Artificial intelligence could launch radiation therapy for cancer patients

19 July 2022, 21:01 | Technologies
photo e-news.com.ua
Text Size:

Artificial intelligence can help cancer patients start radiation therapy earlier and thereby reduce the chance of cancer spreading by instantly translating complex clinical data into an optimal attack plan.

Patients typically have to wait a few days to a week to start therapy while doctors manually develop treatment plans.. But new research from UT Southwestern shows how advanced deep learning models have made this process a fraction of a second..

“Some of these patients need immediate radiation therapy, but doctors often have to ask them to go home and wait,” says Steve Jiang, Ph.D., director of the Medical Artificial Intelligence and Automation (MAIA) UT Southwestern Lab.. “Achieving optimal treatment plans in near real time is important and is part of our broader mission to use AI to improve every aspect of cancer care.”.

Radiation therapy is a common form of cancer treatment that uses beams of high radiation to destroy cancer cells and shrink tumors.. Previous research shows that delaying this therapy by even a week can increase the chance of some cancers recurring or spreading by 12 to 14 percent..

These statistics prompted Jiang's team to explore ways to use AI to improve several aspects of radiotherapy, from initial dosing plans required before starting treatment to dose recalculations that occur as the plan progresses..

Jiang says developing a complex treatment plan can be a time-consuming and tedious process that involves careful analysis of patient imaging data and multiple steps of feedback from the medical team..

New MAIA Dose Prediction Lab study published in Medical Physics demonstrates AI's ability to generate optimal treatment plans within five hundredths of a second of receiving clinical data for patients.

Scientists trained AI to instantly generate 3D images of how best to distribute radiation therapy to cancer patients. Technology could allow patients to start treatment earlier, thereby reducing the chance of cancer spreading.

The researchers achieved this by feeding data from 70 prostate cancer patients into four deep learning models.. Through repetition, AI has learned to develop three-dimensional images of how best to distribute radiation in each patient. Each model accurately predicted the treatment plans developed by the medical team.

Study builds on other MAIA work published in 2019 that focuses on developing treatment plans for lung and head and neck cancer.

“Our AI can cut out a lot of what goes on between the doctor and the dosage specialist,” Jiang says.. "

Jiang's second new study, also published in Medical Physics, shows how AI can quickly and accurately recalculate doses before each radiation session, taking into account how the patient's anatomy may have changed since the last treatment.. The usual accurate recount sometimes requires patients to wait 10 minutes or more, in addition to the time required for anatomical imaging before each session..

Jiang's researchers developed an artificial intelligence algorithm that combined two traditional models that were used to calculate dose: a simple, fast model that lacked accuracy, and a complex model that was accurate but took a much longer time, often around half an hour..



Newly developed AI assessed differences between models - based on data from 70 prostate cancer patients - and learned how to use speed and accuracy to generate calculations within one second.

UT Southwestern Plans to Leverage New AI Capabilities in Clinical Care with Patient Interface Implementation. Meanwhile, MAIA Lab is developing deep learning tools for several other purposes, including improving medical imaging and imaging, automated medical procedures, and improving disease diagnosis and predictive outcomes..

medical-heal. en.

Based on materials: med-heal.ru



Add a comment
:D :lol: :-) ;-) 8) :-| :-* :oops: :sad: :cry: :o :-? :-x :eek: :zzz :P :roll: :sigh:
 Enter the correct answer