Disease Prediction, Machine Learning, and Healthcare - DZone AI


The 21st century has been an era of data-driven decisions. It is said that the segments or industries that generate more data will grow faster and the organizations that this data to make important decisions will be ahead of the curve.

UPMC cuts hospital readmission rates with ML algorithm


The University of Pittsburgh Medical Center's clinical analytics team has leveraged machine learning to develop an algorithm that rates hospital patients for their risks of being readmitted. Specifically, the ML algorithm identifies patients at highest risk of re-hospitalization within seven and 30 days of discharge. To date, re-hospitalizations have been reduced by about 50 percent. "Right now, the main area of focus is on seven days," says Oscar Marroquin, MD, chief clinical analytics officer, UPMC Health Services Division. "The models to predict seven and 30 days are almost identical. The only thing that changes is the statistical weights of each one of the co-variants, which are a little bit different. They are statistically evident but not clinically important."

5 Big Data Trends in Healthcare for 2017


This blog post is an excerpt from the MapR Guide to Big Data in Healthcare. To read more, you can download it here.

Machine learning, EHR data helping to combat hospital infections


Hospitals continue to grapple with clostridium difficile infections, caused by bacteria that are resistant to many common antibiotics and that kill about 30,000 Americans each year. However, machine learning can help predict patient risk in developing C. difficile much earlier than current methods of diagnosis. Using electronic health records for nearly 257,000 patients, researchers from Massachusetts General Hospital, MIT and Michigan Medicine have built hospital-tailored machine learning models that they contend are an improvement over a "one-size-fits-all" approach that ignores important factors specific to medical facilities. "When data are simply pooled into a one-size-fits-all model, institutional differences in patient populations, hospital layouts, testing and treatment protocols, or even in the way staff interact with the EHR can lead to differences in the underlying data distributions and ultimately to poor performance of such a model," says Jenna Wiens, assistant professor of computer science and engineering at U-M. "To mitigate these issues, we take a hospital-specific approach, training a model tailored to each institution." De-identified EHR data from 191,014 adult admissions to Michigan Medicine and 65,718 adult admissions to MGH were analyzed using separate machine learning algorithms tailored to each healthcare institution with different types of variables.

Preventing deadly hospital infections with machine learning


Nearly 30,000 Americans die each year from an aggressive, gut-infecting bacteria called Clostridium difficile (C. New machine learning models tailored to individual hospitals could give them a much earlier prediction of which patients are most likely to develop C. difficile, potentially helping them stave off infection before it starts. The models are detailed in a paper published today in Infection Control and Hospital Epidemiology. Developed by researchers at the University of Michigan, Massachusetts General Hospital and MIT, the models can predict a patient's risk of developing C. difficile much earlier than it would be diagnosed with current methods. Preliminary data from their study, was recently published in Infection Control and Hospital Epidemiology.