Artificial intelligence refines breast cancer diagnosis
Valencian researchers have improved the precision of mammograms thanks to an algorithm in radiology tests
Medicine and technology continue to strengthen ties to improve the diagnosis and prevention of diseases. A team of researchers from the University of Valencia and the Polytechnic of Valencia has managed to program an algorithm based on artificial intelligence to help radiologists.
This work has shown that artificial intelligence, coupled with the knowledge of medical professionals improves the accuracy of cancer detection by mammography.
The research is being carried out with Artemisa, the new computing platform for Artificial Intelligence of the Corpuscular Physics Institute, funded by the European Union and the Generalitat Valenciana within the FEDER operational program of the Valencian Community 2014-2020 for the acquisition of infrastructure and R + D + I equipment.
Mammograms are the most widely used diagnostic technique for the early detection of breast cancer. Although this screening tool is usually adequate, mammograms must be evaluated and interpreted by a radiologist.
Therefore, the practitioner uses his visual perception to identify signs of cancer; An estimated 10% of “false positives” are among the 40 million women who undergo scheduled mammograms each year in the United States.
“An effective Artificial Intelligence algorithm that can increase the radiologist’s ability to reduce the repetition of unnecessary tests while detecting clinically significant cancers would help increase the value of mammography detection, effectively improving the harm-benefit ratio,” explains Christoph Lee of the Washington School of Medicine.
Google on the run
America’s tech giants have also gone to work searching for and perfecting new cancer tests. The magazine Nature published the study in early January where it revealed that Google had developed an algorithm that reduces false positives by 3.5%, are those people who are diagnosed with cancer when in fact they do not have it, and by 8, 1% false negatives, those cases in which cancer is not detected, but it is.