Latest news with #AIinHealthcare
Forbes
3 days ago
- Health
- Forbes
Should You Be Worried If Your Doctor Uses ChatGPT?
6 years ago, I wrote a piece, 'Doctors Use Youtube And Google All The Time. Should You Be Worried?' In 2025, it's time to ask, 'Your doctor may be using ChatGPT. Should you be worried?' In a recent unscientific survey, technology entrepreneur Jonas Vollmer asked physicians how many used ChatGPT. 76% percent of the respondents answered 'yes.' According to Volmer, a physician friend also told him, 'most doctors use ChatGPT daily. They routinely paste the full anonymized patient history (along with x-rays, etc.) into their personal ChatGPT account.' My own unofficial conversations with colleagues bears this out, with younger physicians more likely to regularly use AI than older ones I think AI tools such as ChatGPT, Grok, Claude, and other LLMs can be very helpful for physicians after they take a good patient history and perform a properly thorough physical exam. The physician can describe patient signs and symptoms with appropriate medical precision for the AI to analyze In particular, the AI can frequently suggest diagnoses that would not otherwise occur to the physician. For example, Vollmer noted that in a busy urgent care clinic, a patient might be taking some 'alternative medicines' with unusual side effects that might not be widely known in the traditional medical literature, but have been discussed in recent online articles and discussion forums. Thus, ChatGPT acts as an extension to a good physician, not a replacement. As always, the physician has the final responsibility of confirming any novel hypothesis offered by the AI with their own human judgment, which might include running additional tests to confirm the diagnosis. We've already seen non-physician patients report how ChatGPT made a diagnosis on themselves or loved ones after stumping doctors for years. And there are multiple studies showing that AI tools like ChatCPT can be surprisingly good at diagnoses when offered patient case reports. Of course, physicians need to be careful to adhere to all relevant medical privacy laws in their states/countries. And they may even consider getting explicit consent from their patients ahead of time to run their (anonymized) data through AI. Currently, physicians are allowed to seek second opinions from fellow doctors all the time, as long as privacy rules are met. The same guidelines should apply to consultations with AI. In many ways, this is comparable to how AIs in driverless cars perform comparably to human drivers. Driverless Waymo taxicabs in selected cities like Los Angeles perform as safely (or better) than human drivers in appropriately restricted settings. Tesla owners who use the self-driving mode can rely on the AI to drive safely most of the time, although they still have to be prepared to take control of the wheel in an emergency. Robot cars are not yet ready to replace human drivers in all settings (such as icy Colorado mountain highways in wintertime), but they continue to improve rapidly. Similarly, we may soon reach the point that a physician who does not use an AI consultant to double-check his diagnoses will be considered practicing below the standard of care. We are not there yet, but I can see that coming in the next few years. Summary: Tools like ChatGPT can be enormously helpful for physicians, provided that the doctor retains ultimate responsibility for the final diagnosis and treatments, and respects the appropriate privacy rules.
Associated Press
20-05-2025
- Health
- Associated Press
Can Math Help Detect Heart Disease Even Before You Know You're Sick?
