Using AI to advance cervical cancer screening for all
OVERVIEW
UCLA Health partnered with Slalom Build to reimagine and rebuild their patient digital experience platform, retiring more than 250 legacy subsites and modernizing their tech stack in the process. Prognosis? A vibrant and healthy future.
SCOPE
Software Engineering, Modern Web Development, Experience Design, Quality Engineering, DevOps,
Agile Coaching
There’s no cure without access to care
Cervical cancer is the fourth most common cancer found in women globally. Medical science has fortunately developed reliable screening methods so that, diagnosed and treated early enough, many patients can be cured. So why are so many people still dying from it?
For many patients, human papillomavirus (HPV) vaccines and the requisite Pap or HPV test for cervical cancer screening are widely available and routine. For others, preventative care visits may be few and far between. So cervical cancer may not be identified until long after symptoms develop—when it’s often too late. And because of that basic inequity of care, hundreds of thousands of mothers, daughters, and loved ones die from a preventable disease.
In fact, of the estimated 342,000 deaths from cervical cancer in 2020, about 90 percent of cases occurred in developing countries. Can a cure really be called a cure when it’s only available to the more fortunate?
Re-examining the partnership procedure
As with any procedure, a pre-op check-in was required. Determining patient needs and how they wanted to engage with the website was paramount. Fortunately, our Product Engineering Methodology (PEM), is designed to answer these important questions.
First, some subsites had the same look and structure as the main site, however they led patients off the domain or to dead-ends that lacked clear navigation and integration with the main site, generating a user experience that could be confusing and frustrating.
Second, there were redundancies among the subsites. We needed to examine subsite content to eliminate duplicate information and pages.
And third, some subsites were owned by doctors or organizations that were no longer active or whose sites lacked current information. We needed to eliminate the inactive sites, and for the others, identify new owners and train them on the new platform to ensure consistency and relevancy.
UCLA Health was intent on creating a more cohesive, navigable experience for their patients.
“AI makes cervical cancer screening so much faster while not losing any accuracy,” affirms Ashley Jarosz, Hologic lab supervisor. “I'm absolutely inspired that this work could further improve patient outcomes.”
Potentially the first automated algorithm of its kind used for cancer screening, this technology is already being deployed in Europe with plans to continue expanding globally.
Michael Quick is excited about what this means for the future of patient care, “The technology that we've developed at Hologic with Genius Digital Diagnostics and Genius AI could propel the World Health Organization's goal of eradicating cervical cancer in our generation.”
And in the meantime, life-saving cancer prevention can become more readily available to people everywhere.
With patients at the heart of the rebuild, this work led to a 12% increase in self-scheduled appointments and a 482% increase in online appointment requests. Plus, the new site shows significant performance gains over the legacy platform, including a 26% increase in visitors, 30% improvement in Time to Interactive, 50% increase in online search results brand visibility, and 71% improvement in Time to First Byte.
The strong partnership between UCLA Health and Slalom Build continues to this day—we’re transitioning legacy subsites on a regular basis and invigorating the new site with more features and functionality, such as faceted search.
Because of our commitment to help clients transition successfully, Slalom Build engineering will continue to partner with the UCLA Health marketing team through 2022, until they are fully prepared to take ownership and maintain the health of their powerful and flexible, easy-to-use, patient-loving platform!
The beginning of digital cytology’s future
In initial tests, the algorithm achieved a level of accuracy that far exceeded expectations. The AI autonomously detected abnormal cells within whole slide pap test images with roughly 70 percent sensitive recall.
We proved that our AI algorithm can learn to replicate what human cytologists see—both accurately and at impressive speeds. And as the algorithm improves over time, the results can only get better.
Michael Quick is excited about what this means for the future of patient care, “The technology that we've developed at Hologic with Genius Digital Diagnostics and Genius AI could propel the World Health Organization's goal of eradicating cervical cancer in our generation.”
And in the meantime, life-saving cancer screening can become more readily available to people everywhere.
Next Up