Lucas G.’s Post

🌍✨ Exciting developments in digital health! ✨🌍 On June 2nd, PATH and Digital Square at PATH convened at the 3rd Meeting of the G20 Brasil 2024 Health Working Group for a co-branded event, "AI for Science & Health: Ensuring Equitable Impact." Supported by the Wellcome Trust and co-hosted with the World Health Organization, GIDH, and the World Economic Forum, the event addressed how artificial intelligence (AI) can transform healthcare, especially in low- and middle-income countries. Discussions covered: 🤖 The role of AI in improving clinical decision support and medical product development 🌍 The importance of localizing digital innovations to benefit the Global South ⚠ Overcoming challenges such as algorithm biases, data quality issues, and privacy concerns Pictured below, Ana Estela Haddad, Secretary for Information and Digital Health for the Brazilian Ministry of Health, spoke to Brazil’s perspective on how the G20 can help facilitate and harness the potential of AI in LMICs: “It’s important that we work collaboratively together and with responsibility, especially when we think about ethical use of AI… and that we keep on working on an equitable way to develop strength in our health systems.” Our collective goal is to channel AI's potential while ensuring ethical considerations and equitable access to healthcare innovations. This event was a significant step towards a future where AI-driven solutions support global health equity. 🌐🩺 #G20Health #AI4Health #GlobalHealthEquity #DigitalHealth #DigitalTransformation

This was great! Bilal A Mateen's opening remarks about the threats of removing humans from the center of AI framed the conversations nicely, and closing remarks from Alain Labrique drove the point home by asking not "how" AI should be used, but rather "if" - a question that is best answered only when the problem is well-defined, the target/recipient population has decreased barriers to access (like compute power! H/t Agnes Kiragga), and the foundational models used are very very _very_ well-understood. Curious to hear how funding organizations are creating AI-forward grant strategies, and how/if they are incorporating lessons learned by #digitalhealth/ #tech4dev implementers over the past few decades.

I would like to invite my network to join the Webinar on Artificial Intelligence in Healthcare happening today!

At a recent fireside chat with Jacki Davis during the European Health Summit in Brussels, we discussed how #digitalisation and #ArtificialIntelligence are revolutionising our healthcare systems and particularly the evaluation of medicines. I stressed that we should see #AI as a tool, not a magic solution and we needed human intelligence to help us ask the right questions. To be successful in the health care space, it will be important to adopt the same approaches across the entire ecosystem and this means we need to foster collaboration across sectors, both in Europe and globally. Here are some takeaways: 🚀 Opportunities and challenges: The explosion of access to health data presents immense opportunities to improve healthcare outcomes, but we need more interconnected and interoperable systems to harness the vast amounts of health data across Europe. ⚖️ Balancing regulation: The challenge lies in finding the right balance between enabling innovation and managing risks through adequate regulation. 📊 EMA's use of AI and healthcare data: Through initiatives like Scientific Explorer, EMA is demonstrating the potential of AI to improve processes and data analysis as well as better access to historical regulatory information. DARWIN EU allows us to leverage real-world evidence from 130 million patients to inform our regulatory decisions. 💡 Future proofing to support innovation: Strategic resourcing, upskilling of staff and experts, and seizing the opportunities to adopt more efficient digital procedures. 🎯 My vision for the future: Integrating AI, digitalisation and better data to improve our decision making and added value as regulators to serve European patients.  My call to action is - Let’s work together to unlock the potential of AI in creating a sustainable and competitive healthcare system in Europe! To watch the replay: https://lnkd.in/gWNUeuFQ #HealthcareInnovation #ArtificialIntelligence #DigitalTransformation #EMA #FutureOfHealth

Emer Cooke, has the European Medicines Agency determined exactly what it deems the role of AI to be in relation to the regulation of medicines and revolutionising healthcare? I think this is a critical starting point as, unchecked, and with no clear focus on what AI should/shouldn't be doing, we may be walking blindfolded into a potential minefield. I recently saw a horrendous and quite sickening example of people, in this case the church, blindly jumping on the tech bandwagon. As shocking as it is to believe, but a church in Switzerland was using an "AI Jesus" to take confessions! I don't think I need to expand any further on the inappropriateness of this. But when it comes to revolutionising healthcare with tools like AI, I believe we must ALL keep a tight focus on the right area, i.e. improving the health of the individuals.

Harnessing the potential of GenAI in health innovation in a way that is safe and effective! 🙌🏽 Our CPO, Sarah Morris, explored this topic on a panel yesterday at the Global Digital Health Network Forum with esteemed thought-leaders in healthcare from Google Health, Medtronic, Digital Umuganda, and the World Health Organization. Some of the Key takeouts: 💡Fred Hersch (Google Health): High-quality data and accurate labels are crucial for GenAI models, especially in high-stakes scenarios  💡 Anne Stake (Medtronic): Choosing the right use case avoids wasting resources, and foundational infrastructure like DPI is vital for scaling GenAI 💡Garrett Mehl (WHO): GenAI must be designed to avoid reinforcing inequities and biases in global health 💡Sarah Morris (Audere): It’s critical to understand the nuances of the audience you are building these tools for and prioritize investment in robust monitoring for safety when deployed 💡 Samuel Rutunda (Digital Umaganda): Prioritize investment in native language datasets, measure accuracy, and consider trade-offs with cost to consider viability of your use case Thank you to PATH Chief AI Officer Bilal A Mateen for putting together and moderating such a great discussion! It’s all about creating equitable, locally grounded solutions that can be scaled for global health challenges 🙌🏽 What are your thoughts on the role of GenAI in advancing healthcare? #GDHF2024

