New tool shows promise in detecting colon cancer earlier
ColonFlag uses blood test results and EHR data to identify high-risk patients, says Matt Hornbrook.
A new computational tool may help physicians identify patients with the highest risk of having undiagnosed colorectal cancer while the disease is still in an early stage.
The tool, called ColonFlag, was developed by Medial EarlySign, an artificial intelligence and machine learning vendor. It predicts and stratifies individuals at high risk by analyzing routine blood test results and electronic health records data. The company also is developing additional disease predictors for other types of cancers, diabetes and other life-threatening conditions.
In a study published in the journal Digestive Diseases and Sciences, ColonFlag was able to identify risk for colorectal cancer as much as a year earlier than conventional diagnostic practices, study authors say.
Also See: Utah researchers to leverage clinical decision support for advanced cancer screening
Researchers culled data from blood samples and demographic information of 17,095 patients served by Kaiser Permanente’s Northwest region, which included a random sample of 900 patients with colorectal cancer. ColonFlag then analyzed complete blood count results and factored in age and gender to create an overall risk score for each patient.
Colorectal cancer is the second highest cause of all cancer-related deaths in the United States, says Mark Hornbrook, lead researcher for the study.
“Early screening significantly improves the survival rate,” he adds. “The ability to identify people at high risk for colon cancer and refer them for further testing could help to reduce mortality and prove integral to reducing the overall colorectal cancer burden.”
Cases of colorectal cancer identified by ColonFlag were compared with cases detected by low hemoglobin levels across two time periods: 0 to 180 days and 181 to 360 days before diagnosis. In the 180-day time period, ColonFlag showed an improvement of 34 percent and 36 percent in identifying cases compared with low hemoglobin levels in patients age 50 to 75 and 40 to 89, respectively.
In the 181- to 360-day window, detection was 47 percent higher for the 50-to-75 age group and 84 percent higher in the 40-to-89 age group.
Additional analysis found that ColonFlag particularly excelled at detecting tumors in the cecum (a pouch at the beginning of the large intestine) and ascending colon.
The full study is available for purchase here.
The tool, called ColonFlag, was developed by Medial EarlySign, an artificial intelligence and machine learning vendor. It predicts and stratifies individuals at high risk by analyzing routine blood test results and electronic health records data. The company also is developing additional disease predictors for other types of cancers, diabetes and other life-threatening conditions.
In a study published in the journal Digestive Diseases and Sciences, ColonFlag was able to identify risk for colorectal cancer as much as a year earlier than conventional diagnostic practices, study authors say.
Also See: Utah researchers to leverage clinical decision support for advanced cancer screeningResearchers culled data from blood samples and demographic information of 17,095 patients served by Kaiser Permanente’s Northwest region, which included a random sample of 900 patients with colorectal cancer. ColonFlag then analyzed complete blood count results and factored in age and gender to create an overall risk score for each patient.
Colorectal cancer is the second highest cause of all cancer-related deaths in the United States, says Mark Hornbrook, lead researcher for the study.
“Early screening significantly improves the survival rate,” he adds. “The ability to identify people at high risk for colon cancer and refer them for further testing could help to reduce mortality and prove integral to reducing the overall colorectal cancer burden.”
Cases of colorectal cancer identified by ColonFlag were compared with cases detected by low hemoglobin levels across two time periods: 0 to 180 days and 181 to 360 days before diagnosis. In the 180-day time period, ColonFlag showed an improvement of 34 percent and 36 percent in identifying cases compared with low hemoglobin levels in patients age 50 to 75 and 40 to 89, respectively.
In the 181- to 360-day window, detection was 47 percent higher for the 50-to-75 age group and 84 percent higher in the 40-to-89 age group.
Additional analysis found that ColonFlag particularly excelled at detecting tumors in the cecum (a pouch at the beginning of the large intestine) and ascending colon.
The full study is available for purchase here.
More for you
Loading data for hdm_tax_topic #better-outcomes...