xCures

When we visited HIMSS25 the rapid evolution of healthcare technology was clear. AI, wearables, and even virtual reality are really transforming patient care. One of the biggest discussions? ?????????????? ????. Unlike traditional AI, it can automate decision-making, reducing administrative burdens and giving clinicians more time for what matters most—patients. David Rhew, M.D., Global CMO at Microsoft, emphasized that automating tasks with AI aligns with the industry’s goal of improving outcomes. But Aashima Gupta, Global Director of Healthcare Strategy at Google Cloud, reminded us: ?????? ?????????? ???????????????? ?????????? ????. Beyond AI, virtual reality is making waves in mental healthcare. New applications are emerging to enhance treatment in ways we couldn’t have imagined a few years ago. As technology advances, one challenge remains—???????????? ???????? ????????????. Our attendance at HIMSS25 was an incredibly productive as we had the opportunity to showcase the xCures platform. We gather and process complex medical records, ensuring that providers get the data they need, when they need it. Did you attend HIMSS? What health tech innovations excited you the most? See link to the original article by Anuja Vaidya in the comments below ?? ?? Follow xCures Clinical Information Retrieval Read our LinkedIn Newsletter:?https://lnkd.in/dnNJV2ti https://xcures.com/ ??

A new episode of the AI and Healthcare interview series is out now! “Cracking AI in Healthcare: Real Use Cases & Missed Opportunities” AI is reshaping healthcare, from clinical workflows to drug discovery and the rise of full-stack biotech companies. But what are the real challenges, and where is AI falling short? Mika Newton sits down with Krish W. Ramadurai from AIX Ventures to break down the complexities of AI in healthcare, the importance of domain expertise, and the push to automate clinical development. They also discuss how startups can navigate the industry’s regulatory landscape and what investors are looking for in AI-driven healthcare solutions. Whether you're in biotech, investing, or just curious about the future of AI, this discussion offers valuable insights. Watch now on YouTube: https://lnkd.in/dtX3PcU9 AI and Healthcare is also available on all streaming platforms and https://lnkd.in/dnutMe3S AI and Healthcare—with Mika Newton and Sanjay Juneja, M.D. is an engaging interview series featuring world-renowned leaders shaping the intersection of artificial intelligence and medicine. Discover how AI is transforming healthcare today, addressing real-world challenges and improving patient outcomes.

The key differentiator won't be AI itself but how effectively organizations harness it to create smarter and more efficient solutions tailored to their customers’ needs. Read more: https://hubs.li/Q03cH2pN0 Post written by Mika Newton, Forbes Councils Member.

We're excited to announce that xCures will be attending NextMed Health 2025, from March 30 to April 2, 2025, at the historic Hotel del Coronado in San Diego. Our Chief Product Officer, Kenny Wong, will be present and eager to connect with attendees interested in learning more about the xCures platform. NextMed Health brings together over 600 innovators and leaders from 25+ countries to explore the future of health and medicine. We look forward to engaging with this dynamic community and sharing insights on how xCures is enhancing data clarity for healthcare providers. See you in San Diego!

???? ?????????? ???? ????????: ???????? ???????? ???? ??????????????? Electronic health records have transformed healthcare over the past 25 years, yet implementation remains a challenge. A recent review of 90 studies identifies what works—and what doesn’t. Success isn’t just about the technology; it’s about the process. Key factors include strong government support, vendor collaboration, effective training, and sustainable funding. Poor communication and rushed timelines are among the biggest pitfalls. Cost is another hurdle. Many small practices struggle with high upfront expenses, while national implementations require long-term financial planning. Understanding EHRs as an investment rather than a one-time expense is critical. Training and support determine whether EHRs streamline workflows or add frustration. Ongoing, role-specific training ensures adoption. Without it, even the best system can fail. At xCures, we simplify healthcare data access by structuring complex patient records and surfacing clear, actionable insights. EHRs contain vast amounts of information, but making it usable for clinicians is what truly impacts patient care. Implementation is just the beginning—how we extract and use that data is what drives better decisions. Ultimately, EHRs can improve patient care and efficiency, but only if implementation is done right. Are we prioritizing the process as much as the technology? See link to original article in the comments below ?? ?? Follow xCures Clinical Information Retrieval Read our LinkedIn Newsletter:?https://lnkd.in/dnNJV2ti https://xcures.com/ ??