Can complicated math equations help detect heart disease, even before you realize you're sick? The answer appears to be yes, and they can do it by sifting through potential clues left behind during your trips to the doctor's office. 'We now have the automated intelligence tools to help us collect and analyze mountains of medical data in ways that we couldn't just a short time ago,' said Nina Goodheart, Senior Vice President and President of the Structural Heart and Aortic portfolio businesses at Medtronic. 'Artificial intelligence (AI) allows us to examine that data to identify certain types of heart disease sooner and get personalized treatment to the patient faster.' So what does that actually mean for you? Let's use the example of Aortic Stenosis (AS) — a serious heart problem that affects 250,000 Americans every year and often goes undertreated. AS is a condition where the heart's aortic valve gradually narrows and restricts blood flow from the heart to the rest of the body. When this happens the body sends out a variety of warning signals through symptoms that can be subtle, vary from person to person, and develop slowly over time. All of which makes diagnosing the problem difficult. But whenever you visit the doctor, clues to your developing AS may be left behind. At every medical visit, lots of health information gets entered into your Electronic Health Record (EHR). It could include everything from your history of alcohol or tobacco use to vital signs to lab results to notes and observations from the doctor, and much more. The clues pointing to AS might be buried in that EHR data—but it can be a modern-day version of the needle in a haystack. That's where the math comes in. 'AI helps us bring order to what can sometimes be data chaos,' said Chris Rogers, Sr. Director of Strategic Provider Solutions at Tempus AI, a healthcare technology company based in Chicago that uses data and artificial intelligence to create medical diagnostic solutions. 'Our algorithms can look for unique characteristics in the health data that a clinician might not identify for every one of their patients.' The company's software, known as Tempus Next, includes 60 highly complex algorithms, developed to identify potential care gaps across 15 cardiovascular diseases, including AS. 'Our software is designed to handle the complexity of medical data, which requires sophisticated and adaptable solutions,' Rogers explained. 'There is no one-size-fits-all algorithm in healthcare. Each institution has unique needs and circumstances, and our technology is built to consider as much information as possible to provide tailored insights.' Tempus Next combs through the data points in your EHR, even the handwritten notes from your doctor, and compares your information to the diagnostic criteria in clinical guidelines for AS. If your health information meets enough of the criteria, the system notifies your doctor for possible next steps. It's much more than a theoretical exercise. Tempus recently finished a case study with the John Brancaccio St. Francis Hospital and Heart Center in New York. The Tempus Next algorithm identified 388 patients who met the criteria for AS and/or one other serious heart condition, but who did not yet have an existing plan to treat their disease. The Tempus Next system flagged their cases and alerted their doctors to the findings. Math has made a potentially life-changing difference in each of their lives. Now, Medtronic and Tempus are teaming up to take the next step—with a clinical study called ALERT —that will analyze health records at six more health sites. The objective is not only to identify patients who may be undertreated and get them to their next step in their care, but also to determine why that's happening and help design systems to solve the issue. 'The under diagnosis of severe symptomatic aortic stenosis is a pervasive problem that can be fatal within two years without intervention. The goal of this study is to move towards better solutions for more equitable and timely care and to eliminate barriers to treatment,' said Wayne Batchelor, M.D., M.H.S., M.B.A., Interventional Cardiologist, President of the Medicine Service Line, Inova Health System, Fairfax, Va. and Steering Committee Chair of the ALERT study. The most exciting part, experts say, is the potential impact that such technology can have on human health. 'It's breathtaking to think about the future,' Goodheart added. 'Imagine the day when we can apply this technology to hundreds of thousands or millions of patients. Imagine the number of people we can reach before their AS becomes too severe. The number of lives we can lengthen or improve. It adds up in a hurry.' That's the kind of math everyone can appreciate. This is one in a series of stories about how Medtronic is using math and AI to fight disease. Click here to learn how math can help prevent cancer. Visit 3BL Media to see more multimedia and stories from Medtronic
Yahoo
20-05-2025
- Health
- Yahoo
SIME Diagnostics Announces Results of Clinical Study on AI-Powered Lung Maturity Test in U.S. Neonatal Intensive Care Unit
