Exploring How Artificial Intelligence Can Support Telemedicine

How might we make the remote monitoring experience for cancer patients seamless?

Role
UX researcher
Timeline
10 weeks
(June 2024 -
August 2024)
Team
2 designers +
1 UX intern +
me!
Skills
User research
Experience design
Tools
Figma 

Figjam

Dovetail

Miro
Project Overview
For the summer of 2024, I interned at Memorial Sloan Kettering Cancer Center (MSKCC) as a UX Research Intern. My project focused on the MATCHES (Making Telehealth Delivery of Cancer Care at Home Effective and Safe) Initiative. MATCHES is funded through the White House’s reignition of the Cancer Moonshot whose goal is to reduce the death rate from cancer by at least 50 percent over the next 25 years. Ultimately, it hopes to improve the experience of people and their families living with and surviving cancer, and to ultimately end cancer as we know it.

The focus at MSKCC is to provide more care in the home and less care at clinic sites, which will hopefully improve access for those who face barriers to care (time and financial toxicities, lack of family support, distance to treatment centers, etc,.).
With the introduction of telemedicine, more patients want to receive their care at home which requires additional interventions to support asynchronous communication and health information management. The goal of my internship is to reimagine how to provide a seamless and satisfying experience for these patients receiving care at home. I focused on how patients remotely monitor their health at home using technology.
Why is Care at Home Important?
Traditional Hospitalization
  • Tethered to hospital bed with limited mobility
  • Exposure to infectious diseases
  • Visitors in limited number and only during defined windows of time
  • Interrupted multiple times in their day/night for vital signs, phlebotomy, and medication administration
  • Functional decline
  • Delirium
  • Loneliness
Hospital at Home
  • Comfort of home
  • Virtual rounds and home visits
  • Remote monitoring
  • Phlebotomy, medication admistration as needed at home
  • Limited functional decline
  • Increased patient satisfaction
  • Increase/relieve caregiver strain
What Are Cancer Patients Feeling?
Before conducting patient interviews, I sought to understand cancer patients on a deeper level. I referenced Reddit, an online forum where users are able to post raw, unfiltered thoughts on different topics. All users are anonymous and therefore, I thought this was a good medium to explore the emotions cancer patients were feeling throughout their treatment journey. This was a safe space for them to be vulnerable and share thoughts they might not otherwise feel comfortable sharing.
During my initial search using the key word “feeling”, I wanted to gain a general understanding of the emotions cancer patients are feeling throughout their treatment. The main feeling I encountered was loneliness, and I found dozens of comments about how patients were feeling isolated and were struggling with no support system.
I thought loneliness was a unique feeling that could be exacerbated by being treated at home. By removing the normal day-to-day interactions patients have with the care team and hospital staff when treated on-site, patients lose another form of connection. I wanted to see if this hypothesis was applicable and incorporated these findings into our patient interview guide.
What Should I Be Asking Patients?
While our project focused on 3 specific aspects of a patient’s remote monitoring journey (continuous monitoring, convenient communication, and autonomous self care), I wanted to make sure we were asking the right questions that would allow me to pinpoint the challenges of patients monitoring their health at home. 


From version 1 of the interview guide to the final version we referenced when speaking to patients, I focused on narrowing down the questions to target how patients were able to set up and incorporate different devices into their lives, the different communication channels they used, and the difficulties they faced throughout this process from onboarding to discharge. From the desk research I conducted,  I added questions to explore the feelings remote monitoring patients experienced when monitoring their health at home.



