Skip to content

Using AI to measure prostate cancer lesions could help diagnosis and treatment

Prostate cancer is the second most common cancer in men, and nearly 300,000 people are diagnosed with it each year in the US. To develop a consistent method for estimating the size of prostate cancer, which can help doctors To make more accurately informed treatment decisions, researchers at Mass General Brigham trained and validated an AI model based on MRI scans of more than 700 prostate cancer patients. The model was able to identify and delineate the borders of 85% of the most radiologically aggressive prostate lesions.

Tumors with larger volume, as estimated by the AI ​​model, were associated with a higher risk of treatment failure and metastasis, independent of other factors typically used to estimate this risk. Additionally, for patients who received radiation therapy, tumor volume performed better than traditional risk stratification in predicting metastasis. Researchers believe the tool could be used to help doctors understand the aggressiveness of a tumor, inform more personalized treatment plans, and guide radiation therapy. The study is published in the journal. Radiology.

“Al-determined tumor volume has the potential to advance precision medicine for patients with prostate cancer by improving our ability to understand the aggressiveness of a patient’s cancer and therefore recommend the most optimal treatment,” he said. first author David D. Yang, MD, of the Department of Radiation Oncology at Brigham and Women’s Hospital, a founding member of the Mass General Brigham healthcare system.

MRI has improved doctors’ ability to diagnose prostate cancer and is a routine part of diagnosis and treatment. While human doctors can estimate tumor size based on MRI images, these estimates are somewhat subjective and can vary from person to person.

To develop a more consistent method for estimating tumor size, the researchers trained an AI model based on MRI images of prostate cancer tumors from 732 patients undergoing treatment at a single center. They then investigated whether the AI ​​model size estimates were associated with treatment success in the 5 to 10 years after diagnosis.

They demonstrated that the AI ​​model was able to locate and measure around 85% of prostate tumors that had a PI-RADS (Prostate Imaging Reporting and Data System) score 5 within the patient cohort. The score indicates a very high risk of clinically significant prostate cancer. Model size estimates also showed potential as a prognostic marker: larger tumors were associated with a higher risk of prostate cancer coming back, as measured by blood levels of prostate-specific antigen (PSA), or metastasizing. , both in patients treated surgically. or with radiotherapy.

“Measuring AI itself may tell us something additional in terms of patient outcomes,” said senior author Martin King, MD, PhD, of the Brigham’s Department of Radiation Oncology. “For patients, this can really tell them something about what the chances of a cure are and how likely their cancer is to come back or metastasize in the future.”

In addition to helping doctors and patients understand the aggressiveness of their cancer, the AI ​​model could also help guide radiation oncologists by identifying the focal region of the tumor for more targeted treatment. It is also a much faster test compared to methods currently used to predict the aggressiveness of prostate cancer, which typically take two weeks or more to provide results. AI-based tests could mean patients can start treatment sooner.

Cancer research is a fundamental pillar of Mass General Brigham’s care for patients. Research, coupled with the power of system strengths in innovation, education and community engagement, enables Mass General Brigham Cancer to provide integrated cancer care for all, placing health equity at the center of that support. The vision is to provide a comprehensive, integrated, research-based approach to cancer care, helping patients navigate the entire continuum of care, from prevention and early detection to treatment and survivorship.

Looking ahead, the researchers plan to test their model with a larger multi-institutional data set.

“We want to validate our findings, using other institutions and cohorts of patients with different disease characteristics, to ensure that this approach is generalizable to all patients,” Yang said.

Leave a Reply

Your email address will not be published. Required fields are marked *