Al taps CT datasets, EHRs to personalize radiotherapy dosage

A deep learning framework leveraging computerized tomography scans and electronic health records is able to individualize the dose of radiation therapy for treating cancer patients.

That’s the finding of a new study—supported, in part, by Siemens Medical Solutions USA—and published in the July issue of The Lancet Digital Health.

“The image-based deep learning framework proposed herein is the first opportunity to use medical images to individualize radiotherapy dose,” state the study’s authors. “The most important message in our study is that predictive features can be learned from CT images and can contribute to the individualization of radiation dose.”

Also See: MD Anderson leverages deep learning for radiation targeting tool

The study included 849 lung cancer patients treated with high-dose radiation who had “evaluable” electronic health record data and CT images.

Siemens developed the Deep Profiler, which analyzed the scans to create an “image signature” indicative of radiation outcome. The image signature from Deep Profiler was then combined with EHR data to predict sensitivity or resistance and generate a personalized radiation dose value.

“We input pre-therapy lung CT images into Deep Profiler, a multi-task deep neural network that has radiomics incorporated into the training process, to generate an image fingerprint that predicts time-to-event treatment outcomes and approximates classical radiomic features,” according to the authors. “We validated our findings with the independent study cohort.”

“This framework will help physicians develop data-driven, personalized dosage schedules that can maximize the likelihood of treatment success and mitigate radiation side effects for patients,” says lead author Mohamed Abazeed, MD, a radiation oncologist at Cleveland Clinic’s Taussig Cancer Institute and a researcher at the Lerner Research Institute.

“The development and validation of this framework is exciting because not only is it the first to utilize medical images to inform radiation dose prescriptions, but it also has the potential to directly impact patient care,” Abazeed adds. “The framework has a low implementation barrier to inform radiotherapy-based clinical trial design and ultimately can be used to deliver radiation therapy tailored to individual patients in everyday clinical practices.”