Watson (AI) For Oncology-A thousand case studies
Watson is a computerized question answering system that can answer questions posed by natural language. It was developed by a research team led by principal investigator David Ferrucci on IBM’s DeepQA project. Watson is named after IBM’s founder and first industrial CEO, Thomas J. Watson.
The computer system was originally developed to answer the Jeopardy! Quiz show. And in 2011, Watson Computer Systems competed in Jeopardy! He faced champions Brad Rutter and Ken Jennings and won the first prize of $ 1 million.
In February 2013, IBM announced that the first commercial application of the Watson software system would be used for management decisions in the treatment of lung cancer at Memorial Sloan Kettering Cancer Center, New York City, and WellPoint (now Anthem).
In 2013, Manoj Saxena, IBM Watson’s chief operating officer, stated that 90% of field nurses using Watson now follow the guidelines.
Background
On July 29, 2016, IBM and Manipal Hospitals (India’s leading hospital chain) announced the launch of IBM Watson for Oncology for cancer patients. This product provides doctors and cancer patients with information and knowledge to help them determine evidence-based and personalized cancer treatment options. Manipal Hospital is the second hospital in the world to adopt this technology and the first hospital in the world to provide expert advice to patients online through its website.
Artificial intelligence is used to provide information-intensive decision support. Presented in this report is an experience to show how artificial intelligence can add value to MDT’s decision-making capabilities and pave the way for personalized therapy.
Methodology
Between 2016 and 2018, the Multidisciplinary Oncology Committee of the leading cancer centers in India evaluated 1,000 cases of breast, lung, and colorectal cancer. After the tumor committee’s decision, MDT also received a recommendation from Watson.
MDT reviewed its decision after considering Watson’s recommendations and the evidence presented to support its decision. In the case of changing a decision, an objective evaluation is performed requiring MDT to cite the reasons for reviewing and changing its decision.
Results
Among 1,000 cases, breast cancer, lung cancer, colon cancer, and rectal cancer were 620, 130, 126, and 124, respectively. There are 712 non-metastatic cases and 288 metastatic cases amongst these. The mean age of the patients was 54.3 ± 12.2. In all cancers i.e. 92%, rectal cancer 93%, breast cancer 92%, lung cancer 89%, colon cancer 81% reached treatment consistency, MDT changed the decision of 136 cases (13.6%).
Watson provided the most recent evidence that 55% of patients need up-to-date treatments, 30% of patients use personalized alternative methods and new insights from genotype and phenotype data of 15% of the patients. The clinical experience is ever-evolving, so the Oncology Committee changed its decision.
Click here for relevant content
Conclusions
This study shows that the cognitive computing decision support system is expected to greatly reduce the cognitive load on oncologists by providing experts with the latest and most recent evidence-based knowledge for related decisions.
In tertiary cancer centers with MDT function, the incremental lead of more than 13.6% in itself illustrates the value of having experienced colleagues like Watson for tumor treatment.
In an environment where there is a lack of timely access to high-quality cancer knowledge and information, it will definitely add more value. By providing high-quality cancer care services, these systems can become a strong and valuable link between patients and physicians.
