During these times of pandemic, it's important for us to lend as much hand as we can to our frontliners especially those working in healthcare. Businesses closed, hospitals overwhelmed, the world seemed to stop. It asked for heroes, nurses and doctors and paramedics. But as they fought one front another appeared as insidious as the virus itself - testing, paperwork and supply chain issues, creating barriers between healthcare worker and the patient. Fortunately, we have reinforcements - software robots (UiPath) that's made more intelligent using natural language understanding capabilities of Expert.ai. Imagine the possibilities of having a robot with a healthcare domain expertise helping clear a 160-day registration backlog in a single day, provide more accurate clinical review results in 1/3 time, significantly reduce screening time of healthcare workers to four hours from five days, etc.
When you combine the world's leading RPA (Robotic Process Automation) provider, UiPath, with the best in class Natural Language Understanding platform, Expert.ai - possibilities for advanced AI solutions and truly intelligent robots that can augment human capabilities - are endless!
⏩ What it does
Automated case administration (on EMR systems) and generation of clinical summary for in-patient cases by intelligently suggesting ICD10 CM diagnosis codes and extracting associated verbiage (using custom NLP model built with Expert.ai Discovery API) from unstructured medical records while seamlessly handing off to nurse for final assessment in order to perform upstream and downstream processes and validations fully autonomously.
🛠 How it was built
Built using UiPath Studio and Expert System Custom Activities/Connectors to interact with the custom NLP model via the service endpoint and annotation plan for medical record diagnosis extraction. The integration of Expert.ai and UiPath allows the automation of tasks that require text analytics to extract information from unstructured documents and use it in a robot - thereby enabling human comprehension & insight on everyone's beloved software robots! This integration helps enterprises automate repetitive business processes based on large volumes of unstructured data—text documents, email, customer interactions, call notes etc., reducing errors, improving efficiency and increasing scalability of operations.
⛈ Challenges ran into
Brainstorming and building the workflows to simulate an end to end clinical summary process. Making sure that the clinical summary form is as usable and as comprehensive as possible for the clinicians/nurses make the most out of the automation.
🥇 Accomplishments that we're proud of
Being able to create a functioning automation which can serve as a basis for other use cases within healthcare space, freeing up medical practitioners with their precious hours from these mundane tasks.
Log in or sign up for Devpost to join the conversation.