LONDON and ROCHESTER, Minn., May 20, 2025 /PRNewswire/ -- SIME Diagnostics, a UK-based medical technology company today announced the completion of a clinical study evaluating an AI-powered Lung Maturity Test (LMT) for late-preterm newborns in a U.S. patient population. The findings were presented by Mayo Clinic's Rochester Neonatal team at the Pediatric Academic Societies (PAS) 2025 Conference in Honolulu on the 27th of April. Addressing an Unmet Need in the NICU Each year, an estimated 300,000 babies in the United States are born between 32 and 36 weeks' gestation, according to national health statistics.[1] This population often faces a higher risk of respiratory distress, which is a leading cause of neonatal morbidity and extended hospital stays. Traditional clinical practice often requires a "wait and see" approach to identify which infants may need intervention. This practice can contribute to higher rates of NICU admissions, delayed treatment, increased healthcare costs, and additional pressure on families and infants. The AI-based LMT aims to address this gap by providing an objective screening test to assist in the timely identification of at-risk newborns. Advancing Neonatal Respiratory Diagnostics with AI To address this clinical challenge, a study was conducted at Mayo Clinic in Rochester enrolling 207 late-preterm infants (>30 weeks' gestation).[2] Samples of gastric aspirate were analyzed using a point-of-care device developed by SIME Diagnostics. Results showed that the Lung Maturity Test (LMT) was able to assess biomarkers associated with lung maturity, and support early identification of infants at risk for respiratory distress. The study concluded that this point-of-care device accurately predicted the need for prolonged respiratory support and was effective at identifying patients whose need for respiratory support would resolve within 6 hours. Rapid Lung Maturity Test (LMT) Powered by AI and Photonics This first-of-its-kind platform consists of a point-of-care device and single-use cartridge that rapidly analyzes routinely collected samples without the need for reagents. The device enables real-time analysis of lung biochemistry by measuring key biomarkers—lecithin and sphingomyelin. The ratio of these biomarkers reflects surfactant levels, which are essential for effective oxygen exchange in developing lungs. Supporting Timely, Targeted Respiratory Care for Newborns By providing clinicians with data to guide respiratory support decisions within the first hour of life, the AI platform can help ensure early intervention for at-risk infants and minimize unnecessary treatments and interventions. Specifically designed for intensive care settings, this platform assists healthcare professionals in delivering targeted respiratory care within the first hour of life. Ongoing research and clinical evaluation will seek to refine its application and inform its potential use across various hospital environments. About SIME Diagnostics SIME Diagnostics develops AI-powered medical diagnostics platforms designed for use in acute and intensive care settings. The company's technology leverages machine learning and proprietary hardware to enable rapid, point-of-care testing for respiratory diseases. The Lung Maturity Test (LMT), the company's first product, is currently under FDA review, with additional solutions for adult respiratory diagnostics in development. More information on the company can be found by visiting or following SIME Diagnostics on LinkedIn. Conflict of Interest Statement Mayo Clinic has a financial interest in the technology referenced in this press release. Mayo Clinic will use any revenue it receives to support its not-for-profit mission in patient care, education, and research. Citations 1 2 Logo - View original content to download multimedia: SOURCE SIME Diagnostics Sign in to access your portfolio
Medscape
13-05-2025
- Health
- Medscape
Six Evidence-Based Apps Every Medical Student Should Know
When Ashwin Chetty first started his clinical rotation in pediatric surgery, he would often rely on ChatGPT and the app OpenEvidence, which is designed to help healthcare providers with summaries of peer-reviewed research; the apps helped him think of questions an attending doctor might ask him about the surgeries he planned to see the next day. These artificial intelligence (AI)–driven assists didn't replace studying. Rather, they served as a starting point for Chetty to figure out what material he needed to study more. 'Ultimately, that next day, it was super helpful,' he said of utilizing these resources. Chetty's attending doctor did ask him the questions which he identified and thought about. For Chetty, who is a second-year medical student at Yale School of Medicine in New Haven, Connecticut, these tools also help him structure what subject matter he needs to know for his clinical rotations. Given the sheer volume of material presented to medical students, you often need to pinpoint what areas you need to focus on right now, he said. 'These resources can help me do that type of prioritization,' said Chetty. Which evidence-based apps are best for helping medical students learn to diagnose illnesses, prescribe medications, and stay current on peer-reviewed medical research? Here are our picks. 1. UpToDate This app and corresponding website help medical students make point-of-care decisions by providing evidence-based information on different medical conditions. UpToDate also offers drug-related information. The app is a favorite for fourth-year medical student Isabella Ntigbu. 'This is very much the bread-and-butter clinical knowledge library used by medical providers and can be very useful to look something up on the fly while on your clinical rotations or to just learn more about a medical diagnosis or process,' Ntigbu said in an email. 'The library uses evidence-based resources and references in their articles.' Medical students can type in a question, and a suggestion will pop up for common searches. There are additional features, such as medical calculators and information on drug interactions. Healthcare providers can also access continuing education credits. A history feature allows users to revisit topics they've previously viewed, and they can create bookmarks to organize saved topics, graphics, and calculators. 