Some of the questions I asked were:
1. Do you mind sharing why you agreed to participate the MATCHES IO program?
2. Can you tell whether you have a good reading or bad reading from the pulse oximeter?
3. What do you typically discuss with your care team remotely via [communication method]?
4. What steps of the programs gave you the most confidence and the least confidence?
5. If another patient were to enroll in a similar program, what advice would you give them?
Unexpected Findings
I conducted user interviews with both MSK and non-MSK cancer patients. For the MSK cancer patients, I coordinated with a physician to speak to patients in the MATCHES IO program. For non-MSK cancer patients, I used Userzoom, an external participant recruiting tool, to speak to 4 non-MSK cancer patients who had similar experiences remotely monitoring their health.
  • 30 minute phone call interview
  • 2 patients
  • 2 males
  • 65+ years old
  • Both patients used a pulse oximeter
  • 45 minute video call interview
  • 4 patients
  • 1 female, 3 males
  • 51-66 years old
  • All patients used a blood pressure monitor
Here is what they had to say:
“I think it can be more complicated. It’s all in the instructions, right? Whether it's a written instruction, YouTube instruction, or the care team hand-holding. I think if you get good instructions early on and then have kind of a cheat sheet to back you up, I'd jump on any technology they want to try.”
Patients value ease of use in their devices. They are open and willing to incorporate more difficult technologies to support their care.
“I wanna understand. I wanna know. I wanna see the warning signs. If something's gonna go off the rails so that I can be part of the solution, not just sit here and have it handed to me.”
Patients want to feel involved in their treatment and care journey.
“I just say I was out of town or I'll make up some sort of excuse. I've been known to just write numbers in. That’s not the first time you’ve heard that.”
There is an additional burden devices put on patients.
“Everything I have to do now based upon these numbers. I have to do the research. Is this good? Is this bad? I have to ask all the questions.”
Patients have difficulty understanding reports about their health.
“What was the purpose of it? What were they looking for? Are these good results or bad results? And then I've got to go back and look up the test from last quarter in my portal. Compare the results. Are they better? Are they worse? Are they the same? So there's work on my part.”
Patients seek to understand how the different aspects of their health data connect.
Understanding the Takeaways
Remote monitoring can be burdensome
  • One device is manageable while having multiple devices can become harder to integrate
  • Manually uploading data can become tedious
  • When remote monitoring becomes “too much work”, it can be seen as intrusive

Participating in remote monitoring provides a sense of engagement in their care
  • Feel as if they are actively participating in something that supports their treatment
  • Patients specifically found external motivation from their children
Enhance the remote monitoring experience through playful and easier access
  • Convenience was top-priority for patients
  • Prefer the phone when using applications because of portability and accessibility
  • Prefer to view documents and reports on the desktop due to familiarity and screen size
Defining the Problem Space
After reviewing the findings, I hypothesized that patients’ emotional and mental needs were not being adequately supported throughout their remote monitoring program. Stemming from our findings online to the conversations we had with cancer patients, we were beginning to see what the problem space was.

These insights helped us arrived at my focus:

How might we adequately support the emotional and mental needs of cancer patients remotely monitoring their health at home?
Ideation
To start off my brainstorming session, I generated how might we (HMW) questions to turn the challenges from our insights into opportunities for design. I wanted to explore the specifics and delve deeper into the problem space before brainstorming any solutions. Stemming from my initial hypothesis, much of the questions revolved around how we can empower patients remotely, visualizing aspects of the rm process, and ensuring patients have a clear understanding of their care.
I then grouped our HMW questions into opportunity areas to help us identify which problem spaces to focus on. The problem spaces with the most HMW questions were centered around involvement and encouragement, understanding health data, and technology integration.

After careful consideration, we decided to prioritize involvement and encouragement, as well as understanding health data, because these areas were a focus in our conversations with patients. This led us to refine our focus into two key HMW questions:

HMW simplify and personalize health information to improve patient understanding of their health?

HMW empower patients to actively engage and participate in their care?

While brainstorming solutions to address the questions created in the previous round, I followed IDEO’s brainstorming rules emphasizing the aim for quantity, encouragement of wild ideas, and deferred judgement. I wanted to generate solutions to address patient pain points without considering any constraints – anything was possible!
For our final round, I wanted to narrow down which solutions to move forward with. I, alongside my managers and the co-design intern voted for the concepts we thought would best support the mental and emotional needs of cancer patients during their remote monitoring journey.

One interesting idea we had was an AI activity recommender that would suggest tailored activities based on the patient's current symptoms and treatment status. This tool could potentially help patients engage in appropriate and beneficial activities, enhancing their quality of life. However, we disqualified this idea because it addresses a different problem space and falls outside the scope of our project.

