Computer analysis of organs predicts patient mortality

Using computers to analyze CT images of patients' organs, researchers were able to predict their five-year mortality rate with an accuracy rate of up to 70%. This is the latest research result published in the journal Scientific Reports. Let's follow the medical electronics editor to learn about the relevant content.

Computer analysis of organs predicts patient mortality

Study author Dr. Luke Oakden Rayner of the School of Public Health at the University of Adelaide in Australia and colleagues believe their findings can advance the field of precision medicine.

The National Institutes of Health (NIH) defines precision medicine as "an emerging approach to disease prevention and treatment that takes into account each person's individual differences in genes, environment, and lifestyle."

As the study authors note, precision medicine relies on the discovery of biomarkers, which are precise indicators of disease risk, response to treatment, or disease prognosis. They suggest that radiology plays an important role in this area.

In their study, Oakden Rayner, PhD, and colleagues set out to explore whether they could teach a computer to "learn" information from computed tomography (CT) scans to predict a patient's 5-year mortality.

First, the team collected 15,000 CT images of seven different tissues—including heart and lung tissue—from patients aged 60 and older. Using logistic regression techniques, the researchers identified a number of image features that were associated with 5-year mortality.

The team then combined the data with a technique called "deep learning," a method by which computers can "learn how to understand and analyze images," Dr. Oakden Rayner explained.

He added: “Automated systems are not focused on diagnosing disease, but rather predicting clinical outcomes in a way that doctors are not trained to do: by combining large amounts of data and detecting subtle patterns.”

Next, the researchers used computers to analyze chest CT images of 48 patients over the age of 60. They found that the computer predicted their five-year mortality rate with 69% accuracy, compared with mortality predictions made by health care professionals.

In the meantime, the researchers say their study provides evidence that CT images and computer learning could lead to major advances in precision medicine.

The above is an introduction to medical electronics - computer analysis of organs to predict patient mortality. If you want to know more related information, please pay more attention to eeworld. eeworld Electronic Engineering will provide you with more complete, detailed and updated information.