AI system accurately detects key findings in chest X-rays of pneumonia patients within 10 seconds

From 20 minutes or more to 10 seconds. Researchers from Intermountain Healthcare and Stanford University say 10 seconds is about how quickly a new system they studied that utilizes artificial intelligence took to accurately identify key findings in chest X-rays of patients in the emergency department suspected of having pneumonia.

Read more

First targeted therapy for cholangiocarcinoma shows clinical benefit in phase III trial

New data have shown for the first time that targeted therapy can improve the outcome of patients diagnosed with advanced cholangiocarcinoma. Cholangiocarcinoma is a subtype of bile duct cancer with aggressive behavior and poor prognosis. Despite the low incidence, most patients die from the disease and therefore new effective therapies are urgently needed.

Read more

Wrist-worn step trackers accurate in predicting patient health outcomes

Researchers have found that steps measured through a step tracker worn on the wrist can be used to estimate exercise capacity and determine the health status of patients, rather than the standardized 6-minute walk distance test, which is usually conducted in a clinical setting.

Read more

Simulations characterize turbulence caused by common connection for dialysis

The complex interplay among the arteriovenous grafts, the vessels they connect, and the blood they transport has been difficult to simulate, but one new method provides a way. Researchers report simulations that reconstructed the fluid dynamics affected by the insertion of an AVG. They used a model that considered the ability of AVG tubes and blood vessels to deform and found much of the disrupted flow could be mitigated by this flexibility.

Read more

For hospitalized patients with fungal infections, specialists save lives

Fungal bloodstream infections are responsible for the deaths of more than 10,000 people every year. New research shows that the death rate can be reduced by 20% if infectious disease specialists oversee care of such patients.

Read more

First systematic review and meta-analysis suggests artificial intelligence may be as effective as health professionals at diagnosing disease

Artificial intelligence (AI) appears to detect diseases from medical imaging with similar levels of accuracy as health-care professionals, according to the first systematic review and meta-analysis, synthesizing the available evidence from the scientific literature.

Read more

Discovery of how colorectal cancer drug works will help more patients

Some colorectal cancer patients with a certain gene mutation benefit from a chemotherapy drug called cetuximab, although the mechanism of how this drug worked was unknown. Scientists have combined computational biology with experimental investigations to discover, for the first time, the mechanism for why these patients respond to cetuximab, which will help doctors identify more effective, targeted treatment plans for people diagnosed with colorectal cancer.

Read more

Large-scale enhanced recovery program improves outcomes for bariatric surgery patients

A large-scale implementation of a protocol to improve recovery of patients after weight-loss operations was found to reduce rates of extended hospitalization by almost half at 36 participating accredited bariatric surgery centers nationwide, according to a new study.

Read more

Machine learning could offer faster, more precise results for cardiac MRI scans

Cardiac MRI scans can be read by AI (artificial intelligence) 186 times faster than humans, with comparable precision to experts. Because the greatest source of measurement errors are human factors, AI has the potential to improve future clinical decision making.

Read more