Effective help for planning a more appropriate treatment for breast cancer, thanks to artificial intelligence: this is the focus of a study developed by the research group of the biostatistics and bioinformatics laboratory of the Tumor Institute of Bari, published in the international journal ‘ Computers in biology and medicine’, published by Elsevier.
At the center of the study, first designed by researcher Maria Comes, is an algorithm that early identifies the patients most likely to respond to neoadjuvant chemotherapy. This therapy is usually prescribed before surgery to reduce the size of the tumor, but the desired effects are not always achieved.
And here AI intervenes, capable of interpreting the data from the magnetic resonance images performed during the diagnosis phase and returning personalized feedback to the doctor on the effectiveness or otherwise of the treatment.
The model studied by the Bari oncology researchers, currently tested on 120 patients, fits into the promising research line of precision oncology and relies on a visual explanation of the decision-making process. This makes the interaction between the doctor and the application even simpler, with the advantage of being able to define the therapy for the individual patient, and therefore avoiding, if not necessary, toxic and invasive treatments.
The Apulian laboratory has already published publications on the possibility of extracting this kind of predictions from radiological examinations for different types of cancer: breast, lung, head and neck.
