6 hospital applications for machine learning: algorithms to predict patient violence, HIV risk & more

Here are six recent studies and product launches that incorporate machine learning into clinical settings:

1. A group of five machine learning algorithms identified whether a patient was a suitable candidate for corneal refractive surgery with 93.4 percent accuracy, a level equal to that of expert ophthalmologists.

2. Cleveland Clinic researchers developed a deep learning model that analyzes lung cancer patients' EHRs and imaging data to determine the most effective radiation dosage for their treatment.

3. A predictive machine learning tool was trained to analyze the medical records of millions of HIV-negative patients at Kaiser Permanente Northern California to measure their risk of receiving an HIV diagnosis within three years.

4. Automated analysis of clinical notes in the EHRs of inpatients at psychiatric treatment centers showed "good predictive validity" in flagging patients' risk of violence in the first four weeks of treatment.

5. A deep learning algorithm developed by scientists at the Duke University School of Medicine outperformed several practicing radiologists in determining whether a biopsy was necessary for a thyroid nodule.

6. Geisinger's Steele Institute for Health Innovation partnered with Medial EarlySign to develop and deploy several machine learning solutions to identify patients at risk of high-burden diseases.

More articles about AI:
Viewpoint: The AI revolution will leave us 'struggling to understand'
Michigan Medicine, Atomwise launch research collaboration for AI-driven drug discovery
Machine learning predicts individual malaria outcomes, disease progression

© Copyright ASC COMMUNICATIONS 2019. Interested in LINKING to or REPRINTING this content? View our policies by clicking here.

 

Top 40 Articles from the Past 6 Months