Will AI and Machine Learning Revolutionize Radiation Oncology?
By Mel Kauffman
In October 2020, the Mayo Clinic and Google Health announced a joint initiative that focuses on applying Artificial Intelligence (AI) to radiation therapy planning and medical imaging. Experts from the Mayo Clinic and Google Health will apply AI to medical imaging. The collaboration’s goal is to improve the quality of radiation plans and patient outcomes, and thereby make the delivery of radiation therapy more efficient. This technology will be more accurate in dosage planning and in the proper identification of key organs to protect healthy tissue.
Since the 2000s the use of AI and machine learning in health care has exploded. According to research from UT Southwest, using AI enables cancer patients to start treatment sooner because AI can use complex data and create an optimal treatment plan. AI can also create treatment workflows and determine proper dosage, and it allows for quick recalculations to adjust the treatment plan as needed. UT Southwest uses AI and deep machine-learning to streamline the planning process down to a fraction of a second, which enables doctors to spend more time with patients and decreases the time staff needs to devote to treatment planning.
There are several instances in which AI and machine learning have successfully detected cancer. Breast cancer is one example. AI programs such as MammoScreen aid in early breast cancer detection--their increased sensitivity has reduced the number of false-negative results and led to more accurate readings. AI has the potential to improve breast cancer screening. When used as a second reader, an AI algorithm can also improve lung cancer detection on chest radiographs, according to a recent study published in JAMA Network Open.
An article published in Nature earlier this year discusses how AI can detect subtle patterns that the human eye can miss. A recent article published in The Harvard Gazette discussed some of the risks of AI in medicine.
AI and machine learning in radiation oncology—and in medicine generally—are here to stay. Researchers and clinicians can’t—and likely wouldn’t want to—wind back the clock. Just how they will fit within the U.S. healthcare system is still a work in progress, but the advances should improve patient outcomes.
We want to hear from you!
Do you have experience using AI and has it improved any processes in your department?
Comments