How #health #data and innovations can solve complex healthcare problems. A full day at DMEA - Connecting Digital Health I concluded at the Nordic embassies in Berlin and discussion on #EHDS and #data in healthcare. Key takeaways: ✅ Big gaps in health data use in Europe: when some countries are just slowly getting ready for the EHDS or don’t think about it yet at all, some are worried that the new EHDS compromise will force them to take a step back regarding secondary use of data (Finnland). ✅ Sweden is living proof of how digitalizing primary care can improve patient experience and use smartly limited human resources in healthcare. On-site visits are low since most of the non-emergency health problems can be solved using telecare (remark: high-quality telecare strongly supported and promoted by the state). ✅ The best way to innovate is to start with small solutions (a great example of gamification solutions in a Danish hospital) that patients will enjoy, and that will slowly build momentum. ✅ In all countries, there is a gap between legislation and the speed of development of advanced technologies. It won’t change. Innovators should cooperate with healthcare professionals and start changing the system from there or looking for opportunities within the existing incentives in the reimbursement system. Yes, there are some. ✅ We won’t enjoy the benefits of data, EHDS, and AI if we don’t have high-quality data - this is still a challenge in healthcare settings. EHDS itself won’t make magic. ✅ EHDS is a “dead or alive” question for the future of research in medicine and development of AI in Europe. Wake up, EU! ✅ If we want to build the capacity of EHDS, we need Europe to invest massively in computational resources and AI expertise. ChatGPT or Mistral (France) is not a coincidence but a result of the national solid focus on advanced tech. ✅ Large language models won’t solve all the problems of unstructured data in healthcare, even if we would wish they do. We have to do our homework regarding making data structured following standards. ✅ If data becomes accessible and available through EHDS, the next logical step is to move toward prevention-focused healthcare to transform our collected data into actions. ✅ With better data flow in healthcare, many health problems could be avoided, detected earlier, or solved entirely using virtual care. And while we are getting there, it is time to think about what impact EHDS could have on healthcare systems. ✅ We must be honest: why do we expect 100% precision from AI if humans neither deliver this 100% accuracy? AI is not an ideal solution we should strive for, but it is a supportive tool for humans. ✅ We are missing so much data in the EHR between hospital/doctor visits. And this data collected by wearables and from different data points can make a significant change. Maria Bäcklund-Hassel Päivi Sillanaukee Sveinung Tornås Julian Varghese Michael Wätjen Dr. Viola Henke

For Europe to stay competitive in AI and Data we must match the development of regulation and guardrails with investments in research and development. Almost every project that we fund at @IHI makes use of artificial intelligence and advanced analytics and over the years we have supported the development of many large, important datasets - from deep phenotyped cohorts, via mobile devices to Real World Data. A key question for Innovative Health Initiative (IHI) - and for many other funding organisations - is how these datasets can be used to drive European innovation into safe, unbiased AI tools that makes a difference to patients and healthcare systems. Our unique model of large public-private partnership allow the many groups that must have a say - patients, healthcare professionals, researchers, regulators and industry - to join forces and drive innovation in an open, trusted environment.   Looking forward to speaking on this topic at the European Tech Sovereignty Summit at the French Ministry for Europe and Foreign Affairs today and contribute to the roundtable on "AI for Health: a European story of innovation & shared sovereignty" with Yann-Mael Le Douarin, Head of the Health & Digital Transition department, Direction générale de l'offre de soins (DGOS) Derya Şahin, MD Health Policy Analyst & Advisor, OECD - OCDE Marc Julien, Managing Director - EIT Health France & Senior Director - Accelerator EIT Health #AIforHealth

Great summary and reflections by Artur Olesch ! #Data integrity including high data quality is essential if we want to transform #healthcare and #drugdevelopment. We still tend to focus on the technologies (#AI, #LLM, #automation etc.) and take shortcuts when it comes to the data and its quality. Of course, some value can be obtained without structuring and standardizing data, but the full potential of the technologies and solutions cannot be achieved without recognizing that a solid data foundation is vital 💪

3 take aways from the great conversation: i. #AIforGood in health should really mean directing investment in what matters: prevention>treatment; healthcare>chips. Today the balance *is the opposite. #NVIDIA market capitalization is equivalent to that of 165 healthcare companies... ii. There are great applied examples in radio, portable ECG or healthcare access with chatbots (w/o forgetting covid vaccines). #FF Microsoft Health Futures, Telefónica Tech eHealth, or Innit & Roche iii. (UNESCO) Ethical Impact Assessments are coming. Gabriela Ramos Natasha Crampton Amanda Craig Deckard Joaquina Salado Sabastian V. Niles Alejandro Anderlic Eugenio Minvielle