A team of researchers led by the Cardiovascular Imaging Research Center
Atherosclerotic risk assessmentcardiovascular diseases is carried out for the appointment of preventive statins. Traditionally, this has been done using a statistical model that takes into account many variables, including age, gender, race, systolic blood pressure, hypertension treatment, smoking, type 2 diabetes, and blood tests.
To improve forecasting efficiency and,Accordingly, the researchers trained a neural network to prescribe preventive treatment only to those who really need it. They used data from 147,497 chest radiographs from 40,643 participants in a multicenter randomized trial of screening for prostate, lung, colon, and ovarian cancer.
The researchers tested the model using a secondan independent cohort of 11,430 outpatients who underwent routine ambulatory chest radiography and were potentially eligible for statin therapy. Of these patients, 1,096 had had cardiovascular disease within 10 years of the radiograph, and the rest had no such problems.
The results of the study showed that the levelThe risk predicted by the neural network is significantly correlated with the actual data on the disease: in those patients who encountered problems, the calculated risk level was significantly higher.
Read more:
The creator of the space elevator: "Transit into space will appear within 10 years"
Mummies with golden tongues were found in Egypt. They were needed to communicate with the gods
Look at the terrifying woman generated by the neural network