Artificial Intelligence in Cancer Care: Tools for Better Outcomes
by Preethi Kannan
Artificial intelligence (AI) is an expanding field in technology and has recently made great strides in the health industry. However, few patients are aware that it is used in their care plan and many do not know what it is. According to the Brookings Institution, “opinion surveys show that many people confuse AI with super-powered robots or hyper-intelligent devices” and “even top business leaders lack a detailed sense of AI.” (1) In reality, AI has been around for almost 60 years. (2) As AI tools become more common across hospitals and clinics in the US, it is important for patients to understand how these tools can benefit their care. To explore the role that AI plays in cancer care, we sat down with Dr. Mia K. Markey, Professor of Biomedical Engineering at The University of Texas at Austin and Adjunct Professor of Imaging Physics at The University of Texas MD Anderson Cancer Center, to discuss common concerns and dispel myths about artificial intelligence.
What is Artificial Intelligence and How is it Used in Cancer Care?
Artificial intelligence refers to machines that respond to stimulation consistent with traditional responses from humans. First, an engineer uses existing data to teach the computer a task such as detecting an object in a picture. Then, the computer uses this algorithm to interpret new data and make predictions. In the context of healthcare, these algorithms have the potential to adapt and optimize performance in real-time to continuously improve health care for individuals. (3) Dr. Mia K. Markey explains that, “AI in healthcare creates models based on the experiences of prior patients to make predictions for current patients,” She goes on to define intelligence as a moving target in the AI field. “As soon as a computer can do it, we stop considering it as intelligence.” In other words, many things we considered to be “advanced artificial intelligence” 30 or 40 years ago are now regular tools in our daily lives that we no longer identify as AI. For example, common tools that use AI include navigation apps, music streaming services, and virtual personal assistants such as Apple’s Siri and Amazon’s Alexa.
AI has been used in our healthcare for decades. For example, multiple AI tools communicate with each other from the moment a healthcare provider prescribes a medication to the point when a pharmacist fills the prescription. These tools help streamline procedures, provide a second check, reduce chances of errors, and improve efficiency. These applications range from medical imaging to electronic health records. Dr. Markey explains that AI can help with accuracy, efficiency, and consistency of medical tasks. For example, when conducting breast cancer imaging, radiologists typically look at the mammogram independently first. Then, AI software can provide a second opinion as to whether there is something concerning in the image that might warrant additional attention. Dr. Markey emphasizes that “these systems in no way replace what the radiologist does. They operate in conjunction with the radiologist to assist them.”
Can we Trust AI?
In medicine, perhaps more so than in other aspects of our life, we are often concerned about trusting technology. In the US, the FDA has a very strict regulatory process for all new medical technologies (including AI) before they can be applied in the medical setting. These regulations are frequently updated to keep up with the evolving technology of the AI field. (3) Dr. Markey points out that, “computer-aided detection systems for screening mammography were approved by the FDA in the 1990s and have been in widespread clinical use ever since. In fact, it is quite likely that an American woman who has had a mammogram recently had an AI tool providing a second check.” (4) While artificial intelligence tools have proven to make diagnoses as accurately as physicians in certain specialties, Dr. Markey emphasizes the distinction of these tools being involved clinically as a second opinion rather than replacing physician opinion. She explains that “these are tools, they are not taking over. They are helping a healthcare team provide better care for you.”
AI in Delivering Person-Centered Care
According to the World Health Organization, person-centered care is organized around the health needs and expectations of people rather than diseases. (5) Livestrong’s commitment to person-centered care reorients the focus of cancer care to the whole person — emotionally, physically, and practically. To provide collaborative and coordinated care, individuals are integrated directly into their care team so they can make informed decisions on their own care plan. The goal of person-centered care is to empower individuals to take charge of their health. For the person-centered care approach to meet this goal, individuals need access to the education and support necessary to make informed decisions.
AI presents an opportunity to provide personalized information for patients and providers to use when making informed decisions. This chance to make recommendations that are more specific to the patient allows healthcare providers to not only consider the objective clinical measures but also allow the patient’s own values to come into play. As Dr. Markey described, decisions have three components: the options that are available to you, the possible outcomes of each choice, and the value you place upon each of those possible outcomes. AI allows patients to understand what their options are and what their outcomes may look like so that patients can have more meaningful conversations with their health care team.
During one’s cancer journey, one often feels powerless and overwhelmed by the new information. Simultaneously, having more information does not always mean one immediately makes better decisions. This is why AI aims to provide power back to the individual by structuring information for patients and providers to make better-informed decisions about their care plan.
Dr. Markey offers this advice for any survivor or caregiver wanting to know more about AI’s role in their cancer care. “Take a step back and consider the tools in your life that you already use that rely on sophisticated algorithms. They may make mistakes sometimes, but we still have our human judgment to help us in those cases, like when the GPS might tell us to drive off the road. Our prior knowledge helps us recognize the mistake, which is why your health care team still has professional judgment over the suggestions made by AI tools. Above all, talk to your health care team if you have any questions about how AI tools are being directly used in your care plan.”
Dr. Mia K. Markey is a Professor of Biomedical Engineering at The University of Texas at Austin and Adjunct Professor of Imaging Physics at The University of Texas MD Anderson Cancer Center. She is a Fellow of both the American Association for the Advancement of Science (AAAS) and American Institute for Medical and Biological Engineering (AIMBE) and is a Senior Member of both the IEEE and the SPIE.
Preethi Kannan is a Mission Intern at the Livestrong Foundation and an undergraduate Public Health student at The University of Texas at Austin.
Bibliography:
Bibliography:
1. West, Darrell M. “What Is Artificial Intelligence?” Brookings (blog), October 4, 2018. https://www.brookings.edu/research/what-is-artificial-intelligence/.
2. “The AI Paradoxi Is What Everyone Gets Wrong about AI — Business Insider.” Accessed October 18, 2020. https://www.businessinsider.com/misconception-artificial-intelligene-2015-10.
3. Health, Center for Devices, and Radiological. “Artificial Intelligence and Machine Learning in Software as a Medical Device.” FDA, October 5, 2020. https://www.fda.gov/medical-devices/software-medical-device-samd/artificial-intelligence-and-machine-learning-software-medical-device.
4. Keen JD, Keen JM, Keen JE. Utilization of Computer-Aided Detection for Digital Screening Mammography in the United States, 2008 to 2016. J Am Coll Radiol. 2018 Jan;15(1 Pt A):44–48. doi: 10.1016/j.jacr.2017.08.033. Epub 2017 Oct 6. PMID: 28993109.
5. WHO. “WHO | What Are Integrated People-Centered Health Services?” World Health Organization. Accessed October 22, 2020. http://www.who.int/servicedeliverysafety/areas/people-centred-care/ipchs-what/en/.
