New Performance Analysis Research Shows Medial EarlySign's Machine Learning Flagging System Helps Identify Individuals at High Risk for Colorectal Cancer
KFAR MALAL, Israel, March 7, 2017 /PRNewswire/ --
Medial EarlySign (http://www.earlysign.com), a developer of machine learning tools for data driven medicine, today announced new research with one of Israel's leading integrated delivery networks, Maccabi Health Services (MHS), that found the company's MeScore (hereafter 'ColonFlag') tool can potentially be used to identify individuals who are at 10 to 20 times increased risk of harboring an occult colorectal cancer (CRC).
The peer-reviewed study, published by PLOS ONE, sought to determine if information contained in complete blood count (CBC) reports could be processed automatically and used to predict the presence of occult CRC in the setting of a large health services plan. The researchers reviewed Maccabi's CBC reports for 112,584 study subjects, of whom 133 were diagnosed with CRC in 2008, and analyzed these with Medial EarlySign's tool.
"Leveraging our advanced EMR database, the results of this performance analysis indicate that ColonFlag can help identify individuals in the population who should be sent for further evaluation for CRC, including those with no clinical signs or symptoms of CRC," said Revital Kariv, Chief Gastroenterologist of Maccabi Health Services.
This is the first time that ColonFlag has been evaluated on previously collected prospective data. When using a 1% cut-off, the odds ratio for colorectal cancer was 21.9, while a 3% cut-off yielded an odds ratio of 10.9, significantly higher than traditional indicators. Current clinical indicators to estimate the risk of colorectal cancer include family history, lifestyle factors such as smoking and physical exercise, obesity, diabetes and laboratory results. These indicators influence a doctor's decision about which screening tool to use, whether it be a fecal occult blood test (FOBT) or colonoscopy. For some of the indicators, research shows 1.5-2.5 times higher risk than the average population.
Reviewing the charts of patients diagnosed with colorectal cancer, the researchers found that for the majority of individuals with cancer, CRC was not suspected at the time of the blood draw. The study also indicated that frequent use of anticoagulants, presence of other GI pathologies, and non-GI malignancies were associated with false positive ColonFlags.
"Our research opens the door to further insights relating to the successful performance of ColonFlag and an understanding of how false positives may bear other significant clinical implications," said Dr. Ran Goshen, Chief Medical Officer of Medial EarlySign.
Maccabi operates six regional centers, including 150 branches and clinics, providing full access to Maccabi Healthcare services for all members throughout the country. The organization invests significant resources in research and the acquisition of innovative and managerial systems which promote its medical and administrative organization.
View the full study here.
ColonFlag is not yet cleared by the FDA for commercial use in the USA.
About Medial EarlySign
Medial EarlySign's advanced algorithm platform accurately detects the likelihood of disease for subpopulations using basic medical information, such as blood test results, and other EMR data. The company's predictive tools provide physicians with actionable insight, while providing insurers with effective models to flag and focus on patients at risk, helping to prioritize resources, save money and improve care. Medial EarlySign's platform addresses numerous potential clinical outcomes, including cancers, diabetes and other life-threatening illnesses. Founded in 2009, the company is headquartered in Kfar Malal, Israel. For more information, please visit http://www.earlysign.com.
Follow Medial EarlySign on LinkedIn: Medial EarlySign and Twitter: @MedialEarlySign
Media Relations Contact: Ellie Hanson Finn Partners +1-929-222-8006 email@example.com
SOURCE Medial EarlySign