Explainable AI for breast cancer shows multidisciplinary logic ‘pixel by pixel’
Researchers in Europe have demonstrated an explainable AI technique for administering precision medicine to patients with breast cancer.
The advance represents the first cancer-care pathway offering an automatically combined analysis of morphological, molecular and histological data, according to a study published March 8 in Nature Machine Intelligence.
The team, led by a pathologist at Charité–Universitätsmedizin Berlin and a professor of machine learning at Technical University of Berlin, built their system to supply an easily readable visual aid—heatmaps—to show clinicians the AI’s “thinking” as well as its results.
“Pixel by pixel, these heatmaps show which visual information influenced the AI decision process and to what extent, thus enabling doctors to understand and assess the plausibility of the results of the AI analysis,” the institutions explain in a joint announcement. “This represents a decisive and essential step forward for the future regular use of AI systems in hospitals.”
The team’s next moves include clinically validating the approach in routine tumor diagnostics.
Dave P. has worked in journalism, marketing and public relations for more than 30 years, frequently concentrating on hospitals, healthcare technology and Catholic communications. He has also specialized in fundraising communications, ghostwriting for CEOs of local, national and global charities, nonprofits and foundations.