ACDM: Association for Clinical Data Management’s Post

AI/ML Breakout Stream 1 at ACDM24…… The ACDM: Association for Clinical Data Management would like to thank the presenters of the sessions that took place during the first AI/ML Breakout Stream: > Lisa Moneymaker (Saama) who presented “Unleashing the power of AI to transform clinical development - today” > Kian Norouzi (Novo Nordisk) who presented “Leveraging Language Models for Clinical Data Management: A Case Study on Text Cleaning and Mapping” > Richard Yeatman (IgniteData) who presented “Where next for Clinical Trials EHR eSource data transfer – and is AI the answer?” #clinicalresearch #clinicaltrials #clinicaldatamanagement #clinicaloperations #edc #ecoa #acdm #ACDM24

👉🏼 Clinical Accuracy, Relevance, Clarity, and Emotional Sensitivity of Large Language Models to Surgical Patient Questions: Cross-Sectional Study 🤓 Mert Marcel Dagli 👇🏻 https://lnkd.in/gZXYWQdk 🔍 Focus on data insights: - The study evaluated the clinical accuracy, relevance, clarity, and emotional sensitivity of responses provided by large language models (LLMs) to surgical patient questions. - High performance of LLMs was observed across various metrics, including accuracy, relevance, clarity, and emotional sensitivity. 💡 Main outcomes and implications: - LLMs, particularly Anthropic's Claude 2, demonstrated superior performance compared to other models like OpenAI's ChatGPT and Google's Bard. - The findings suggest that LLMs have the potential to enhance information delivery and improve patient-surgeon interaction. 📚 Field significance: - This study highlights the promising role of LLMs as adjunct tools in patient communication and education within the surgical field. 🗄️: [#clinicalaccuracy #LLMs #surgicalpatients #datainsights]

🎯 Nature Paper Alert: How LLM Agents Are Transforming Healthcare Decision-Making 🔬 Just Read: Groundbreaking Article in Nature Machine Intelligence on LLM-Based Agentic Systems in Healthcare[1] (https://lnkd.in/gWSQek5g) Introduction Large Language Models (LLMs) are undergoing a transformative evolution in healthcare, moving beyond simple text generation to become sophisticated agentic systems (LLM agents). While traditional LLMs excel at following instructions and providing information, their limitations in memory, multimodal processing, and continuous learning have restricted their practical applications in medicine and biotechnology. The emergence of LLM-based agentic systems represents a significant leap forward, combining the linguistic capabilities of LLMs with the ability to process multiple types of input, maintain long-term memory, and actively engage with external tools and other AI agents. “In healthcare, an LLM agent will be able to not only suggest diagnoses but also schedule appointments while taking into account patient location, medical urgency, and physician availability. Possession of long-term memory enables agents to recall a patient’s medical history, medications and preferences, which facilitates personalized care. ”[1] As a side note for those in biotechnology, LLM agents could revolutionize biotech operations by serving as intelligent coordinators across research, development, and manufacturing processes. These systems could simultaneously manage lab automation, analyze complex multi-omics datasets, and maintain regulatory compliance while interfacing with various instruments and databases. LLM agents could function as collaborative partners in biotech research, coordinating between different specialized AI agents (like those focused on genomics, proteomics, or drug discovery) while maintaining long-term memory of experiments and institutional knowledge. Key takeaways from this fascinating piece by Qiu, Lam, et al. on the future of AI in medicine: 💡 Four Major Healthcare Opportunities with LLM agents: 1. Clinical workflow automation (potentially automating up to 47% of tasks) 2. Enhanced medical AI trustworthiness through self-verification 3. Multi-agent diagnostic systems mimicking medical team collaboration 4. Health digital twins for personalized care ⚠️ Critical Challenges to Address: - Safety and security concerns - Bias mitigation - Over-reliance risks - Regulatory framework needs 🔮 Future Vision: AI evolving from mere assistant to true healthcare colleague, with personal AI agents helping manage individual health data. [1] Qiu, J., Lam, K., Li, G. et al. LLM-based agentic systems in medicine and healthcare. Nat Mach Intell (2024). https://lnkd.in/gP_KCYUb Visit us at AINovaStrategy.com and sign up for our newsletter or schedule a chat. #HealthcareAI #ArtificialIntelligence #DigitalHealth #FutureOfMedicine #MedTech

