Blog

Algorithm mines infants’ health data for autism risk: An algorithm that analyzes health records for data on infant health care visits for brain, sleeping and stomach issues is helping physicians at Duke University identify babies at risk for autism spectrum disorder in their first month of life. The tool, combined with conversations with parents and doctors, identified nearly half the babies with ASD within a few days of birth, which may give families early access to interventions and support

Parkinson’s gloves designed to reduce tremors: Gloves that were designed to reduce the hand tremors of patients with Parkinson’s disease have been launched by the Chulalongkorn University Communication Center. The gloves automatically deliver electrical stimulation to the hand muscles without the need for brain surgery or increased medication.

Researchers develop risk-scoring tool for Diabetes: A new assessment model that combines basic health data and a methylation analysis identified more people at high risk for type 2 Diabetes than standard methylation-free assessments. The new assessment model might someday be used as a risk-scoring tool for conditions other than type 2 Diabetes, researchers wrote in Nature Aging.

Machine Learning (ML) could help guide depression treatment: A machine learning algorithm accurately predicted antidepressant treatment outcomes based on EHRs, suggesting the algorithm could be developed as a “clinical decision-making tool for personalized management of depression,” according to a study published in Psychiatric Research and Clinical Practice. Researchers tested various models and found the gradient boosting decision tree algorithm produced the most accurate results.

AI products must be made safe before release per Biden: President Joe Biden told the President’s Council of Advisors on Science and Technology that artificial intelligence shows great promise for addressing big problems such as disease and climate change; however, AI also poses risks to society, the economy and national security that must be addressed. Biden said technology companies are responsible for ensuring their products are safe and called on Congress to pass legislation limiting collection of personal data, prioritizing health and safety in product development and banning advertising that targets children.

Happiest Baby bassinet to reduce SIDS risk cleared: The FDA has given de novo clearance to Happiest Baby for its Snoo Smart Sleeper, which is designed to keep babies from sleeping on their stomach. The motorized bassinet aims to aid in reducing cases of sudden infant death syndrome.

Reliability, biocompatibility and other standard medical device safety concerns still apply. However, a few areas stand out in robotic-assisted surgery, said Medical Microinstruments CEO Mark Toland, who led Corindus Vascular Robotics until helping sell it to Siemens for $1.1 billion.

Bill would require Medicare to cover breakthrough devices: Rep. Suzan DelBene, D-Wash., and Rep. Brad Wenstrup, R-Ohio, introduced legislation to require Medicare to cover medical devices that received breakthrough designation from the FDA for four (4) years. The bill would also require the Centers for Medicare and Medicaid Services (CMS) to make a permanent coverage determination during the four-year period.

Technology could help predict miscarriage risk: Researchers in the Netherlands are combining 3D ultrasound imaging with virtual reality technology to determine if a developing embryo is maturing normally and help predict miscarriage risk, and they noted that the longer it takes to develop, the greater the risk of miscarriage appears to be. The technique creates a hologram of the embryo and offers clinicians more detailed information than they would get from simply measuring its size at various pregnancy stages.

AI quickly detects genetic mutations in brain tumor samples: A machine learning tool called DeepGlioma demonstrated a 93.2% average accuracy in molecular genetic classification of tumors in patients with diffuse gliomas within 90 seconds, according to a study in Nature Medicine. “Rapid methods for molecular classification hold great promise for rethinking clinical trial design and bringing new therapies to patients,” said Study Senior Author Daniel Orringer.