Stanford Pioneers Medical LLM ChatBot Model
Researchers at Stanford are setting up a pilot project that uses the power of large language models to help clinicians handle patient information in new ways.
Basically speaking, clinicians can ask questions to a chatbot called ChatEHR and automate charting, as well as promoting enhanced diagnostics.
Some of what people like about the software is that it's secure, direct and seamless.
Stanford chief data science officer Nigam Shah points out that for AI to be effective, it should be embedded in their workflows and have a high degree of accuracy.
'ChatEHR is secure; it's pulling directly from relevant medical data; and it's built into the electronic medical record system, making it easy and accurate for clinical use,' Shah said.
Why it Matters
Is this software valuable to clinicians?
You have to look at the current context, and what healthcare professionals need.
Too many clinicians spend a lot of their time dealing with chart notes.
Take this testimony from one frustrated medical worker posted on Reddit:
'Hello fellow docs - third year Family Medicine Resident here. Recently ramped up to 20 patients per day in clinic at my FQHC and I feel like I'm drowning in notes. I spent all of this weekend finishing notes. I know parts of the problem are: 1) using NextGen, 2) having to write extra to prove to attendings you can think.'
The poster talks about current methodology:
'I've started using the computer as much as possible in the room to get all the HPI down. I type out the whole plan before the patient leaves. I am having patients come back more instead of addressing a lot each visit. But I STILL am spending all my time doing notes it seems. This is not sustainable. I feel like I will never have a life. A scribe isn't in my near future as usually attendings have to work a few years to get one.'
Then this person asks for help from the forum.
'Any tips? How do you guys feel about charting taking over your free time?'
These new technologies have enormous potential to help with exactly this kind of workload. And from reading this person's feedback, the workload seems heavy indeed.
Feedback from Anurang Revri
At a recent IIA event, I sat in on a lecture with a panel, including Anurang Revri, Chief Enterprise Architect for Stanford Medicine.
Here are a couple of quotes that Revri brought up around the idea of automating tasks like this:
'We support research education as well as clinical care … so it's an interesting sort of place to be, where we have this kind of amazing research that we can translate into clinical workflows or external vendors and solutions. So that's sort of our focus, to implement the responsible AI life cycle.'
'When you start combining all those pieces of data and building models, multimodal models, around that, then it becomes a really powerful thing in clinical care.'
Use Cases for the Technology
It's expected that the ChatEHR technology will lead to various types of advances in automated healthcare.
A resource from Nelson Advisors puts this into five categories, so I'll outline them with bullet points:
Enhanced clinical decision support and diagnostics
Streamlined administrative and clinical workflows
Enhanced patient engagement and personalized care
Global health applications and bridging access gaps
Advanced data utilization and research
Some of these are very interesting. Under decision support, you have advances in treatment. With administrative issues, you see how this works with the patient life cycle.
Under the fifth column of 'advanced data utilization and research,' you have the idea of digital twins, in order to house a person's health data in a certain digital entity – and I think that wearable devices will come in handy here.
All of it is likely to help us make medicine more effective, and do earlier interventions for patients.
And that's a big deal. Look for this type of technology to blossom in the healthcare industry as a whole.
Hashtags
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Business Wire
21 minutes ago
- Business Wire
Paige and Oncoclínicas Strengthen Leadership in AI-Driven Cancer Diagnostics Across Brazil
NEW YORK & SíO PAULO--(BUSINESS WIRE)--Paige, a leader in next-generation AI technology, announced the continued and expanded use of its innovative AI-assisted diagnostic applications by long-time customer, Oncoclínicas&Co (Oncoclínicas), the largest private group of cancer care in Latin America. With a commitment to enhancing diagnostic capabilities and improving patient safety, Oncoclínicas has utilized Paige's technology across prostate and breast cancer diagnosis, supporting its commitment to advancing pathology standards throughout Brazil. Oncoclínicas employs a rigorous approach where all digitized prostate and breast cancer biopsy whole-slide images are analyzed by pathologists using AI as an adjunct diagnostic tool. This process provides extra layers of patient safety, enhancing both sensitivity and efficiency for the pathologists using the technology. Since August 2021, Oncoclínicas has utilized AI to support the review of over 23,000 prostate and breast biopsies. These results demonstrate the scale and commitment of Oncoclínicas to AI integration in their diagnostic workflow, which has led to Oncoclínicas' achieving: Faster second reader sign-out for senior pathologists using FullFocus™ and AI support. 1 Enhanced detection rates, including the identification of approximately .5% of precursor non-obligatory breast lesions that were not initially detected by pathologists. 1 Improved diagnostic consistency and efficiency through the use of Paige Prostate and Paige Breast AI Suites. 1 Dr. Leonard Medeiros da Silva, pathologist at OC Medicina de Precisão, an integrated laboratory of Oncoclínicas&Co dedicated to pathological anatomy, genetic testing and Big Data, reports a positive experience with Paige Prostate. "It aligns with Oncoclínicas published data showing nearly 100% sensitivity and 93% specificity,' he notes. 'The specialist also commends the technology's effectiveness in both cancer detection and accurate Gleason score determination.' 