Another idea we had was creating a simulation to connect the nervous system of the physician to the patient. The doctor could then clearly and easily understand the pain the patient was experiencing. Though this idea could help potentially enhance the doctor’s empathy and help patients clearly communicate the severity and context of their symptoms, we disqualified it due to ethical concerns surrounding the direct simulation of pain and the potential risks involved.
Finalizing Concepts
After voting, I decided to move forward with 2 concepts that I thought would best address the needs of cancer patients.
The first concept focuses on the idea of a motivational garden. As a patient consistently takes their readings, the garden will flourish as a reward. With the integration of artificial intelligence, the application will learn when patients normally take readings and when they are available to send a reminder if they ever forget to take their readings. One common trend among all patients was how they enjoyed being involved in their care. By providing the patients with control over something, I hoped it would give them a sense of ownership. I also believed that this would alleviate the loneliness some patients were feeling as an interactive garden provided a sense of companionship and progress. By engaging with the garden, patients would see the direct results of their efforts and adherence to their remote monitoring program.
The second concept is a data visualizer that simulates the health data a patient receives and displays the data on a human body in an easily understandable way. I believed this concept would offer patients a clear way to understand their health data by mapping it onto a human body. Every patient I spoke to had difficulty understand the reports they receive and would reference multiple external sources alongside consulting their physician. I hoped this approach would make complex information more relatable and less intimidating, providing a consolidated space where patients can refer to when trying to understand what is happening in their body and how it all connects.
Visually Communicating Information to Patients
To visually communicate my concepts, we decided to create storyboards for each idea. I led the development of the motivational garden concept. My goal was to elicit honest feedback from patients by placing them in hypothetical scenarios, allowing them to visualize how they might incorporate and use these concepts in their daily lives to manage their conditions. During concept testing, we aimed to capture their initial reactions, assess their acceptance, and gauge the desirability of the concepts. This feedback was crucial in refining and validating our solutions to ensure they meet the needs and preferences of the patients.
During the concept testing sessions with externally recruited non-MSK cancer patients, I walked them through the storyboard and asked them to image that they were a cancer patient who has just begun a remote monitoring program and experiencing this journey themselves.
Concept Testing Takeaways
From the concept testing, I was able to observe patients' reactions to my story in real-time. When presented the motivational garden concept, most patients reacted very positively, with one patient noting that it "definitely made him smile." Many valued the reminders for tracking readings, although one patient found the garden aspect to be childish and preferred only the reminders. Despite this, most felt that seeing the garden grow would provide a sense of accomplishment. For patients who didn’t find the garden theme valuable, we’ve determined that providing customizable themes and simplified indicators that they’ve completed their readings would help them better achieve a similar sense of accomplishment.
If I Had More Time...
I would explore further customization options to tailor our concepts to individual patient preferences, as many patients expressed a desire for customizable themes in the virtual garden to enhance its appeal. After refining our prototypes, I would conduct usability testing to observe how users interact with and integrate our products into their daily lives, ensuring that our solutions effectively meet their needs.
Internship Learnings
One of the most invaluable takeaways I gained from this project was learning how to adapt and overcome obstacles that arose during my project. I initially encountered significant delays in our project due to challenges in connecting with and scheduling patient interviews, which stalled several project milestones. To address this, I pivoted to conducting UserZoom interviews with patients who had similar experiences to those in MSK’s remote monitoring program. Having a contingency plan to navigate obstacles allowed us to gain a different perspective on the patient experience with remote monitoring and keep the project moving forward.

Through my interactions with MSK patients and their care team, I learned how to manage and interact with different stakeholders. What initially seemed like a straightforward process of contacting the lead doctor of the MATCHES program and his patients evolved into a more complex, multi-faceted process. First, we needed to persuade the doctor of the value of interviewing his patients, which led us to prepare and present a project proposal. We then coordinated with his team to review and approve our interview scripts before proceeding with the user interviews. Following these steps, we stayed in contact with the remote monitoring program coordinator to schedule patient interviews and arrange meetings with other teams involved in remote monitoring.

After conducting user interviews, I synthesized my UX research findings into a slide deck which we presented to various stakeholders in the remote monitoring space. From the feedback I received on my research share-out, I am glad to hear that my work will serve as a centralized space for stakeholders to reference how patients both in and outside of MSK feel about their remote monitoring programs. For the doctor and the clinical team involved in remote monitoring, our work will inform and support how he approaches and encourages patients to participate in these programs. For the product team focused on remote monitoring, I hope these insights will refine their development processes in creating user-centered solutions. Ultimately, in the future, I hope that my work this summer helps to inform the way future remote monitoring programs engage and support their patients.
The People That Make the Place!
I wanted to pay a special thanks to the people who made this summer one of my favorites so far!

I am so grateful for my mentors, Qingyan Ma and Bo Kim, and the entire UX team at MSK for their guidance these past few weeks. Thank you for giving me the opportunity to delve into such a meaningful area of work. I’ve learned so much about the world and about myself than I thought possible! I hadn’t expected my work to revolve so heavily around user research but my interactions with cancer patients have been so deeply impactful. It’s been incredibly rewarding to contribute to a project that will inform the future work of designers, technologists, and clinical teams.

A special shoutout to Mandy, the co-intern who worked on this project with me, and the other interns I met this summer at MSK. Thank you all for making this summer so incredibly fun, joyous, and full of laughter. I will cherish these friendships for the rest of my life <3