We may now be closer than ever to effective personalized cancer treatment, with the discovery of a viable method of determining the best drug dose for each patient. Researchers at the University College, London, have developed an AI model capable of replicating tumors and simulating the possible response by the tumors to treatments.
With this discovery, it is now possible for doctors to ascertain the best medication and drug dose that will help a patient defeat the disease, with minimized side effects and treatment resistance risks. And it is possible to have the model on the NHS in the next ten years.
While discussing the discovery, the joint lead author of the research, Dr. Simon Walker-Samuel, said, “It advances the move towards truly personalized medicine, with the potential aim that, one day, clinicians might be able to predetermine the most effective therapeutic plan for each patient’s unique tumor makeup.”
Image shows how drugs tagged with fluorescent molecules move through the bloodstream and cells of a human bowel cancer tumour. Using AI computer models, the scientists then predicted how well the growths would respond to these medications
For every cancer, there is a unique tumor, usually with various growths that differ considerably in cell numbers and blood flow. What determines the extent of absorption and movement of drugs in the bloodstream is the tumor structure, and it ultimately decides if the tumor growth will respond to the drugs or not.
As a means of testing the model, cancer tissues of the human bowel were implanted under the skin of mice whose immune systems have been weakened. After about 10 to 14 days later, drugs tagged with fluorescent molecules were administered into the tumors, after which the tumors were removed and made transparent using simple chemical reactions. Thus, the researchers were able to monitor how the various administered drugs fared in the blood cells and vessels.
Then the AI mathematical model, named REANIMATE by the scientists, was used in predicting the probable responses the growths would show to the various medications administered. “The new framework has a vast potential impact in helping to develop new cancer drugs and potentially providing a cost-effective way to test their efficacy before going to human trials,” Dr. Walker-Samuel explained.
Also speaking on the study, which was published in the journal Nature Biomedical Engineering, Co-lead author Dr. Rebecca Shipley also added: ‘This is a novel approach that provides an entirely new framework for therapy prediction in tumors, and we are now developing ways of applying it to images taken from patient biopsies.”
In the above tumour, blood vessels are shown in grey. The coloured regions show where blood was delivered, which contained anti-cancer drugs. The red areas had the greatest amounts of blood, the green and yellow places had medium amounts, and the blue had the lowest
In the course of the next few years, the scientists are seeking to get similar drug-response predictions using smaller tumor tissue samples from biopsies. While commenting on the project, Geoff Parker, a Professor of Biomedical Imaging at the Manchester University, who was not involved in the research, told iNews: “This work is a tour de force that provides a significant step forward in our understanding of key areas of cancer diagnosis and treatment that could lead to more efficient and effective drug development.”
With the increasing rate of cancer prevalence in the US, there have been various investigations into this disturbing subject matter. Just last August, there was the research by the National University of Singapore, which suggested that AI halted the spread of prostate cancer in a patient already having an advanced form of the disease. According to the study, a patient with tumors spreading outside his prostate cancer was treated using a blend of an approved cancer medication and an experimental drug. Using CURATE AI – an AI Technology, a continuous assessment of the patient’s response to the treatment was carried out, while the doses of the medications were adjusted as required.
The Underlying Technique of Artificial Intelligence
Artificial Neural Networks (ANNs) is what AI systems run on, and it works by attempting to simulate the way the brain works, with the aim of making it learn. When programmed well, ANNs can detect patterns in text data, speech, visual images, and other information. And it has been the basis for several advancements in the AI world in the past years.
The traditional AI relies on input in ‘teaching’ an algorithm about a certain subject. Input here means feeding the system with enormous amounts of information consistently. And this has been applied practically in several technologies, including Google’s language translation services, Facebook’s facial recognition software, and Snapchat’s image altering live filters. However, a lot of time is invested in the data input, and the system can only handle one kind of knowledge.
There is now a new type of ANNs, which is the Adversarial Neural Networks, which attempts to bring two AI bots together to allow them to learn from each other. This will not only speed up the learning process but also enhance the output of the AI systems.