Researchers at the Regenstrief Institute have developed and tested a precise system that can track colonoscopy quality and appropriate procedure intervals, improving health outcomes and lowering overall costs.
The system was tested in the country’s first multiple institution colonoscopy quality measurement study that used natural language processing and reports that proved as accurate as human review and less expensive.
This new system consists of natural language processing through sophisticated software that is capable of extracting meaning from written language. It allows a computer to read and process free text very quickly while at the same time processing reports that gastroenterologists wrote about colon growths’ compositions removed with colonoscopies.
The capacity of a computer to process, understand and excerpt meaning from human language is one of the oldest goals in science fiction, familiar to everyone who has ever watched Star Trek and Captain James T. Kirk having a conversation with the ship’s computer.
Over the last few decades, research has finally allowed computer software to be side by side with fiction since IBM’s Watson computer responded correctly to tricky questions on the TV game show Jeopardy. In this recent research project, Regenstrief and colleagues from the IU School of Medicine assessed the capacity of language processing, adapting it to interpret complex medical information concerning colon cancer detection.
“We found that rapid and inexpensive natural language processing, which utilizes free-text data that was previously unusable for efficient computer-based analysis, was extremely accurate in measuring adenoma detection rate during colonoscopy. The presence of adenomas in the colon is predictive of a patient’s risk of later developing colon cancer, and the detection rate has been identified as the critical measure of a high-quality endoscopist, the specialist who performs colonoscopy,” said Dr. Timothy Imler in a press release, the professor who conducted the study.
This evaluation concluded that the software could perform the same job as humans in interpreting and correlating reports in a much faster and economic way.
“Confirming that humans and the computer had similar assessments on procedures performed at facilities across the country gives us a powerful, scalable tool to assess quality and determine the appropriate interval between colonoscopies based on the procedure findings. Natural language processing will enable comparison of adenoma detection rates across populations to determine geographic, racial, ethnic or gender disparities. It also could be used across health systems or colonoscopy centers or doctors to enable referrers or patients themselves to make informed decisions about where they wish to go for a colonoscopy,” concluded Dr. Imler.
The post Natural Language Processing Software Could Improve Colon Cancer Detection appeared first on Colon Cancer News Today.