'We are proud to support Oncoclínicas in their mission to offer cutting-edge cancer diagnostic solutions,' said Dr. Juan Retamero, Medical Vice President at Paige. 'Their success in implementing AI technology is a testament to their dedication to clinical excellence and their forward-thinking approach to pathology. Our long-term partnership underscores Paige's commitment to providing access to high-quality AI technology globally, helping pathologists provide better diagnoses for their patients.' 'Over the past few months, I've had the opportunity to witness how the integration of digital pathology, biomarkers, AI, and genomics is accelerating the advancement of precision medicine, and, more importantly, improving patients' quality of life and offering new hope in the fight against cancer,' said Rodrigo Alvarenga, Operations Executive at Oncoclínicas&Co. 'At both the 2025 USCAP in Boston and ASCO in Chicago, it became clear that precision medicine is no longer a distant goal, but a present-day reality that can expand access to advanced technologies and enable more cost-effective care. We've reached an exciting point of no return, the convergence of AI, pathology, and genomics is not only transforming cancer care but is also critical to the long-term sustainability of our healthcare systems.' In Brazil, FullFocus™ and all AI assisted diagnostic applications such as the Paige Prostate Suite and the Paige Breast Suite are for Research Use Only. Not for use in diagnostic procedures. 1 Figures are based on independent clinical research conducted by Oncoclínicas&Co. Finding an average of 12,000 breast biopsies and surgical specimens were analyzed in 2024. The Paige Breast Suite as a second reader detected 55 non-obligatory precursor breast lesions that were not initially detected by pathologists. Additionally, after utilizing FullFocus™ with the Paige Prostate Suite, the average time to read a slide and complete a report reduced to one minute and 10 seconds. About Oncoclínicas&Co Oncoclínicas&Co is the largest group dedicated to cancer treatment in Latin America, with a highly specialized and innovative model focused on the oncology journey. With a medical staff of over 2,900 oncology specialists, the company operates in 40 Brazilian cities, totaling more than 140 units. Focused on research, technology, and innovation, the group has conducted approximately 682,000 treatments in the past 12 months. Oncoclínicas follows international standards of excellence, integrating outpatient clinics with high-complexity cancer centers, enhancing treatment through precision medicine and genomics. As the exclusive partner of the Dana-Farber Cancer Institute in Brazil, affiliated with Harvard Medical School, it has acquired Boston Lighthouse Innovation (USA) and holds a stake in MedSir (Spain). It is also part of the IDIVERSA index of B3, reinforcing its commitment to diversity. With the goal of expanding its global mission to overcome cancer, Oncoclínicas has entered Saudi Arabia through a joint venture with the Al Faisaliah Group, bringing its oncology expertise to a new continent. Learn more at: About Paige Paige is pushing the boundaries of AI to solve cancer's most critical issues, revolutionizing cancer care with next-generation technology. By leveraging exclusive access to millions of digitized pathology slides, clinical reports, and genomic data, Paige gains a holistic understanding of cancer, encompassing diverse factors such as gender, race, ethnicity, and geographical regions. This comprehensive data enables Paige to create advanced AI solutions that redefine cancer detection, diagnosis, and treatment. With a unique, intricate understanding of tissue, Paige sets new standards in precision diagnostics, earning the distinction of being the first FDA-cleared AI application in pathology. Paige has also developed the first million-slide foundation model for cancer, continuing to lead the way in uncovering novel insights and transforming them into life-changing products. For more information, visit
Forbes
22 minutes ago
- Forbes
Your Business' AI Has A Digital Immune System—Here's How To Protect It
Anand Oswal, SVP & GM of Network Security, Palo Alto Networks. In the last couple of years, artificial intelligence (AI) has transformed from a lab concept into the central nervous system of modern business, powering everything from developer tools to customer support. Now, with the rise of autonomous AI agents that can learn, reason and act on our behalf, we are not just using a new tool; we are introducing a new, dynamic entity into our corporate bodies. This represents a profound architectural shift. The old three-tier application stack was like a simple organism. The move to cloud computing modernized this system, but the fundamental constructs were similar. But the AI stack—with its models, new uses of data for training and inferencing, plugins and autonomous agents—is a whole new level of biological complexity. It can learn, adapt and interact in ways that are often unpredictable, which creates an entirely new landscape for potential attacks. Just like any highly evolved organism, this new digital nervous system is born vulnerable. It doesn't yet have the mature, battle-hardened immune system needed to defend against a new generation of sophisticated threats. To truly protect it, we need an equally sophisticated, built-in defense mechanism. Business leaders I talk to every day are keenly aware of this challenge. They see the immense promise and competitive advantage that AI offers, but they're also grappling with the hidden risks that come with adopting it. A common concern is the lack of visibility into which AI apps are running, what data they're accessing and what permissions they have. To navigate this new world confidently, security can no longer be a simple shield; it must evolve to be a comprehensive, adaptive immune system. This means building a defense system that is deeply integrated, intelligent and flexible enough to tell the difference between helpful and harmful actions, neutralize threats in real time and continuously learn