'I use that a lot to do background reading and understand what the latest guidelines are for something,' Chetty said about UpToDate, which he likens to a 'Wikipedia for doctors.' Medical students must register to access UpToDate. It also requires an individual subscription or institutional support to use. 2. OpenEvidence OpenEvidence— which has a content agreement with The New England Journal of Medicine — enables users to quickly generate answers to medical questions. The app and corresponding website are based on peer-reviewed academic medical literature. OpenEvidence is a favorite of Rico Carter, a first-year medical student at Wright State University Boonshoft School of Medicine in Dayton, Ohio. 'You may also ask follow-up questions or ask for points of clarification that will drive into the topic in more detail,' Carter said in an email. Carter likes the app's ability to 'link specific literature to the topics that I discuss, ensuring the explanations are evidence based,' he said. OpenEvidence is currently free and unlimited to licensed healthcare professionals, but they must register first. 3. MDCalc Medical Calculator This app and corresponding website serve as a clinical reference and decision-making tool. The app offers a wide variety of calculators for dose calculations. It is another favorite app of Ntigbu, who plans to start her residency in emergency medicine in June. It helps with common evidence-based calculations, such as the Pulmonary Embolism Rule-out Criteria score to exclude a diagnosis of pulmonary embolism and the Clinical Opiate Withdrawal Scale score for measuring the severity of opioid withdrawal, Ntigbu said. The app offers evidence-based tools written by physician experts and supports over 50 specialties that cover over 200 patient conditions, according to the website. 'That's one I've definitely used a lot,' said Chetty, 'especially when I was in the emergency department when you're trying to risk stratify patients.' He has utilized the app to help with questions like: 'What's the risk of this patient having a deep vein thrombosis or having a pulmonary embolism?' MDCalc's core functions are free, but some parts of the app — such as earning continuing medical education credits — require a paid subscription. 4. AMBOSS The AMBOSS app and website provide a medical knowledge platform that is also evidenced-based. According to the app, it allows users to quickly define medical terms, learn how to identify details in medical imaging, and diagnose medical conditions with accuracy. The AMBOSS platform also helps medical students study for shelf exams and the United States Medical Licensing Step 1 and 2 exams, said Ntigbu. This app is 'very helpful for third-year medical students on their clinical rotations,' she said. To use most of AMBOSS's features, you need to pay a membership fee. 5. USPSTF Preventive TaskForce App This app is part of the United States Preventive Services Task Force, an expert panel that puts forth clinical preventive services' recommendations — including screenings and counseling — that are evidence-based. For example, it makes recommendations for breast and colon cancer screenings. There is also a website. The app allows medical students to 'put in different characteristics of the particular patient that you're working with, and it'll show you the different guidelines that might be relevant to that particular patient,' Chetty said, 'so that's really handy to have.' This app is free. 6. VisualDx The VisualDx app and its corresponding website bring together illustrations of disease with diagnostic frameworks. The app — which started out as a dermatology image program and now covers general medicine — allows a healthcare professional or medical student to build a differential diagnosis based on a patient's chief complaint and to search conditions in order to find treatment and testing options. 'It combines expert knowledge, a leading medical image library, and AI to support better clinical decisions, treatment, and patient communications,' Robert Sklar, director of Implementation at VisualDx and a paramedic by training, said in an email. It is 'uniquely positioned to help students by providing a comprehensive, evidence-based tool that enhances diagnostic accuracy and clinical decision-making,' Sklar added. VisualDx requires a paid subscription. Other apps that are not necessarily evidence-based but those Chetty relies on during his day-to-day work as a medical student, include ChatGPT and Sketchy— a visual depiction app and website that 'a lot of people use to study microbiology and pharmacology,' he said. Sketchy requires a paid subscription. Chetty also likes the flashcard app Anki, which is free. There are 'tens of thousands of cards that are about different medical school-related topics,' he said. 'For me, Anki is enormous,' said Chetty. '… there's so much material in med school and one of the hardest parts for me is retaining it. So being able to have cards come back and kind of have this automated review mechanism — that's been really helpful.' Which medical or study-related apps work best is a personal decision each medical student needs to make, according to Chetty. 'I think ultimately it comes down to understanding how you study and how you like to collect information and retain it and just figuring out the workflow that works best for you,' he said.
