IBM Watson Health
In January 2016, I was tapped to join the Watson Health Design team as a User Experience Designer. Since then I have been working on Watson for Genomics, in partnership with Quest Diagnostics and Memorial Sloan Kettering, a tool that analyzes genomic reports and recommends potential treatments based on literature review, and Watson for Patient Safety, a pharmacovigilance reporting and workflow management tool in partnership with Celgene. While much of the work I have done is confidential, I will discuss the challenges I faced as a designer working in the healthcare space, as well my learnings, below.
Design culture at IBM is driven by Design Thinking, a problem solving methodology pioneered by the d.school at Stanford. I started my time at IBM by participating in a series of projects aimed at teaching design thinking principles and workshop facilitation.
My design process at IBM begins with research: domain/competitive research and generative user research, in which I help to craft an interview protocol alongside our Design Researchers, and research synthesis using methods such as system mapping and affinity diagramming. Once we have gathered enough information on the users, the team develops personas and as-is scenarios to represent the pre-existing user experience.
While research is ongoing during the length of a project, after synthesis we move to form project goals by facilitating a stakeholder workshop. During this workshop, we prioritize the pain points we need to address in our product and begin to discuss the ideal to-be scenario for each user persona (in contrast to the current as-is scenario). This allows us to better understand the major milestones for the project and form a clear picture of an end-to-end experience, from MVP to beyond.
Once goals have been formed, I work on information architecture, wireframing, and prototyping. Based on the results of user testing or feasibility reviews with our engineering team, we iterate on the designs accordingly.
Watson for Genomics
Recent advances in DNA sequencing technology have enabled more patients to seek personalized care, yet researchers and pathologists often lack the ability to fully utilize this trove of medical data. Watson for Genomics is a cognitive tool that analyzes a patient’s lab results and recommends treatments it has identified from a constantly updated database of medical literature.
Because most researchers and physicians had never interacted with cognitive systems before, some of them were reluctant to trust Watson’s recommendations. For this project, I explored ways of presenting Watson’s recommendations and developed design solutions to explain its rationale to end users.
Another central question was, what if researchers were able to share their knowledge on gene alterations? Since genomics is a rapidly developing scientific field, we needed to develop ways for researchers to collaboratively share their findings with a community of pathologists.
Watson for Genomics has been featured on CBS’s 60 Minutes and was credited with saving the life of a cancer patient in Japan. To learn more about the technology, visit the Watson for Genomics page at IBM.
Watson for Patient Safety
Each year millions of patients around the world are affected by adverse events resulting from drug side effects. Pharmaceutical companies are under strict timelines to report adverse events to regulatory agencies, but due to medical complexity, manual processes, and lack of reported data, they find it difficult to detect signals which could lead to adverse events. As a part of the Watson for Patient Safety design team, I am using insights gathered from interviews and contextual inquiry with drug safety specialists, case processors, and safety physicians at Celgene to design an end-to-end pharmacovigilance experience that will manage and interpret adverse event case reports.