Cedars-Sinai published research in JAMA Cardiology this week on a algorithm it developed to detect atrial fibrillation in people not yet showing symptoms. The study found that the AI model does a good job of identifying hidden signals in common medical diagnostic testing and could help physicians better prevent strokes and other heart-related complications in people with atrial fibrillation.
Lucem Health recently closed a $7.7 million Series A funding round. The startup's platform takes in the data required to fuel AI algorithms that give clinicians insights to improve patient care.
As technology advances, AI-powered tools will increasingly reduce the administrative burdens on healthcare providers.
Consumers want the healthcare industry to meet them where they are, and this shift in patient mindset has expanded the definition of home healthcare and who benefits from it, according to a panel at MedCity INVEST Digital Health conference.
A review of 12 studies about AI systems used to screen patients for breast cancer found that they lacked evidence to prove that they would work well in a clinical setting. Researchers highlighted concerns about study quality, including the fact that all of the studies were retrospective and many of them had a high risk of bias.
In an analysis of Epic’s sepsis model, researchers at the University of Michigan found that it performed “substantially worse” than what the developer reported. According to their analysis, it didn’t detect many undiagnosed sepsis cases, but still gave alerts for one in five hospitalized patients.
It's important for clinicians to consider AI recommendations with a critical eye, the same as they would for any other test, said Dr. Erich Huang, Director of Duke Forge. In our latest podcast episode, he spells out the stakes for implementing algorithms in healthcare, which can potentially affect thousands of patients.
The future looks bright for the integrated specialty pharmacy model – for health systems, providers, and most importantly, for patients.
The nation's largest public health system will stop using clinical assessments for kidney function and vaginal delivery after C-sections that use race-based calculations for determining illness severity and risk. These assessments can result in people of color not receiving the diagnoses and treatments they need.
Of 26 health systems surveyed by MedCity News, nearly half used automated tools to respond to the Covid-19 pandemic, but none of them were regulated. Even as some hospitals continued using these algorithms, experts cautioned against their use in high-stakes decisions.
In a review of 130 FDA-cleared AI tools, researchers found that the vast majority of them relied on retrospective studies. Most of the developers also didn’t share information on how many sites the algorithms had been tested in, or how they performed in different patient demographics.
A recent study published in Nature found that an algorithm was able to detect causes of knee pain not seen by a radiologist, helping account for racial disparities in knee pain. But in other cases, algorithms can further disparities, allocating clinical resources in ways that disfavor underserved patients. Dr. Ziad Obermeyer, a professor at UC Berkeley’s School of Public Health, shared how he approaches algorithms in a biased world.
Closing cancer health equity gaps require medical breakthroughs made possible by new funding approaches.