Artificial intelligence to improve treatment of cancer patients with heart attacks

(CONNECT) Cancer patients with a heart attack are exposed to a significantly higher mortality risk. In order to develop personalized therapies, researchers from the Center for Molecular Cardiology at the University of Zurich, together with international colleagues, have developed a tool supported by artificial intelligence (AI) that enables a more accurate prognosis based on symptoms. According to a press release, the researchers examined a study of more than 1 million heart attack patients from England, Sweden and Switzerland, including over 47,000 people with cancer. The study showed that the heart attack patients had a remarkably poor prognosis: one third of those examined died within six months.

The newly developed Onco-Acs tool uses AI to combine cancer-related and cardiovascular factors from the extensive data set. The doctors then want to use the findings to predict both mortality and events such as severe bleeding or ischemic incidents. "Depending on the characteristics of the tumor, cancer patients may have an increased risk of bleeding, an increased tendency to arterial blood clots, or both - each of which requires different treatment," Florian A. Wenzl from the Center for Molecular Cardiology and the National Health Service England and lead author of the study, is quoted in the press release. "For targeted treatment, it is therefore important that we can better assess the individual risk profile in the clinic."

In the future, Onco-Acs is intended to provide treating physicians with information for developing a personalized therapy that weighs up the benefits and risks. The current study has been published in the journal The Lancet. ce/ww