Recently, the "2nd National Digital Health Innovation Application Competition - Health and Medical Big Data Theme Contest" concluded successfully. Yidu Tech, as a joint applicant, participated in multiple competition categories and was awarded five prizes, leveraging its strengths in large language models (LLMs) and artificial intelligence! The 2nd National Digital Health Innovation Application Competition was hosted by the National Health Commission of China, with the Health and Medical Big Data Theme Contest organized by the National Health Commission's Statistics and Information Center. Following on-site evaluations by expert judges, 158 teams were awarded prizes. Yidu Tech, in collaboration with several medical institutions and health information centers, excelled in areas such as generative medical AI model evaluation, LLM-based medical guidance model evaluation, and health and medical big data applications, winning second place, third place, and excellence awards, respectively. #DigitalHealth #AIHealthcareSolutions #GenerativeAIinMedicine #LLM #MedicalAI

The rapid advancement of generative AI, exemplified by LLMs and GPTs, has initiated a paradigm shift in drug development. At #ISPOR2024, we will host a workshop to examine latest guidelines on external control arms (ECA) worldwide, with a specific focus on integrating generative AI to enhance the utilization and efficacy of RWD in ECA development. Thanks Paul Arora from University of Toronto, and Mei Yang from Happy Life Technology (HLT) for working together on this.   If you're involved in HEOR, clinical trials, RWD, RWE, AI or any other health care/life sciences sectors, we encourage you to take part in this brief survey. We aim to collect diverse viewpoints on the role of generative AI in ECA development and its integration with RWE. Your insights will significantly enrich our workshop, offering a well-rounded perspective to the community. Please note that the survey is completely anonymous, and no personal information will be collected. We will share the aggregated responses and discuss them in detail publicly. #generativeAI #LLMs #ECA #GPT #clinicaltrials #RWD #RWE

🤖 AI-Driven Success: Meet Ask Vivo, your AI assistant for clinical development. Our generative AI interface plugs into Vivo’s clinical trial data warehouse, allowing you to obtain rapid answers to questions about your trial data and documents. Get actionable insights to enhance your decision-making. See Vivo in action: ▶️ https://lnkd.in/gkbaQ4aC ◀️ #clinicaltrials #clinicalresearch #biopharma #datascience

Here’s a latest developments in clinical research: "Advancements in Clinical Research” The clinical research landscape is evolving rapidly! From the rise of precision medicine to the adoption of AI and machine learning, the industry is transforming. - AI-designed trials for greater success rates - Real-World Data (RWD) guidance from regulators - Causal machine learning for accurate insights - Data privacy technology for secure research - Precision medicine for targeted efficacy - Operational excellence through AI adoption - Increased funding for AI-enabled drug discovery Stay ahead of the curve and let's shape the future of clinical research together! #clinicalresearch #precisionmedicine #AIinresearch #dataprivacy #machinelearning #researchinnovation"

What is unlimited, helps machines fare, Balanced and tunable if crafted with care, Safe and scalable, privacies keep, Fast and available, yet fairly cheap? Synthetic data use across the AI lifecycle (registration required) #osel #regulatoryscience #cdrh

It's almost impossible to get through a day without hearing about artificial intelligence at home, work or in your LinkedIn feed. But beyond the chatbots and image generators lies a far more powerful set of use cases within the clinical research ecosystem. We'll uncover it all at Medidata Solutions NEXT New York! Don't miss my session "Shortening Trial Timelines with Better AI Enabled Trial Design”. I will be joined by Bristol Myers Squibb's Revathi A. and Fusion Pharmaceuticals’ Michael Groaning, for a deep dive into how AI and clinical research data are being used to unlock exciting medical breakthroughs. Learn more and register here: https://mdso.io/nvq #MedidataNEXT #LifeSciences #Medidata

AI and Research.. Exploring the intersection of technology and research opens up a world of possibilities. The application of AI in research brings efficiency; however, it also raises ethical considerations. Oversight and regulatory compliance are crucial to ensure all aspects are covered. The question remains - who oversees AI? Building trust in AI will be a gradual process. This podcast was an informative listen. #AI, #innovation, #research Read more about the ethical implications of AI in clinical trials: https://lnkd.in/g7wFPaH3