Seven reasons why the time has come for AI technology in cancer care
- Data is proliferating: Worldwide healthcare data doubles every 24 days1—and much of that is unstructured data hidden in electronic health records
- EHRs are a key source of data—and burnout: For every hour a physician spends with a patient,
they spend two hours doing EHR-based clerical work. 2 The EHR
burden has pushed burnout rates up more than 10 percent since 2011.3
- Meanwhile, fewer doctors are practicing oncology: A shortage of more than 2,300 medical
oncologists in the United States is anticipated in 2025.4 Other
countries are facing shortages, as well. A recent study found that there were eight countries with no clinical
oncologist available to provide care for patients with cancer. In 39 countries, a clinical oncologist would
provide care for more than 500 patients with cancer. An extreme shortage of clinical oncologists, greater than
1,000 incident cancers per clinical oncologist—existed in 25 countries in Africa (78%) and two countries (11%)
- And cancer is spreading: There was an estimated 18 million cancer cases worldwide in 2018—and
that number is projected to reach 29.5 million by 2040.6
- More care is needed: It comes as no surprise, then, that the American Society of Clinical
Oncology predicts a 42 percent increase in the demand for cancer treatments over the next decade.7 And the global market for oncology therapeutic medicines will
reach as much as $200 billion by 2022, averaging 10–13 percent growth over the next five years, with the U.S.
market reaching as much as $100 billion by 2022, averaging 12–15 percent growth.8
- Clinical research keeps growing: The countries generating the most cancer research produced
88,529 publications between 2010-2014.9 This amount of research
is impossible for a human to keep up with.
- Yet, clinical trials struggle to enroll patients. Only 3 percent of adults with cancer are
enrolled in clinical trials10 and 80 percent of US clinical
trials fail to meet recruitment timelines.11
More data, more research and more patients with fewer oncologists at risk for burnout. While the outlook may sound
bleak, the good news is that technology is catching up with
healthcare data—which could help patients and the oncologists who treat them.
Connecting patient data with potential treatment options
Artificial intelligence (AI) technologies like IBM
Watson® for Oncology are helping physicians worldwide keep up-to-date with the growing body of medical
literature to make connections with key insights in patients’ medical records. Using natural language processing (NLP),
Watson for Oncology also consumes massive amounts of medical literature to extract potential evidence-based treatment
recommendations that may be a good fit for a patient. Parsing the data quickly and perceptively, Watson for Oncology
presents treatment options ranked by level of confidence and includes supporting evidence. The oncologist can then apply
their own expertise to identify the most appropriate treatment options.
Matching patients to potential clinical trials
AI is also helping to match more patients with potential clinical trial opportunities. Again,
massive amounts of patient and clinical trials data are consumed and analyzed using NLP and advanced cognitive
algorithms. Explicit eligibility criteria are weighed against specific patient characteristics to determine a potential
match. The output is an ordered list of relevant trials a patient is eligible for, as well as the trials they were
excluded from. This technology is a boon for oncologists and their patients searching for treatment options and clinical
trial coordinators trying to meet enrollment targets.
Cancer is the second leading cause of death globally, and is responsible for an estimated 9.6 million deaths in 2018,12 and the healthcare industry is racing to find answers. AI
technology offers exciting potential and is already hard at work helping oncologists connect patients with treatment
options and clinical trial opportunities, even in the face of exponentially increasing amounts of data.
1 Marconi, Katherine and Lehmann, Harold. Big Data and Health Analytics. CRC
Press, 2014. Accessed at: http://bit.ly/1UjEtLL
2 NLP: Enabling The Potential of a Digital Healthcare Era. Chilmark Research. July
3 NLP: Enabling The Potential of a Digital Healthcare Era. Chilmark Research. July
4 Yang W, Williams JH, Hogan PF, et al: Projected supply of and demand for
oncologists and radiation oncologists through 2025: An aging, better-insured population will result in shortage. J Oncol
Pract 10:39-45, 2014
5 DOI: 10.1200/JGO.17.00188 Journal of Global Oncology. published online
February 8, 2018. Accessed at http://ascopubs.org/doi/abs/10.1200/JGO.17.00188
7 The State of Cancer Care in America, 2014: A Report by the American Society of
8 Global Oncology Trends 2018. IQVIA Institute for Human Data Science. Accessed at
9 Cancer Research Current Trends & Future Directions. Elsevier, 2016.
10 Schuler, Peter and Buckley, Brendan. Re-Engineering Clinical Trials: Best
Practices for Streamlining the Development Process. 2015
11 “Clinical Trial Educator Program—A Novel Approach to Accelerate Enrollment in
a Phase III International Acute Coronary Syndrome Trial,” Clinical Trials, 2012.
12 World Health Organization.