Scientists have developed a deep learning model that predicts a 10-year risk of death from heart disease using a single chest image. Research results submitted at the annual meeting of the Radiological Society of North America.
A standard cardiovascular risk assessment is performed for prescribing prophylactic statins. Classical analysis provides a statistical method that takes into account many factors such as demographics, biomarkers, pathologies, and bad habits. According to scientists, this data is not always available, which is why the accuracy of the prediction is significantly reduced.
To improve the prediction of heart disease risk, scientists have developed a deep learning model. They trained her on 147,497 chest X-rays from 40,643 participants in a multicenter randomized screening trial for prostate, lung, colon and ovarian cancer.
They then tested the algorithm using other data from 11,430 patients who were previously potentially eligible for statin therapy. Of these, 1096 had suffered a cardiovascular disease within the last 10 years after the radiograph. The rest of the patients had no such problems.
As a result, the AI model predicts disease risk with high accuracy. In patients with cardiovascular problems, the calculated level was significantly higher.
According to study lead author Jacob Weiss, the algorithm will help identify people who would benefit from starting statin treatment.
“Based on a single existing chest x-ray image, our deep learning model predicts future serious adverse cardiovascular events with similar performance and added value to the established clinical standard,” he stated.
Weiss added that more research is needed to test the model, which could serve as a decision support tool for clinicians in the future.
Recall that in September, scientists created an AI that detects diseases by radiographs.
Subscribe to Cryplogger news in Telegram: Cryplogger AI – all the news from the world of AI!
Found a mistake in the text? Select it and press CTRL+ENTER
Scientists have developed a deep learning model that predicts a 10-year risk of death from heart disease using a single chest image. Research results submitted at the annual meeting of the Radiological Society of North America.
A standard cardiovascular risk assessment is performed for prescribing prophylactic statins. Classical analysis provides a statistical method that takes into account many factors such as demographics, biomarkers, pathologies, and bad habits. According to scientists, this data is not always available, which is why the accuracy of the prediction is significantly reduced.
To improve the prediction of heart disease risk, scientists have developed a deep learning model. They trained her on 147,497 chest X-rays from 40,643 participants in a multicenter randomized screening trial for prostate, lung, colon and ovarian cancer.
They then tested the algorithm using other data from 11,430 patients who were previously potentially eligible for statin therapy. Of these, 1096 had suffered a cardiovascular disease within the last 10 years after the radiograph. The rest of the patients had no such problems.
As a result, the AI model predicts disease risk with high accuracy. In patients with cardiovascular problems, the calculated level was significantly higher.
According to study lead author Jacob Weiss, the algorithm will help identify people who would benefit from starting statin treatment.
“Based on a single existing chest x-ray image, our deep learning model predicts future serious adverse cardiovascular events with similar performance and added value to the established clinical standard,” he stated.
Weiss added that more research is needed to test the model, which could serve as a decision support tool for clinicians in the future.
Recall that in September, scientists created an AI that detects diseases by radiographs.
Subscribe to Cryplogger news in Telegram: Cryplogger AI – all the news from the world of AI!
Found a mistake in the text? Select it and press CTRL+ENTER
