See all results

Education City Road Closures

For further information about road closures and diversions in Education City during the FIFA Arab Cup™, click here

Story | Education
4 February 2021

World Cancer Day: QF expert developing AI tech to reduce side effects of radiotherapy

Share
World Cancer Day: QF expert developing AI tech to reduce side effects of radiotherapy

Breast cancer is the most common form of cancer in Qatar.

Image source: Guschenkova, via Shutterstock

Dr. Othmane Bouhali, Director of Research Computing and Professor at TAMU-Q, says personalized radiation dose for breast cancer can reduce the risk of cardiovascular disease.

Cancer treatment through radiotherapy has significantly increased the number of cancer survivors, but it has also increased the chances of these patients going on to develop cardiovascular disease – particularly in the case of breast cancer.

“Global statistics say in the past 20-25 years, a significant fraction of women who were treated for breast cancer using radiotherapy have gone on to develop cardiovascular disease (CVD) within 5-10 years of receiving their first dose,” said Dr. Othmane Bouhali, Director of Research Computing and Research Professor at Texas A&M University at Qatar (TAMU-Q), a Qatar Foundation (QF) partner university.

With breast cancer being the most common cancer in Qatar, this is a serious concern. However, Dr. Bouhali said he believes this side effect can be significantly reduced by optimizing the radiotherapy dosage.

World Cancer Day - QF expert developing AI tech to reduce side effects of radiotherapy - QF - 02

Dr. Othmane Bouhali of Texas A&M University at Qatar.

Building on his years of experience in medical physics and supercomputing, he is attempting to achieve this with the help of artificial intelligence (AI). “The increased risk of CVD is not a direct consequence of radiotherapy but is a product of the dosage,” Dr. Bouhali said. “If we can optimize the dose quantity to be the minimum amount needed to be effective, and also angle the X-rays such that they hit the tumor cells while staying as far as possible from the heart, I believe the risk of CVD can be reduced.

To be able to calculate the precise dosage, much more sophisticated software needs to be used

Dr. Othmane Bouhali

“At present, radiotherapy dose is calculated using a classical software that is based on approximations as it’s time-efficient. It is reliable but not precise and in most cases, it gets the job done. To be able to calculate the precise dosage, much more sophisticated software needs to be used.”

Monte Carlo simulation is one such software. However, it takes hours to compute the right dosage, making it impractical to be used in a hospital setting. And Dr. Bouhali’s group is attempting to combine the Monte Carlo simulation with a powerful machine learning algorithm that his team has developed. Once developed, it could considerably reduce the time needed to determine the precise dose that would be suitable for a particular patient.

“We have already developed the algorithm; we are now in the process of ‘teaching’ it using existing patient data.”

AI has so much potential, especially in the medical sector, but it is crucial to understand it can’t and won’t progress without access to data

Dr. Othmane Bouhali

Commenting on the challenges faced, Dr. Bouhali said: “Access to data remains AI’s biggest challenge, not just in Qatar but all over the world. AI has so much potential, especially in the medical sector, but it is crucial to understand it can’t and won’t progress without access to data. Unless researchers are given access to data, we will not be able to deliver the promises of AI.”

In another project in collaboration with a local hospital, Dr. Bouhali’s team is looking at ways to identify which stage a patient’s cancer is in by looking at tumor images and processing them using artificial intelligence. At present, doctors determine a cancer's grade by performing a biopsy, an invasive procedure in which doctors remove a piece of the affected tissue and analyze it to determine how much a cancer has spread.

AI combined with medical physics has the potential to bring drastic changes in medicine

Dr. Othmane Bouhali

“Our project aims to bypass this invasive method and instead use imaging data and artificial intelligence,” Dr. Bouhali said. “The algorithm has been developed, we are now in the process of teaching it how to recognize the cancer grade by feeding it images of previous patients and details of their cancer. When enough information has been fed into it, eventually the algorithm will be able to differentiate between grades of cancer by analyzing their images and picking up on differences in pixilation, resolution, contrast and several other things.”

Working closely with local and international hospitals, Dr. Bouhali said he hopes his work will pave the way for data-driven, personalized dosage that will maximize the probability of treatment success and minimize radiation side effects for patients.

World Cancer Day - QF expert developing AI tech to reduce side effects of radiotherapy - QF - 03

Dr. Bouhali says access to data remains the biggest obstacle to AI’s role in advancing medicine

“Medical physics has not received the attention or the funding it deserves in the past, fortunately, that’s changing now. Thanks to the funding of Qatar National Research Fund, this area of research is now taking off in Qatar.

“In the past five years, the United States Food and Drug Administration has approved eight new radiotracers. This has never happened before. I think AI combined with medical physics has the potential to bring drastic changes in medicine, particularly in fields like oncology; and I am very excited to be part of the team that will hopefully deliver these changes in Qatar.”

Related Stories