SHREVEPORT, La., Aug. 22, 2023 /PRNewswire/ — LSUS alumni developed an innovative medical coding chatbot that is designed to reduce time and improve accuracy of medical coding. 

Phillip Kilgore and Keyvan Shahrdar have created an ICD code reference powered by artificial intelligence to help medical professionals sift more quickly through the approximately 9,000 International Classification of Diseases (ICD) codes that describe everything from heart attacks to impacted teeth. 

“People usually go through an insane amount of training just to learn the codes, and then you have these lists you’re referring back to,” said Kilgore, a machine learning expert and a former LSUS computer science instructor. “Take something like cancer – there are potentially hundreds of ICD codes for cancer. 

“But with this tool, you can describe the symptoms, and the bot will produce a list of codes that are related to what you’re describing. This is especially helpful when a patient might have comorbidities that impact the primary diagnosis. Our goal with this software is to provide a reliable tool that enables medical professionals to rapidly access accurate ICD codes, freeing up more time for patient care.” 

The code list can be viewed digitally on the Discord server or imported into an Excel spreadsheet for download. 

Shahrdar, an LSUS computer science instructor and CEO of Shahrdar Enterprises, believes this technology can improve patient outcomes by reducing coding error and enhancing efficiency. 

“This is the first of its kind Discord server that’s able to use a chatbot to return ICD codes,” Shahrdar said. “As an educator and entrepreneur, I’m thrilled to introduce this advanced AI tool to the healthcare community. 

“The intersection of technology and healthcare holds immense potential, and we’re proud to contribute to this evolving landscape. We believe this software will not only support medical professionals but will also inspire students and researchers with its innovative application of AI in healthcare.” 

The medical coding chatbot is built upon the ChatBotCPA framework, a sophisticated conversational process automation (CPA) platform.  

ChatBotCPA harnesses the power of advanced natural language understanding and machine learning to grasp and process user inputs. 

These capabilities allow the chatbot to interpret medical terminologies and context, enabling the chatbot to reference the most relevant ICD codes in response. 

Medical coders typically endure months of training, but the medical coding chatbot could lessen that learning curve and better prepare coders. 

Kilgore, the chief technology officer for Shahrdar Enterprises, experienced the complexity of the medical coding world firsthand as a researcher. 

He worked on projects involving massive databases of cancer diagnoses, which required ICD codes to navigate, to pinpoint specific forms of cancer. 

One major goal of medical research is to identify trends in comorbidities – multiple diseases in a single patient that may or may not act in concert to affect patient health. 

Efficiency in navigating ICD codes could positively impact medical research and insurance underwriting along with improving patient care. 

The chatbot uses the Discord platform, a social media app which allows large communities to access this tool. This free platform could introduce the medical coding chatbot to a wider audience than other paid platforms like Slack. 

The chatbot, which uses advanced natural language processing to discern the coder’s commands, can accurately produce the proper list of codes from which the medical professional can select the appropriate codes. 

“We’ve found that doctors or other medical professionals can more easily and more quickly pick out the correct code from the chatbot list than having somebody navigate the entire list of ICD codes,” Kilgore said. “Patients will want their doctors to accurately summarize the nature of their disease, and this has a research benefit as well because we’re improving the accuracy of how diseases and comorbidities are coded and recorded.” 

Medical professionals wanting to test the medical coding chatbot:  https://discord.gg/x5BXTPqG7U

For more information, visit www.ChatBotCPA.com. 

Matt Vines
Communications Specialist, LSUS Public Relations
Louisiana State University Shreveport
318-797-5000
364067@email4pr.com
http://www.lsus.edu

View original content to download multimedia:https://www.prnewswire.com/news-releases/lsus-alumni-create-medical-coding-chatbot-to-reduce-time-improve-accuracy-301905207.html

SOURCE LSU Shreveport