Advertisement

AI in Cancer Care: Embrace the Change


Advertisement
Get Permission

Anant Madabhushi, PhD

Anant Madabhushi, PhD

Jame Abraham, MD, FACP

Jame Abraham, MD, FACP

According to Google Chief Executive Officer Sunder Pichai, artificial intelligence (AI) is “the most profound technology humanity is working on—more profound than fire or electricity or anything that we’ve done in the past.” The impact of AI on health care and especially cancer care will not be an exception. This new technology will change our present and future. With that in mind, The ASCO Post will be publishing an occasional series of articles to help practicing oncologists and cancer caregivers understand the scope of AI and how it may affect their daily lives. In this commentary, we will outline our perspectives on some of the potential opportunities for AI in the cancer care space.

According to Prof. John McCarthy, a founder in the field of AI, artificial intelligence “is the science and engineering of making intelligent machines, especially intelligent computer programs. It is related to the similar task of using computers to understand human intelligence, but AI does not have to confine itself to methods that are biologically observable.”

Types of AI

Although there are several different types of AI approaches, broadly speaking, AI can be designated as belonging to one of the following three categories.

1. Narrow AI: This type of AI performs a specific task, such as a Web search or facial recognition. In general, it will not learn more than it is programmed to do. It is considered “weak” AI.

2. General AI: This type of AI can perform a broad range of tasks that humans can do, such as reason, learn, and perform intellectual functions. This is also known as “strong” AI.

3. Superintelligent AI: Superintelligent AI would be able to perform beyond a human level of intelligence. This potential future form of AI would clearly outsmart us in every field of knowledge and activity.

The leading technology companies are already deploying AI in ways that have become part of our daily lives. These companies and others are investing billions of dollars in the health-care space, from patient care to advanced drug discovery through synthetic biology. These developments have the potential to transform health care (and cancer care specifically) beyond our imagination. As AI shapes the future, we should be prepared to embrace the change.

AI Applications in Cancer Care

There are at least six potential areas where AI might enhance cancer care.

1. Optimizing Patient Access (eg, With an AI-Informed Cancer Concierge): For a patient with newly diagnosed lung cancer, getting to the right specialists in a timely manner can be a daunting task. Choosing the right surgeon, medical oncologist, radiation oncologist, and pulmonologist can be confusing and frustrating. We can find the best hotel, make dinner reservations, and rent a car in Tokyo with a few clicks. Why can’t it be just as easy for our patients to access the right doctor the same way?

Critically, being able to provide an AI-informed guide to patients with newly diagnosed cancer could help them navigate the complicated cancer journey. A typical patient with a new cancer diagnosis spends many hours scheduling multiple appointments, tests, and infusions. Why couldn’t this be done using a computer interface that is agile and adaptable?

Many hospitals in the United States are running at 80% to 90% capacity. When hospitals are at maximum capacity, there is a tremendous backlog in emergency rooms, operating rooms, and outpatient clinics. Cleveland Clinic recently announced a partnership with Palantir Technologies to develop a virtual command center that would help optimize hospital throughput. Using the virtual command center, Cleveland Clinic has achieved a 10% increase in hospital transfer admissions at its main campus over the past year.

2. Doctor-Patient Interaction and Documentation: The electronic medical record is a significant improvement over the paper charts we used to dig through for each patient encounter. There is no doubt it has improved continuity of care for patients across hospitals and systems. It has changed our work environment completely and transformed the way patients connect with us. However, electronic medical record systems have led to nurses, doctors, and other providers spending hours at computers “charting.”

“[The AI boom has] the potential to transform health care (and cancer care specifically) beyond our imagination. As AI shapes the future, we should be prepared to embrace the change.”
— ANANT MADABHUSHI, PhD, AND JAME ABRAHAM, MD, FACP

Tweet this quote

Documentation, or spending time on electronic medical records, is the number one cause of health-care burnout. A recent study published in the Annals of Family Medicine showed that doctors spent more than 90 minutes of their precious downtime at home on electronic medical records.1 With AI, we have an opportunity to fix that. An AI assistant could help providers complete this mundane but important part of our jobs. A recent paper published in Nature Medicine suggested that large language models could summarize patient reports better than summaries created by medical experts.2

More than half of messages sent by patients through electronic health records, such as questions about scheduling, lab values, appointments, or refills, could be managed by AI. That could reduce the burden on providers and possibly help to reduce errors. Many hospital systems, including Cleveland Clinic and Emory University, are testing AI applications to help with administrative responsibilities, such as charting and inbox messages.

