Using AI to Help Cancer Patients Stick to Medication Regimens

“If you’ve ever downloaded a health app, those tend to be pretty dumb,” said Susan Murphy, the Mallinckrodt Professor of Statistics and Computer Science.

Cancer patients who undergo stem cell transplantation face a long and challenging recovery, requiring medications with debilitating side effects and support around the clock. However, studies have shown that more than 70 percent of patients do not adhere to their drug regimens, making it difficult for them to recover and manage their conditions effectively. To address this issue, statistician Susan Murphy and her team are developing mobile apps that use artificial intelligence and sensing technologies to provide personalized support and encouragement to patients, caregivers, and healthcare professionals. These apps are designed to provide real-time adaptations, offering psychological rewards and leveraging social networks to help users stick to their goals. The approach, known as “just-in-time adaptive intervention,” aims to provide support at just the right time by registering changing needs and contexts.

1. Health management, especially for the sickest, typically requires involvement of others, such as family-care partners, who have primary responsibility for managing cancer-related medications.
2. The researchers are working with software engineers, cancer clinicians, and behavioral scientists to develop an app for stem-cell transplant patients and their primary caregivers, usually parents.
3. The app will employ reinforcement machine learning, in which the software will “learn” from previous interactions and tailor timing and content according to when they have been most useful to patients.

The app will inform sequential decisions, including when and whether to send motivational prompts to the patient, and whether to send messages and reminders to both patient and caregiver. The application includes a word-guessing game that fosters social support and collaboration between patient and caregiver.

Harvard postdoctoral fellow Ziping Xu is leading the ADAPTS HCT algorithm development, which will focus on adolescent and young adult patients who have had stem-cell transplants in the 14 weeks post-surgery. Xu hypothesizes that improving the relationship between patients and their caregivers will lead to better medication adherence and overall patient function. Just-in-time adaptive intervention is a key component of the app, which provides support at just the right time by registering changing needs and contexts. The algorithm will continually learn and adapt from interactions with each patient, tailoring its decision rules to improve patient outcomes. The Murphy lab is deploying its algorithmic expertise across other domains, including a recent pilot-testing of MiWaves, a program aimed at young adults who are abusing cannabis. MiWaves uses a similar approach to ADAPTS HCT, employing reinforcement machine learning to help patients reduce their daily cannabis use. In addition, the lab is several years into a project called Oralytics, which aimed to refine the delivery of push notifications to help patients adhere to a tooth-brushing protocol. The project recently wrapped up a 10-week randomized trial, which included 70 participants who used a wireless-enabled toothbrush that sent data to the team’s collaborators at Proctor and Gamble. Graduate student Anna Li Trella led the Oralytics project through the first trial, which collected valuable data on the challenges of running an algorithm in real-life settings. The team is now working to improve the algorithm’s ability to handle messy problems like missing data and software errors. Susan Murphy sees her lab as creating practical pocket coaches who can help people get where they want to go. She believes that these digital supports can provide an effective alternative to human coaching, especially for those who may not want or need intensive human interaction.

news

news is a contributor at AlgoHay. We are committed to providing well-researched, accurate, and valuable content to our readers.

You May Also Like

Fundamentals of Quantum Algorithms

The Quantum Algorithm Revolution: Unveiling the Future of Computing through Quantum Mechanics In an era where computational power is the...

Algorithm Design: Divide and Conquer

Algorithm Design: Unraveling the Core Strategies Behind Efficient Solutions Designing algorithms is at the heart of computer science, enabling software...

Search Algorithms for Pattern Matching

The Power of Search Algorithms in Modern Computing In the digital age where data reigns supreme, search algorithms have become...

Algorithm Applications for Recommendation Systems

The Power of Algorithms in Real-World Applications: A Deep Dive for Developers In an era defined by data, algorithms have...