Microsoft’s Artificial Intelligence Approach to Solve Cancer Problem
The Microsoft’s research laboratories have computer scientists, engineers, programmers and an array of experts working on solutions to the technology industry’s vast problems, which include security, system design and quantum computing. What may come as a surprise is that among these experts, there are those focused on applying computer science to find solutions for one of the human race’s greatest challenges; Cancer.
Microsoft scientists are looking into cancer as if it was a computer virus and, instead of the traditional test tubes and beaker research, they are using algorithms and computers. Their approach uses artificial intelligence to target affected cells and completely eliminate the disease. One such research team is applying natural language processing and machine learning to aid leading oncologists in finding out effective and individualized cancer treatment by efficiently sorting through the bulk of research data that is currently available.
Another group within the Microsoft research team is incorporating machine learning into computer vision. This is to offer radiologists a better understanding of the progress of their patients’ tumors as another group works on “moonshot” efforts for scientists to be able to fight diseases such as cancer through programming cells.
The Computer Science Approaches Being Applied
According to Jeannette M. Wing, the Microsoft VP running the tech giant’s research labs, individual projects vary widely but the cancer solving initiative focuses on two basic approaches. The first is based on the idea that cancer alongside other biological processes form information processing systems and from that perspective, the tools applied to understand computational processes like programming languages, model checkers and compilers are also used in understanding biological processes.
The second approach is data driven. It is based on the idea that machine learning techniques can be applied to the biological data that is currently available and the sophisticated analysis tools can be used to gain a better understanding of cancer and thereafter treat it. The two approaches, however, share common ground and that is the fact that biologists and computer experts need to jointly bring in their knowledge to deal with the problem.
According to Wing, Microsoft is using computer science to fight cancer and is heavily investing in this venture since it is in line with the software firm’s core mission. Moreover, the company’s massive investment in the field of cloud computing, makes it a natural fit for the cancer research as it requires great computing capacity to solve the big challenges.
Better Treatment from Knowledge Organization
The research teams have continued to put in more effort after breakthroughs have revealed the role played by genetics in getting and treating cancer. This has resulted in greater focus on individualized treatment, which is also referred to as precision medicine.
According to David Heckerman, a senior director of Microsoft genomics group, the knowledge on cancer treatment has undergone a revolution. This is due to the understanding that it’s important to treat the cancer genomics and not just treat the tissue as has been the case. The ability to map the human genome is one of the recent advances that have enabled this but the amount of data available for research has been acknowledged as presenting a challenge.
The engagement of artificial intelligence is seen as the problem solver here, and the use of machine learning to automatically do much of the data reviewing is of great help to biological researchers. The machine learning system is applied in cancer research to sort and organize the millions of research and medical records. Cloud computing platform hosts the medical tools to be used for biologists who may lack powerful hardware to run them independently.
Liternome is one such tool that Microsoft has developed where researchers can leverage cloud computing to go through research papers and find those that are relevant to specific diagnoses. The Knight Cancer Institute is currently using the tool to research on treatment for myelod leukemia.
Microsoft computer experts point out that they have gone an extra mile to make their systems easily usable even for those who may lack interest or have no background in computer science and technology. This also includes designing computing tools similar to those already in use by laboratory scientists and learning to interact in the language of biologists and doctors.
The company’s long-term plan is to take the initiative further in the next ten years. By then, they hope to have developed a molecular system that detects disease and that could have an impact beyond cancer treatment.