A Deep-Dive into the HHS Artificial Intelligence Use Cases
Makpar’s Innovation Lab provides insights into how HHS can further augment its overall AI efforts
In January 2021, the Department of Health and Human Services (HHS) launched its Artificial Intelligence (AI) strategy, which aims to broaden and accelerate AI-centered pursuits across all of HHS.
As a result, the agency is prioritizing the application and development of AI and machine learning across common enterprise mission areas, and has already shared several dynamic AI use cases in action.
Mojtaba Heidarysafa, an AI/ML Engineering Intern in the Makpar Innovation Lab, recently shared his insights into how HHS can further bring the following AI use cases to life.
FDA Counterfeit Detection Device Version 5 (CD5): The FDA developed CD5, the latest version of a counterfeit detection device, and is a handheld device that uses LEDs at different wavelengths to help forensic workers to examine FDA-regulated products and detect counterfeit pharmaceuticals.
Makpar Innovation Lab Insights: The use the colorimetric information and tablet images generated by the CD5 to detect differences between authentic and counterfeits products can be augmented by enhanced (binary) image classification solutions – helping to further determine which pharmaceuticals are counterfeit.
CMS Center for Program Integrity (CPI) Fraud Prevention System Models: The goal of this project is to more effectively detect, prevent, and prioritize potential cases of Medicare and Medicaid fraud, waste, and abuse for future investigations.
Makpar Innovation Lab Insights: In addition to using tree-based models and deep learning approaches (in use), this effort can be augmented through the use of anomaly detection, and fraud classification solutions.
FDA Emerging Chemical Hazard Intelligence Platform (ECHIP) and Warp Intelligent Learning EnginE (WILEE): The goal of this project is to leverage AI to better forecast the emergence of a new chemical hazard, by screening and connecting a broader range of data sources for detection of potential chemical signals.
Makpar Innovation Lab Insights: In addition to using AI to develop a horizon-scanning application and intelligent knowledge discovery platform, this offering can be further augmented by using advanced predictive analytics and time-series forecasting solutions.
CMS Rapid Authority to Operate (ATO): The CMS Authority to Operate (ATO) security planning process is a burdensome and lengthy process that requires an average of 500 pages worth of documentation.
Makpar Innovation Lab Insights: In addition to using an automated AI pipeline, consisting of natural language processing (NLP) and supervised and unsupervised machine learning, this effort could be enhanced with additional document and text classification solutions.
CMS OSFLO Help Desk Chatbot: The goal of this project is to assist the OSFLO help desk to more rapidly and effectively respond to employee and contractor questions by automating voice and text responses for general security and badging questions.
Makpar Innovation Lab Insights: In addition to using Google DialogFlo’s natural language understanding model for enhancing responses, this effort could be enhanced by using NLP and state-tracking chatbot solutions.
HRSA Electronic Handbooks (EHB) AI Chatbot: The goal of this project is to develop and deploy the EHB Chatbot, a self-service platform with action-based responses to grantees using regular natural conversational expressions.
Makpar Innovation Lab Insights: In addition to using Google DialogFlo’s natural language understanding model for enhancing responses, this effort could also be enhanced by using NLP and state-tracking chatbot solutions.
OIG Grants Analytics Portal: The goal of this project is to turn grants data into actionable intelligence to support oversight of HHS grants recipients and grant programs.
Makpar Innovation Lab Insights: In addition to developing a deep neural network model was built to extract A-133 single audit finding text from PDFs, this effort could be augmented by using optical character recognition (OCR) systems combined with next-generation visualization solutions.
NIH eRA Internal Referral Module (IRM): The goal of this project is to eliminate referral bottlenecks by automatically referring grant applications to Program Officers once they are received.
Makpar Innovation Lab Insights: In addition to using an AI system to analyze grant applications inputs, this effort could be further augmented by using NLP document classification capabilities.
We would like to thank Mojtaba for sharing his insights with us!
The Makpar Innovation lab is pioneering new ways for helping government to solve critical business problems with data. Learn more about the Makpar Innovation lab here.
If you would like to learn more about how Makpar can help your agency develop the most comprehensive AI solutions for enhancing mission success, please contact us here.