from every interaction to become stronger and more resilient over time. Just as our own biological immune systems have evolved over millennia to protect us from countless unseen threats, our approach to cybersecurity must also evolve dramatically for the age of AI. Here's what that essential digital immune system needs to do: 1. Know Your Genetic Makeup: Proactive Scanning and Posture A healthy immune system starts with understanding its own body. For an AI ecosystem, this means having a complete, dynamic inventory of all components and actively screening every component before it enters your ecosystem. Start by integrating automated model scanning directly into your development pipeline. Before any open-source model is downloaded (from a repository like Hugging Face), it must be scanned for embedded malware, unsafe code within files or signs of model poisoning. To maintain the equivalent of a "healthy lifestyle," you will need to enforce strong posture management, monitoring your entire AI ecosystem for risks like excessive permissions or sensitive data exposure. This ensures your AI doesn't develop chronic vulnerabilities. You will also want to ensure that automated alerts notify you if, for example, a customer service chatbot attempts to access a sensitive financial database or if a developer grants an AI model overly broad permissions that violate the principle of least privilege. 2. Build Resilience Through Exposure: AI Red Teaming An immune system strengthens itself through controlled exposure, and your AI is no different. Go beyond theoretical assessments and implement an automated AI red teaming program that acts as a training ground against real-world attacks. You can begin by using established frameworks, such as the OWASP Top 10 for large language models, to simulate common attacks like prompt injection and data leakage. Take it to the next level by expanding your tests to include more than just technical risks, but also business-specific threats. For example, could an agent be tricked into issuing unauthorized refunds? Could a compromised internal tool manipulate a report for an executive? You must create a direct feedback loop where the findings from these red team exercises are used to immediately fine-tune models, update security filters and provide targeted training for your developers. 3. The Real-Time Immune Response: Runtime And Agent Security When an active threat emerges, the immune response must be swift and immediate. A key step is to implement protections at runtime, capable of analyzing the prompt intent and responses to block sophisticated injection attacks that built-in protections would miss. For autonomous agents, the stakes are even higher since agents have memory and the authority to use other tools. A truly robust digital immune system must be able to distinguish a healthy agent from one that has been hijacked, neutralizing its ability to cause harm from within. To do this, enforce the principle of least privilege by scoping out the tools and APIs each agent can access. For example, an agent designed to schedule meetings should have no ability to read emails. An additional safeguard would be ensuring you can automatically disable an agent if it is compromised or begins behaving erratically, like performing an unusual number of actions or making unusual data updates. Also, be sure to log every action taken by an agent, from the prompt it received to the tools it used and the output it generated, to ensure you can conduct effective forensic analysis if an incident occurs. Forging A Path In The AI Era To gain a competitive edge in today's fast-moving market, businesses must embrace AI's transformative power. But they simply can't do this with a compromised digital immune system. Relying on fragmented security tools that act merely as simple, isolated shields is no longer sufficient. What's needed is a holistic and integrated security approach that protects every single AI model, every dataset and every autonomous agent across the entire AI lifecycle, from development to deployment and ongoing operation. By proactively building and maintaining this comprehensive digital immune system, organizations not only protect their enterprise from evolving threats but also empower themselves to confidently navigate the future, innovate at a faster pace and lead with assurance in the exciting and complex age of AI. Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?
Yahoo
an hour ago
- Yahoo
PagerDuty (PD) Climbs 7% on Analyst Upgrade, Sale Reports
We recently published . PagerDuty, Inc. (NYSE:PD) is one of the best-performing stocks on Monday. PagerDuty rallied for a second day on Monday, adding 7.06 percent to close at $16.83 apiece as investors took heart from an analyst upgrade and news that it was exploring a sale. In its market note, investment firm TD Cowen raised its price target and recommendation for PagerDuty, Inc. (NYSE:PD) to $22 from $17 and to 'buy' from 'hold' previously. This followed a report by Reuters on Friday, quoting sources privy to the matter, that PagerDuty, Inc. (NYSE:PD) was exploring a potential sale after receiving interest from buyers. Reuters said PagerDuty, Inc. (NYSE:PD) is currently working with Qatalyst Partners to facilitate the potential acquisition, and that investment bankers are now soliciting further buyer interest. Copyright: stokkete / 123RF Stock Photo According to TD Cowen, Qatalyst Partners has a strong track record of facilitating software mergers and acquisitions, adding that many of its facilitated transactions resulted in sales over the years. PagerDuty, Inc. (NYSE:PD) is a California-based software maker that helps businesses monitor their IT systems and respond to cyber incidents and outages. While we acknowledge the potential of PD as an investment, our conviction lies in the belief that some AI stocks hold greater promise for delivering higher returns and have limited downside risk. If you are looking for an extremely cheap AI stock that is also a major beneficiary of Trump tariffs and onshoring, see our free report on the . Sign in to access your portfolio
