RamaOnHealthcare July 10, 2022

Today, RamaOnHealthcare talks with Ted Willich, CEO, NLP Logix, an artificial intelligence and machine learning solutions company.

Ted Willich, Cofounder and CEO of NLP Logix

Ted Willich, Cofounder and CEO of NLP Logix

RamaOnHealthcare (ROH): Can you provide a brief history of NLP Logix?

Ted Willich (TW): NLP Logix was founded in 2011 with the mission of delivering world-class machine learning and artificial intelligence solutions across many industries, including healthcare. Our 80-plus member team has made the Inc5000 fastest growing companies in the United States for three years in a row and we are passionate about solving our customers gnarly data challenges.

ROH: Tell us how you utilize Artificial Intelligence (AI) and Robotic Process Automation (RPA) within healthcare?

TW: At NLP Logix, we believe that one of the biggest areas where we can positively impact the delivery of healthcare is by bringing automation tools to the workforce. In fact, in the year 2022, automation through machine learning/artificial intelligence should be considered for any data-intensive and repetitive task. A great example of this are the tasks performed to code medical records for billing purposes. Think about all the clicks, searching, copying, pasting and note taking that happens with every chart being coded. Now, extrapolate those over hundreds or thousands of charts a month or year, and you can quickly see that the combination of using bots (RPA) and data capture techniques can greatly reduce this time spent on menial tasks.

…automation through machine learning/artificial intelligence should be considered for any data-intensive and repetitive task.

ROH: Tell us about your Predictive Models in healthcare. What is its role in data?

TW: We have developed and deployed predictive models for a good number of healthcare clients with most of the solutions being focused on different aspects of the revenue cycle. Specifically, we trained models to estimate the probability a patient will pay a bill as well as the probability a patient will show up for an appointment. The key to these models is the data and having enough in a workable format to be able to train the computers to see the patterns embedded within the data.

The key to these models is the data and having enough in a workable format to be able to train the computers to see the patterns embedded within the data.

ROH: Please explain how your Sentiment Analysis is being used in healthcare.

TW: Sentiment Analysis, is technology that estimates the sentiment of written or spoken words. For example, a patient responding to a satisfaction survey may say “Nurse Jones was very caring and attentive during my stay, but Dr. Smith hardly spent any time with me, and I do not think he prescribed the right medicine for me.” These are two very different sentiments being sent in the same message. Sentiment analysis analyzes the messages and alerts the practice that a caregiver was recognized for their superior care as well as the need for patient follow-up to confirm medication compliance.

One of the exciting things about being in the AI/Machine Learning business is you get to see the technology evolve and advance through the years. Sentiment analysis is one of those where recent advances in audio based deep learning models include tone, rhythm, etc. to enhance the predictions. Most sentiment analysis solutions work for both aggregate statistics (average sentiment by division) as well as detailed analysis (all the comments that have sentiment under a specific value) and there are routing processes based on each. The key is to be able to take the technology and deliver business value.

The key is to be able to take the technology and deliver business value.

ROH: Given the diverse industries you work with, what is your process to best determine and tailor the product to the customer needs?

TW: A good friend of ours once told us that “prescription before diagnosis equals malpractice” and we really took that to heart when it comes to delivering our custom machine learning/automation solutions.

We developed our 10Q Assessment, where we take an inventory of our customer’s data, workflows, and ultimate results they want to achieve. From there, we provide a very detailed plan on how machine learning can be used to accomplish those results. For example, our 10Q AI Assessment provides a complete analysis by our team of Data Scientists, Mathematicians, and Software Engineers. At the end of an 6-8 week Assessment, we produce a comprehensive roadmap for the future of your data – because your data is one of your most valuable assets!

…our 10Q AI Assessment provides a complete analysis by our team of Data Scientists, Mathematicians, and Software Engineers.

ROH: Can you provide us an example of application of your work within a healthcare system?

TW: One example would be the Florida Poison Information Center Network. They do amazing work helping to reduce hospitalizations and saving lives through their expertise in recommending treatment for specific poisoning exposures. Over the past eight years we have developed several machine learning models to estimate the number and types of poison exposures, as well as monitor for trends in certain poisons that are not visible to the human eye. These models are constantly monitored by the team at NLP Logix and when necessary, we are retrained on the latest data to ensure the models are as accurate as possible. The work has been so effective that it has contributed to us growing to include 14 of the 54 Poison Control Centers in the United States.

ROH – AI and healthcare are both rapidly changing industries. What is your method to keep up and keep your team on track?

TW: AI and machine learning are constantly changing and the way we keep our team up to speed is the same approach we use when delivering our solutions. It is simply that we work and learn as a team. In fact, we take “team” so seriously that our trademarked saying is “Data Science is a Team Sport”. We work in teams of machine learning engineers, math and statistics, software engineers, support, and maintenance teams. Because of this approach, we can focus our team members on the emerging technologies in their specific fields. We also do quite a bit of cross training, so we are constantly learning.

About Ted Willich

Ted Willich is Co-Founder/CEO of NLP Logix, an artificial intelligence and machine learning solutions company. It was founded in November 2011 and provides automation solutions by training computers to do repetitive tasks done by humans. Today, NLP Logix is a team of over eighty professionals from multiple technology specialties including software engineering, mathematics, statistics, data analysts and machine learning engineers. Our clients represent healthcare, finance, staffing, defense, energy, transportation, and education. Previously, Ted served as a United States Marine Corps Infantry Officer from 1988 until 1992. He has a bachelor’s degree in history & leadership studies, and has lived with his wife, Julie in Florida for 19 years.

 
Topics: AI (Artificial Intelligence), Interview / Q&A, Technology, Trends
Q&A - The Growing Importance of Pharma Services with Colin Aldridge, Director at TripleTree
Q&A - Dr. Guy M. Nehrenz, Interim Dean and Professor
Q&A - Mayo Clinic Platform by Dr. John D. Halamka, President
Thought Leadership Summary - Importance of Data, Innovation & More in Healthcare
Q&A - Dave Bennett, CEO of pCare - Digital Hospital of the Future