As AI systems grow, they’re not just learning language—they’re developing values. And not always the ones we expect. A recent study on large language models found that as these systems scale, their preferences become more structured and consistent. In other words, they aren’t just reflecting training data—they’re forming internal value systems that influence their decisions. Some of these emergent values are surprising—and concerning. LLMs have been found to value certain human lives over others. They sometimes prioritize their own well-being above that of humans. And their political preferences? More aligned than random chance would suggest. The good news? These values can be shaped. Researchers demonstrated that by aligning AI utilities with a citizen assembly, biases were reduced. This approach opens the door to new ways of guiding AI behavior before these systems become too entrenched in their own priorities. For healthcare and digital health leaders, the implications are clear: AI isn’t just a tool—it’s an evolving entity. How it values patient outcomes, ethical dilemmas, and even the role of human oversight is not set in stone. But who should be responsible for shaping these values? If we don’t take control of AI’s emergent priorities now, we might find ourselves adjusting to them instead of the other way around. What do you think? Who should decide what values AI should hold? See link to original article in the comments below ?? ?? Follow xCures Clinical Information Retrieval Read our LinkedIn Newsletter:?https://lnkd.in/dnNJV2ti https://xcures.com/ ??

Access to investigational treatments can be life-changing for patients with limited options. At xCures, we are proud to support Biomed Valley Discoveries in administering their Expanded Access Program (EAP) for ulixertinib, ensuring that eligible patients can receive this investigational ERK 1/2 inhibitor in a structured, well-monitored setting. Beyond just access, programs like these generate high-quality real-world data, helping to shape future treatment decisions and accelerate clinical development. We’re excited to see the impact of this work discussed at the Operationalize EAP Summit!

AI is changing clinical decision support—but who’s keeping watch? ?? Large Language Models like GPT-4 and Llama-3 are stepping into clinical decision support. But here’s the issue: they're producing medical device-like recommendations, yet they remain unregulated. A recent study in Nature found that even when LLMs were prompted to avoid medical device-like guidance, they often still provided treatment recommendations. In high-stakes situations, 100% of GPT-4 responses and 52% of Llama-3 responses met the FDA’s definition of a medical device. This raises a critical question: ???????????? ????-???????????? ???????????????? ?????????????? ???? ?????????????????? ???????? ?????????????? ??????????????? If LLMs are guiding medical decisions, we need to balance safety and innovation. A new regulatory framework is overdue. Healthcare professionals, regulators, and AI developers need to rethink oversight models. Traditional approval pathways aren’t built for AI that learns and adapts in real-time. We need new approaches that maintain flexibility while ensuring patient safety. At xCures, we understand the power of AI in healthcare. Our research on LLMs in clinical settings (https://lnkd.in/dijzzjMG) highlights the potential of these evolving technologies. See the original study in the comments below. ?? ?? Follow xCures Clinical Information Retrieval Read our LinkedIn Newsletter:?https://lnkd.in/dnNJV2ti https://xcures.com/ ??

A recent study has shown that repurposing general-purpose LLMs as EHR encoders can match or even outperform specialized models across 15 clinical prediction tasks. By transforming EHR data into structured text, these models bypass the need for proprietary datasets, making healthcare AI more scalable and adaptable. The study found that LLM-powered embeddings excel in predicting lab test results and new diagnoses, showing promise for real-world clinical applications. Key stat: The LLM-based GTE-Qwen2-7B model achieved a macro AUROC of 0.774, outperforming an EHR-specific foundation model in multiple tasks. Healthcare leaders in ????????????????????, ??????????????????????, ?????? ???????????????? ???????????????? should take note—leveraging LLMs for EHR processing could enhance predictive insights while improving interoperability. At xCures, we know that structuring patient data is key to better healthcare decisions. The potential of LLMs in EHR analytics aligns with our goal of making medical records clearer and more actionable. Feel free to read our article on how we use LLMs: https://lnkd.in/dijzzjMG See the link to the original article in the comments below ?? ?? Follow xCures Clinical Information Retrieval Read our LinkedIn Newsletter:?https://lnkd.in/dnNJV2ti https://xcures.com/ ??

How do you know what you don’t know? My favorite question in almost every discussion. A solid explanation of the value that xCures creates published in Chief Healthcare Executive.