Democratize Healthcare Insight Generation Through GenAI

Komodo’s NLP-powered AI makes healthcare insights accessible to everyone and enables faster adoption of MapLab across the enterprise. 

GenAI and NLP are revolutionizing healthcare interactions by enabling more users to engage with data, applications, and platforms in a more natural and intuitive way. 

At Komodo, we’re excited to announce that we’ve developed and embedded an NLP-based generative AI assistant — MapAI™ — in our MapLab™ enterprise platform, empowering pharmaceutical organizations like never before.  

In essence, MapAI puts insight generation and data analytics into the hands of every team member, regardless of technical skill level. Individuals who’ve often found themselves waiting weeks or even months for data insights to be delivered (e.g., brand leads, Medical Affairs managers, clinical researchers) can use MapAI to quickly get the answers and direction they need, ultimately accelerating decision-making, the product lifecycle, and commercial success.

With MapAI, users can ask simple questions using layman’s English and quickly receive data insights and graphic visualizations. Cohorts and codesets can be saved and reused in deeper analyses. 

How Does It Work?
Our AI assistant leverages the power of NLP and GenAI to provide personalized and informative responses to healthcare queries. By employing an intelligence agent to understand the intent behind each question, our system can effectively delegate tasks to specialized GenAI agents capable of querying our extensive Healthcare Map™. 

We experimented with several foundational models, including Llama 3.1, Mistral 7B, and Phi-3, to identify the most suitable options for our specific use cases. Our API-first approach allows for rapid agent development and ensures seamless integration with our platform’s comprehensive set of APIs, which manage user medical definitions, resource libraries, and access to the industry’s most comprehensive source of patient-centric insights. To streamline the deployment and orchestration of these models, we’ve implemented frameworks like LangChain and LangGraph. To stay at the forefront of generative AI advancements, we’ve automated the entire LLM lifecycle, enabling the adoption of future foundational models, orchestration, and training strategies.  

Getting Started
MapAI’s chat-style interface and self-service functionality encourage every team member to explore disease or therapeutic areas, test hypotheses, and obtain directional insights using our Healthcare Map. 

Simple questions and prompts posed in layman’s English start the exploration; for example:  

• What are the diagnosis codes for NSCLC?
• How many patients were diagnosed with NSCLC in 2023?
• What is the patient volume distribution by zip3? 
• Show me the patient breakdown by race and ethnicity. 
• What is the patient breakdown by payer? 
• Show me the patient volume breakdown by HCP specialty.
• Who are the top 10 HCPs by patient volume in California?
• Build me a drug market basket for NSCLC.
• How many NSCLC patients were treated with nivolumab in 2023?

Galvanize the Enterprise
MapAI is a feature of our MapLab platform, which was built to accommodate the varying skill levels of teams across the product lifecycle — from those with no technical or analytics experience to the most seasoned data scientists. By using one platform and data source to support myriad use cases, organizations gain consistency in insights and increase efficiency as they reduce the burden of managing multiple vendors and contracts. 

MapAI can be accessed via: 

• MapExplorer™ – A no-code workflow that lets users generate high-level insights on a disease or therapeutic area (e.g., patient population, HCOs, HCPs, payers) to gauge the comprehensiveness and accuracy of our Healthcare Map.
• MapView™ – A no-code workflow that enables users to take the insights they create using MapAI (e.g., saved cohort definitions and codesets) and leverage prebuilt template dashboards to conduct more complex analyses. 
• MapLab Enterprise (Coming soon!) – A solution that includes access to no-code, low-code, and high-code environments to support all teams, including advanced users who wish to build their own algorithms and models. 

Are you ready to eliminate the wait and transform every team member into a data analyst? Schedule your personal demo of MapAI today!

If you’d like to learn more about MapAI, MapExplorer, and MapView, register for our Oct. 10 webinar!

To see more articles like this, follow Komodo Health on X, LinkedIn, or YouTube, and visit Insights on our website.