Sudhakar Jha
... Read More
By clicking “Accept All Cookies”, you agree to the storing of cookies on your device to enhance site navigation, analyze site usage, and assist in our marketing efforts. Cookie Policy.
Written By: Sudhakar Jha | Published : August 14, 2018 1:15 PM IST
Chemotherapy and radiations are the available treatments for cancer patients and though they are accurate to many extents in treating the diseases, it is the severity of them that takes a toll on the body way long time after it's over. And that led the Massachusetts Institute of Technology to form an AI platform that could help in the treatment of brain cancer.
The software that has been created by the researchers takes important metrics into account that are based on the clinical records of other patients before recommending a regimen designed to shrink brain tumours which lessens the side effects.
For creating the software, the MIT team started with a simple Reinforced Learning (RL) system similar to the one Google's DeepMind used to create an unbeatable Go-playing AI. The programme consisted of "agents" that recommend actions which get you closer to the desired outcome. If the recommendation is good, the agent is rewarded with greater weighting. If not, the agent is penalised and has to come up with newer recommendations accordingly. With every feedback and penalisation, the AI optimises itself at solving a specific problem with higher accuracy, and in this case creating treatments for glioblastoma, a form of brain cancer.
The researchers used the model software on 50 glioblastoma patients who had sought treatment in the past. The AI conducted 20,000 trial-and-error tests to optimise itself for the best results. After that, the team ran 50 new patient profiles through the software to see what the machine's recommendations. The AI came out with the conclusion that the patients can get the same or better treatment with much less chemotherapy and radiation.
Doctors refer to a number of standard treatment regimens for glioblastoma, which has a survival time of five years. And during the treatment, the goal is to poison the tumour cells faster than the non-cancerous cells, but the side effects are devastating. The problem with the traditional treatment methods are that they don't take into account differences in tumour size, medical histories, genetic profiles and biomarkers. The system developed by MIT does that, resulting in lowered dosages and in some case even skipping doses altogether.
The new software has only been tested in simulations so far, but the researchers believe that the science has been sound. Medical councils world over have been looking for ways to enhance medical treatment with technology, and this could provide a more precise way of modifying patient care.
Image source: Shutterstock
Disclaimer: The content on TheHealthSite.com is only for informational purposes. It is not at all professional medical advice. Always consult your doctor or a healthcare specialist for any questions regarding your health or a medical condition.