Medical professionals have created a digital intelligence resource that can anticipate the likelihood of adverse effects in breast cancer patients following treatment.

Every year, 2 million women around the world receive a diagnosis of the most prevalent form of cancer among females in the majority of countries.

In recent years, survival rates have improved due to increased awareness, earlier detection, and a larger variety of treatment choices. However, following treatment, many patients may still suffer from severe side effects.

A group of doctors, experts, and scientists from different countries have created an artificial intelligence (AI) tool that can estimate the probability of a patient encountering complications following surgery and radiotherapy. This technology is currently being tested in the United Kingdom, France, and the Netherlands and has the potential to assist patients in receiving tailored medical treatment.

Dr. Tim Rattay, a consultant breast surgeon and associate professor at the University of Leicester, expressed gratitude that the long-term survival rates for breast cancer have been improving. However, for certain patients, this also means living with the aftermath of their treatment, such as skin changes, scarring, painful swelling in the arm known as lymphoedema, and even potential heart damage from radiation therapy.

We are in the process of creating an AI tool that will provide doctors and patients with information about the potential for chronic arm swelling following surgery and radiotherapy for breast cancer. Our goal is to aid doctors and patients in making informed decisions about radiation treatment and minimize side effects for all patients.

The AI technology was taught to forecast lymphoedema occurrence within three years post-surgery and radiotherapy by analyzing information from 6,361 women with breast cancer. By identifying patients who are more susceptible to arm swelling, healthcare professionals can propose alternative therapies or extra assistance throughout and after treatments.

Dr. Guido Bologna, an associate professor at the University of Applied Sciences and Arts of Western Switzerland in Geneva, and a co-investigator on the project, stated that the most optimal model for predictions utilizes 32 various patient and treatment characteristics. These include factors such as whether the patients received chemotherapy, underwent a sentinel lymph node biopsy in the armpit, and the type of radiotherapy administered.

The AI tool had an 81.6% success rate in accurately predicting lymphoedema, and correctly identified patients who would not develop it 72.9% of the time. The overall accuracy rate for the model was 73.4%.

Rattay mentioned that patients who are identified as having a higher risk of developing arm swelling could potentially benefit from additional support, such as wearing a compression sleeve during treatment. Studies have shown that this measure can help reduce arm swelling in the long run. Furthermore, clinicians can use this information to discuss the potential benefits of lymph node irradiation for patients, especially when the potential benefits may not be clear-cut.

At the European Breast Cancer Conference in Milan, Rattay mentioned a technology that is considered an “explainable AI tool,” meaning that it provides the rationale behind its decision-making process.

According to him, this simplifies the decision-making process for doctors and also allows them to give patients explanations supported by data.

The research group aims to enroll 780 participants in the Pre-Act project clinical trial, with a two-year follow-up period. They are also working on a tool to forecast potential adverse reactions such as skin and heart complications.

The Director of Research, Support, and Influencing at Breast Cancer Now, Dr Simon Vincent, emphasized the need for ways to enhance treatments. This project will examine the potential of using AI to provide tailored care and support for individuals with breast cancer, with the goal of reducing side effects like chronic arm swelling after surgery and radiotherapy.

This study is in its preliminary phases and additional proof is required before determining the potential use of the AI tool in medical environments. We are anticipating the outcomes of the trial.

At the conference, other advancements were reported by Italian researchers. They discovered that utilizing a combination of positron emission tomography-magnetic resonance imaging (PET-MRI) scans allowed medical professionals to identify the spread of a breast cancer patient’s tumor. This finding allowed for the possibility of alternative treatments, such as chemotherapy or a different surgical approach.

Dutch researchers reported that young women with breast cancer who received a small dose of radiation at the site of their tumor in addition to whole breast radiation had no local recurrence after a decade.

Source: theguardian.com