3. Improving Patient Safety by Reducing Errors: According to an Institute of Medicine report, about 44,000 to 98,000 unnecessary deaths happen in U.S. hospitals per year.3 Human error is a major cause of patient injury. AI could be used to help identify early on those patients who are at high risk for falls, infection, or delirium, which could help to prevent a serious safety event. Administration of the right medicine at the right time to the right patient, controlling a patient’s pain during a hospital stay, or treating infection can be ensured by technology.4

4. Transforming Radiology and Pathology: With the rapid advancement of imaging AI, providers may be able to obtain a real-time report on a pathology or computed tomography scan, instead of waiting days for results. AI incorporated with radiology can potentially give providers a preliminary read, pending confirmation by the radiologist, the way ECGs are done now.5 AI in imaging will transform care delivery in the future, which could lead to quicker time to treatment for our patients. AI tools for analyzing pathology images could transform the clinical workflow for pathologists, making it easier to triage slides and cases with no obvious cancer and helping them focus on the more challenging cases.6

The explosion of genomic information—from screening and monitoring for early-stage diagnosis to metastatic or advanced cancer diagnosis—can be daunting. New targets and novel therapies are difficult to keep up with for even highly specialized academic doctors. AI programs would be able to filter through massive amounts of genomic information and guide doctors in managing patients appropriately. In addition, AI would allow for the integration of genomics with clinical and imaging-related data to provide a more comprehensive and rigorous assessment of patient risk and response to different treatment regimens. In addition, AI could allow for the integration of genomic-, imaging-, pathology-, and electronic medical record–related information, enabling more accurate cancer prognosis and treatment response predictions and, thereby, better informing clinicians with regard to treatment decisions.

5. Revolutionizing the Research and Drug Discovery Process: Currently, drug discovery is a long and expensive process. It costs about $2 billion to $3 billion and takes 7 to 10 years for the development of one drug. But the rapid evolution of AI and synthetic biology will change this dramatically. Currently, clinical trials require coordination of hundreds of sites around the world and thousands of patients, involving much paperwork and regulation. AI can help streamline those efforts and reduce redundancy, as well as help to screen patients to identify the right patients for the right trials.7

The revolution in synthetic biology, which combines engineering principles with biotechnology such as gene editing (CRISPR) and DNA sequencing to create new molecules, will change the drug discovery process. With major players like Google entering into synthetic biology and drug discovery using their vast AI technology, drug development could be accelerated. It is poised to grow to a $40 billion market in the next 3 years.

6. Reducing Disparities and Addressing Social Determinants of Health: According to the Centers for Disease Control and Prevention, social determinants of health are the nonmedical factors that influence health outcomes. They are the conditions in which people are born, grow, work, live, and age, including the wider set of forces and systems shaping the conditions of daily life. Social determinants of health are thought to be responsible for 50% to 80% of patient outcomes. Generative AI can potentially filter through data already collected in the electronic medical record and identify social determinants of health, so providers can better personalize care based upon that information.8

We are excited about the exponential growth of this technology. As health-care leaders and researchers, it is up to us to help ensure that the power of AI benefits our patients and our fellow caregivers. 

DISCLOSURE: Dr. Madabhushi and Dr. Abraham reported no conflicts of interest.

REFERENCES

1. Arndt BG, Micek MA, Rule A, et al: More tethered to the EHR: EHR workload trends among academic primary care physicians, 2019–2023. Ann Fam Med 22:12-18, 2024.

2. Van Veen D, Van Uden C, Blankemeier L, et al: Adapted large language models can outperform medical experts in clinical text summarization. Nat Med. February 27, 2024 (early release online).

3. Kohn LT, Corrigan JM, Donaldson MS (eds): To Err Is Human: Building a Safer Health System. Washington, DC; National Academy Press; 2000.

4. Yang J, Hao S, Huang J, et al: The application of artificial intelligence in the management of sepsis. Med Rev 3:369-380, 2023.

5. Bera K, Braman N, Gupta A, et al: Predicting cancer outcomes with radiomics and artificial intelligence in radiology. Nat Rev Clin Oncol 19:132-146, 2022.

6. Bera K, Schalper KA, Rimm DL, et al: Artificial intelligence in digital pathology: New tools for diagnosis and precision oncology. Nat Rev Clin Oncol 16:703-715, 2019.

7. Vamathevan J, Clark D, Czodrowski P, et al: Applications of machine learning in drug discovery and development. Nat Rev Drug Discov 18:463-477, 2019.

8. Guevara M, Chen S, Thomas S, et al: Large language models to identify social determinants of health in electronic health records. NPJ Digit Med 7:6, 2024.

Dr. Madabhushi is the Robert W. Woodruff Professor in the Wallace H. Coulter Department of Biomedical Engineering, Georgia Institute of Technology and Emory University, and Executive Director of the Emory Empathetic AI for Health Institute, Atlanta. Dr. Abraham is Chairman of the Department of Hematology and Medical Oncology at Cleveland Clinic and Professor of Medicine at Lerner College of Medicine.

Disclaimer: This commentary represents the views of the author and may not necessarily reflect the views of ASCO or The ASCO Post.


Advertisement

Advertisement




Advertisement