Geisinger, Medial EarlySign find ca algorithm


DANVILLE, Pennsylvania – A machine learning algorithm detected potential signs of colorectal cancer (CRC) in patients identified as high risk who had missed a routine colonoscopy, according to a new study by Geisinger and Medial EarlySign.

The findings, published this month in NEJM Catalyst Innovations in Care Deliverypresent a non-invasive method to increase screening among people at risk of having CRC.

Despite evidence of the benefits of regular CRC screening and significant efforts by providers and healthcare systems to increase screenings, approximately 32% of eligible-age adults in the United States do not meet current CRC screening guidelines, according to the National Cancer Institute. Serious illness and death from CRC can be prevented if asymptomatic polyps and other early-stage cancers are detected and treated early.

In the study, Geisinger identified a group of 25,610 patients who were overdue for CRC screening and used a machine learning algorithm to flag those most at risk of developing cancer. The algorithm, developed by EarlySign, identified patients as high risk by analyzing age, gender and a recent outpatient complete blood count (CBC). A nurse then called the patients to inform them of their risk and offered to schedule a colonoscopy.

Of the patients flagged as high risk, 68% were scheduled for colonoscopy, and of these, approximately 70% had a significant outcome.

“When carefully implemented and supported by healthcare providers, machine learning can be an inexpensive and noninvasive complement to other colorectal cancer screening efforts,” said Keith Boell, DO, director Quality for Population Initiatives at Geisinger and co-author of the study. “This technology can act as a safety net, potentially preventing missed or delayed diagnosis in some patients who may already be showing undiagnosed signs of disease.”

“Our partnership with Geisinger has focused on tackling the devastating impact of CRC with predictive algorithms that can impact early detection, coupled with integration into clinical workflows that lead to a personalized approach to care that engages patients in prevention and treatment,” said Ori Geva, Co-founder and CEO of EarlySign. “The inclusion of our joint study with Geisinger in NEJM Catalyst Innovations in Care Delivery is a great honor for our team, and we are grateful to all EarlySign and Geisinger co-authors and project teams for their achievements in research and quality results.


About Geisinger
Geisinger is committed to facilitating better health for the more than one million people it serves. Founded more than 100 years ago by Abigail Geisinger, the system now includes 10 hospital campuses, a health plan with more than half a million members, a research institute and the Geisinger Commonwealth School of Medicine. With nearly 24,000 employees and more than 1,600 physicians employed, Geisinger boosts the economies of his hometown in Pennsylvania by billions of dollars a year. Learn more about or join us on Facebook, instagram, LinkedIn and Twitter.

About Medial EarlySign™

Medial EarlySign helps healthcare players keep patients healthier, longer with software solutions that derive actionable, personalized clinical insights from real healthcare data. With a focus on improving outcomes and reducing costs, EarlySign’s AlgoMarkers and predictive solutions enable early detection of complications from critical illness and help customers more accurately identify and prioritize patients for multiple conditions for interventions to stop or prevent serious complications early in the disease. . The company’s machine learning platform and development environment enables rapid, high-quality development of custom models and prebuilt models backed by peer-reviewed research published by healthcare organizations and hospitals internationally famous. Founded in 2013, Medial EarlySign is headquartered in Tel Aviv, Israel. For more information, please visit:

Warning: AAAS and EurekAlert! are not responsible for the accuracy of press releases posted on EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.


Comments are closed.