AI uses pre-op mammograms to predict post-op cancer recurrence
Artificial intelligence (AI) tools may help predict the risk of cancer recurrence after patients have been treated for ductal carcinoma in situ (DCIS).
While survival rates for DCIS are high, delayed detection can increase the likelihood of cancer recurrence down the line. This can occur in another DCIS diagnosis but also can take the form of something more invasive. As such, it is important to be able to understand patients’ risks following their initial diagnosis.
Experts believe AI can help. A new paper in the American Journal of Roentgenology describes the utility of a commercially available AI tool capable of quantifying patients’ risk of recurrence after DCIS using pre-operative mammograms.
Researchers conducted a review of whether the AI system’s scores were correlated with second cancer diagnoses for women treated for DCIS. The women had undergone surgery for DCIS between 2012 and 2017 and had one or more years of postoperative follow-up for the team to analyze. Patients’ medical records were reviewed to identify second breast cancers, while the AI’s scores were dichotomized using the Youden index for second breast cancer prediction.
Of the more than 1,700 women included in the study, post-operative recurrence was noted in 28; seven developed post-mastectomy ipsilateral recurrence, while 25 developed contralateral breast cancer. Post-operative ipsilateral recurrence was found to be independently associated with the AI’s dichotomized threshold score of ≥73.5% at 5 and 10 years.
“AI scores, readily obtained noninvasively on preoperative mammography, may help inform DCIS treatment and surveillance strategies,” Jung Min Chang, MD, PhD, with the department of radiology at Seoul NaConal University College of Medicine, and colleagues suggested.
The team noted that the tool performed as well as or better than other clinical risk models, suggesting that it could have clinical utility in informing post-operative care decisions.
Read more about the findings here.
In addition to her background in journalism, Hannah also has patient-facing experience in clinical settings, having spent more than 12 years working as a registered rad tech. She began covering the medical imaging industry for Innovate Healthcare in 2021.