Enhancing the usability of nuclear medicine imaging device (PET-CT)
The Challenge
A major healthcare equipment manufacturer wanted to enhance the usability - application software, as well as at department ecosystem level, of its nuclear medicine imaging machine (PET-CT) which is used to diagnose and stage cancer.
Approach
This research was planned across 4 countries with regional and cultural differences - Turkey, Egypt, China and India. The objective was to unearth the various local and cultural nuances that come into play in high-stress environments as the Nuclear Medicine Imaging Department. Through an understanding of the pain points and stress-inducing tasks, we identified opportunities for improving the usability of nuclear medicine (PET CT) imaging device. This, in turn, led to the enhancement of the care-provider experience and enabled them to provide care to more patients per day.
Outcomes
The deliverables of the study were:
-
Identified key areas of improvement in the PET-CT scan ecosystem - priority touchpoints and key opportunities categorized across 2 segments i.e. user interface, and ecosystem interventions.
-
Detailed mapping of the task-flows. Deviations from the standard workflow gave insight into users' mental models and led to the redesign of PET-CT workflow.
-
Effort and complexity mapping of each touchpoint in the workflow leading to insights for simplification.
-
Perception, Cognition, and Action (PCA) analysis of the entire workflow. This resulted in the discovery of new considerations and lost opportunities that have the potential to enhance the PET CT imaging process.
-
A book of UX/UI recommendations was created for the PET CT software on the basis of heuristic evaluation and insights from the user study.
Year
2019
Project Duration
8 months
Role
Senior Researcher,
Project Coordinator
Teammates
Devyani Lal
Mangesh Ashrit
Swati Krishnan
PROCESS
The research plan was designed with several iterations of discovery and reflection to deepen the insights.
DIVING INTO A COMPLEX DOMAIN
The first step for the team was to understand the context of the nuclear medicine (PET-CT) department through in-depth desktop research.
HEURISTIC EVALUATION
We evaluated the current PET-CT imaging software via Heuristic principles in order to assess the usability of the software. We then created a primary research plan which focused equally on observations and interviews.
PILOT STUDY
A pilot study was planned to understand the real-world practices within the NM imaging department. This was a testing ground for the questionnaires, observation sheets and process mapping frameworks we had designed. The pilot study helped us make our field tools more robust and allowed us to improvise on data capturing techniques.
FIELD VISITS
We visited both hospitals and diagnostic centers, across 4 countries – India, China, Turkey and Egypt. Multiple personas were interviewed at the NM department; the primary persona being the NM Technologist - responsible for conducting the PET-CT scan. For a deeper insight into the day-to-day routine of NM Technologists, NM Nurse and NM Physician, we used the following research methods:
SHADOWING
We tagged along with the NM Team at every site, closely observing them for 3 hours. We tracked their positions and movements within the NM Department throughout the day. Recording their movement helped us define zones of high stress and different behaviours.
CONTEXTUAL INQUIRIES
Next, we started interacting actively with the NM Technologists while they performed their tasks. This helped us gain highly detailed information about their perception, concerns and apprehensions regarding the PET CT software.
BEHAVIOURAL MAPPING
We observed the interactions and contexts of the NM technologist, patients and other personas within the NM Department. By mapping their behaviour, emotions and stress factors we built an understanding of their latent and tactic needs.
TIME STAMPING
We tracked the amount of time taken to complete a PET-CT scan for every patient throughout the day. Measuring time per screen and activity in addition to the duration of the task, enabled us to isolate the steps that were taking the most time or adding unnecessary seconds.
To see the hidden, we visualized the data sets in multiple ways - journey map, emotion map, frequency v/s duration plot, ecosystem flow map, scan workflow map and nested workflows. This gave us rich insights and macro-level pain points in the ecosystem. Through this, we identified the touchpoints that needed to be simplified or eliminated in order to ease out the workload of the NM department professionals.
For nuanced insights, we analyzed the data collected through various frameworks and models - PCA analysis, effort and complexity analysis, cross-cultural comparisons, heat map analysis and heuristical analysis.
Here are some high level yet really interesting insights (and conundrums) that were identified after analyzing the data through the above-mentioned techniques.
Inconsistencies in adherence to regulations
-
While multispeciality hospitals across the various countries visited adhered to regulations stringently, we observed a lot of inconsistencies at the diagnostic centers, especially the ones situated in smaller towns. This had huge repercussions on the patients' health.
-
We found, that the inconsistencies were primarily because the diagnostic centers were understaffed to manage costs, resulting in the NM Technologist being overworked.
-
We proposed simple feature additions and solutions that would assist the Technologists in their routine tasks and thus enable them to match regulation standards.
Technologists work with a fear of exposure to radiation
-
NM Technologist is a high-risk profession as the technologist is exposed to nuclear radiation on a daily basis. Despite the use of protective gear, we found this to be a critical concern for most technologists.
-
Interestingly, many NM Technologists had created their own workarounds to avoid radiation exposure, but in the process, they were drastically deviating from the PET CT task flow. They were also unaware of the impact this had on the scan image quality, and consequently on the patient diagnosis.
-
We improvised the workflow and identified opportunity areas to reduce the deviation from process and the exposure to radiation. This also turned out to be the key differentiating factor from the competition.
What is a novelty for the expert is also a burden for the novice
-
Though the scanning software was loaded with state of the art features to run efficient and quick PET CT scans, most NM Technologists were not trained well enough to use these features. As a result of this, they were executing all the scans on a "template" that was set by the field engineer.
-
On the other hand, these were the very features that piqued the interest of Cancer Researchers and highly specialized Nuclear Medicine Technologists.
-
We ideated on solutions to bridge this gap through UX/UI interventions. The result was a completely new approach to user interaction and workflows.
Most Technologists receive training on-the-job
-
A lot of new NM Technologists in the developing countries are trained on-the-job. Due to this, they have limited understanding of the system and are completely dependent on the knowledge and time spared by their senior. This was a massive opportunity.
All the findings and recommendations were categorically arranged into a research book. We created a rating scale to identify recommendations that would be highly impactful in achieving the objective.
Additionally, the opportunity areas were also classified based on the kind of intervention.
REFLECTION:
Being in the healthcare domain, this project involved understanding a highly complex ecosystem with an extremely low tolerance for error. Planning and executing ethnographic research in this environment was the most challenging aspect. We prepared and practiced 'researcher journeys' before executing the field research in order to map the activities within the NM Department in detail, without hindering the processes.
This project spanned multiple realms, i.e. software application user experience, and the operational ecosystem of the NM Department. Hence, it became crucial to define clear boundaries of the research in order to deliver outputs within the agreed timeframe. In retrospect, this was extremely important as it would've been fairly easy for the team to deviate onto low relevance tangents, given the expansiveness of the research.
This project introduced me to a highly specialized research domain and was a great learning experience. It was very interesting to understand how cultural nuances play a role even in seemingly very technical and by-the-book domains.