking, and Joint Analysis." offers a great framework for integrating #data from randomized trials and #observationalstudies. https://lnkd.in/dqNFm_xT #HealthcareAnalytics #ClinicalTrials #ObservationalStudies #TrialEmulation #Transportability #EvidenceBasedMedicine #DataScience #HealthOutcomes #patientoutcomes #ComparativeEffectiveness #PublicHealth #pharma #MedicalResearch #Epidemiology #rwe #realworldoutcomes #realworldevidence #rwd #dataanalytics #heor #healtheconomics #access #marketaccess #rct #outcomesresearch Johns Hopkins Bloomberg School of Public Health

25

This paper examines the economics and equity challenges of implementing LLMs in healthcare, focusing on three pathways: training from scratch, fine-tuning pretrained models, and using out-of-the-box models, and evaluates their trade-offs in cost, customization, and accessibility. 1️⃣ Training from scratch offers maximum customization and transparency by building LLMs on specific healthcare data but requires significant resources in infrastructure, workforce, and upfront costs. 2️⃣ Fine-tuning pretrained models balances cost and customization by adapting existing LLMs with healthcare-specific data but risks biases from the original training data and has limited transparency. 3️⃣ Using out-of-the-box models enables rapid deployment with minimal infrastructure and workforce needs but offers limited customization and transparency, often relying on general-purpose LLMs. 4️⃣ Cloud service providers play a critical role in scaling LLMs through managed services, federated learning, and hardware accelerators, but affordability and equitable access remain challenges. 5️⃣ Privacy, ethics, and regulatory considerations, including data deidentification and compliance with standards such as HIPAA and GDPR, are essential for LLM adoption in healthcare. 6️⃣ Potential applications of LLMs include chatbots, clinical decision support, and data summarization, but ongoing efforts are needed to mitigate biases, toxicity, and ensure model transparency. ✍🏻 Radha Nagarajan, FHIMSS, FAMIA, Midori Kondo, Franz Salas, Emre Sezgin, Yuan Yao, Vanessa K., Sandip Godambe, Naqi Khan, MD, MS, Alfonso Limon, Ph.D., Graham Stephenson, Sharief Taraman, MD, DABPN, DABPM, FAAP, Nephi Walton, Louis Ehwerhemuepha, Jay Pandit, Deepti Pandita, Michael Weiss, Charles Golden, Adam Gold, John Henderson, Angela Shippy MD, Leo Anthony Celi, Bill Hogan, Eric Oermann, Terence Sanger, Steven Martel, MD. Economics and Equity of Large Language Models: Health Care Perspective. Journal of Medical Internet Research. 2024. DOI: 10.2196/64226 📄 Planning to submit your research to a peer-reviewed, open-access journal? Use the discount code AMJB200 at checkout to save $200 USD on the article processing charge at any JMIR Publications journal. Start your submission here: https://lnkd.in/deuAEN2r. It's important to me that you know this isn't a sponsored post. I'm simply committed to keeping you informed about opportunities to disseminate your research effectively.

110

16 Comments

Some very interesting points from Yossi Matias that worth looking at, and the main one is: "Learn to code" remains crucial, even as AI advances. Here are some additional points: • Understanding tech basics is vital for everyone, not just software engineers • AI will enhance fields like biology, chemistry, and medicine • Healthcare presents a "huge opportunity" for AI transformation • Education could be revolutionized with AI-powered interactive tools • AI can be integrated in life-saving applications like Google's flood forecasting • Junior professionals may face challenges, but opportunities still exist • AI is becoming "both a tool and a science" https://lnkd.in/d5XSWb5Z #artificialintelligence #ai #chatbot #programming #healthcare #aiinhealthcare

1

👉🏼 Using Large Language Models to Address Health Literacy in mHealth: Case Report 🤓 Elliot Loughran 👇🏻 https://lnkd.in/db46KrZv 🔍 Focus on data insights: - ChatGPT effectively reduced complex language in health-related educational materials. - Content from SmartSHOTS mobile app was adapted to improve readability. - ChatGPT simplified sentence structures to align with a third-grade reading level. 💡 Main outcomes and implications: - Improved accessibility of health information for caregivers of children aged 0 to 24 months. - Enhanced adherence to vaccination schedules through clearer educational content. - Potential for ChatGPT to be a valuable editorial tool for simplifying health-related materials. 📚 Field significance: - Advancing health literacy in mHealth through AI-driven language simplification. - Enhancing communication effectiveness in healthcare education. - Promoting better understanding of medical topics among the public. 🗄️: #mHealth #HealthLiteracy #ChatGPT #DataInsights

2

1 Comment

👉🏼 Does using artificial intelligence take the person out of personal statements? We can't tell 🤓 Jenna N Whitrock 👇🏻 https://lnkd.in/eQ-PpF7U 🔍 Focus on data insights: - 📊 The ability to identify AI-generated personal statements by residency reviewers was only 59% accurate. - 🤔 Only 10% of participants correctly identified all AI-generated statements, indicating a significant challenge in detection. - 🏆 Participants rated AI-generated statements as "best" 60% of the time, suggesting that these statements are perceived favorably despite being machine-generated. 💡 Main outcomes and implications: - 🚫 A majority

1