Reimagining Lung Cancer
This is a healthcare hackathon project hosted by MIT. We were challenged to tackle lung cancer problems. During this hackathon, I teamed up with a physician, a pharmacy student, a data scientist and two software engineers. We built an empowering solution for lung cancer patients to have an in-home diagnostic technology as well as a navigator for outreach to physician specialists and clinical trial options worldwide. Now we are continue working on this project and restarting from the research phase.
Time Span: November 2018 – Present
Time Project: 6 people (1 physician, 1 pharmacy student, 1 data scientist, 2 software engineers and me as designer and business consultant)
There are many perspectives to tackle lung cancer problems, such as smoking prevention, detect pre-symptomatic cancer, or encourage health behaviors with systems, workflows and data strategies to stop smoking or detect lung cancer early. After conducing some domain research, including interviewing cancer patients, consulting doctors and conducting secondary research. We chose to focus on streamlining the process of seeking clinical trials for patients.
Throughout the whole project, we always keep the healthcare ecosystem in mind. There are many stakeholders involved in each cancer patient treatment.
I talked to people whose family member has experienced lung cancer and doctors to understand the whole process of treating lung cancer. The issues mainly occur when the patients want to confirm their diagnosis results with other sources, extracting and transferring medial records, as well as finding clinical trials while competing with time.
Fragmented healthcare information
Lack of medical literacy
Slow transfer of medical records
Limited TIME for terminal patients
Product Opportunity Gaps
Clinical Trial Search Engine
Smart Diagnosis & Analysis
Standard EHR & Cloud Storage
Clinical Search Tools
Patients use short terms length terms (single or two word phrases) to search clinical trials. However, they might use terms that are not appropriate for their needs and get non-specific results.
Patel CO, et al. AMIA Annual Symposium proceedings AMIA Symposium. 2010;2010:597-601.
Abel, et al. (2015). Cancer Epidemiol Biomarkers Prev 24(10): 1629-1631.
ClinicalTrials.gov is the primary sources of information which receives hundreds of thousands of visitors every month.
But the problems with ClinicalTrials.gov are the followings:
- Hard to navigate for those with low health literacy
- Contain technical information
- Does not allow for specific searching based on eligibility criteria
The patients will upload their screening results and consent us to extract medical records from the hospitals. Using image processing technology, it will provide the diagnosis results. Combined with the natural language processing technology, it will match the patients with clinical trials tailored to their conditions and needs.
Upload CT scans
Transfer medial records
Provide diagnosis results
Match clinical trials