Moderna - Building AI to Generate Targeted and Actionable Medical Insights

Options
danaja
danaja Registered Posts: 1

Team members:

Dan Aja, Director of Medical Affairs, with:

  • Marktus Atanga
  • Jordan Alcott

Country: United States

Organization: Moderna

In over 10 years since its inception, Moderna has transformed from a research-stage company advancing programs in the field of messenger RNA (mRNA), to an enterprise with a diverse clinical portfolio of vaccines and therapeutics across seven modalities, a broad intellectual property portfolio and integrated manufacturing facilities that allow for rapid clinical and commercial production at scale. Moderna maintains alliances with a broad range of domestic and overseas government and commercial collaborators, which has allowed for the pursuit of both groundbreaking science and rapid scaling of manufacturing. Most recently, Moderna's capabilities have come together to allow the authorized use and approval of one of the earliest and most effective vaccines against the COVID-19 pandemic.

Moderna's mRNA platform builds on continuous advances in basic and applied mRNA science, delivery technology and manufacturing, and has allowed the development of therapeutics and vaccines for infectious diseases, immuno-oncology, rare diseases, cardiovascular diseases, and auto-immune diseases. Moderna has been named a top biopharmaceutical employer by Science for the past eight years. To learn more, visit www.modernatx.com.

Awards Categories:

  • Best Acceleration Use Case
  • Best Moonshot Use Case
  • Best MLOps Use Case

Business Challenge:

Medical insights collected from healthcare providers play a crucial role in combating infectious diseases such as COVID-19, RSV and influenza. Timely insights are essential as they assist Moderna to identify knowledge gaps in specific therapeutic areas, enabling targeted interventions and educational initiatives to enhance the overall response to these diseases.

Moderna receives on average 200 to 300 monthly entries in the form of free text, from interactions with doctors, discussions on medical results, feedback from the field, etc. This manual data collection poses many challenges, being time-consuming, inefficient, and difficult to scale in a timely manner. Additionally, the rapid advancements in mRNA science create gaps in scientific data, learning, and clinical knowledge.

Consequently, medical insights data is already outdated when the team is able to analyze and summarize it, which induces a pressing need for timely and actionable insights aligned with Medical Affairs (MA) strategy. Finding more efficient and effective methods to collect and analyze structured and unstructured data is vital for supporting the fight against infectious diseases.

The collaboration between a pharmacist within the Medical Affairs team and data engineers from the Advanced Analytics & Data Automation team aimed to enhance the efficiency of generative medical insight analysis. This unique combination of expertise brings together pharmaceutical knowledge and data analytics capabilities to drive improvements in medical insights generation. By leveraging their respective skills, this collaboration sought to optimize the analysis process and generate valuable insights more efficiently.

Business Solution:

Moderna is currently developing an on-demand AI model called "Nitro" to analyze collected medical insights and sentiments from diverse and unrelated data sources. This AI model is designed to align healthcare providers (HCPs) with Moderna Medical Affairs strategy and has the potential to be scaled globally.

Nitro takes the data in and performs Natural Language Processing (NLP) for producing summaries, using a combination of recipes, plugins and Python code in Dataiku. Sentiment analysis is then applied to produce insights, in the form of trends. Dataiku finally feeds the data to Tableau, where interactive charts are updated to show sentiment trends over time.

By leveraging Nitro, Moderna is able to extract valuable insights from various data sources, enabling the generation of targeted and actionable medical insights. This technology is instrumental in supporting Moderna's US and global Medical Affairs efforts to advance science.

Day-to-day Change:

Nitro's capabilities allow Moderna to respond rapidly to shifting sentiments from healthcare providers, enabling Moderna to leverage their insights in a more timely manner. The speed of analysis and generation of insights is significantly improved, reducing the time frame from months to days. This agility helps Moderna stay ahead of the curve, adapt its strategies and provide value to HCPs.

Overall, Nitro's capabilities not only enhance Moderna's ability to react quickly to changing sentiments and leverage healthcare providers' insights, but also optimize resource allocation within the Data Science team, leading to more efficient and effective operations.

Business Area Enhanced: Analytics

Use Case Stage: In Production

Value Generated:

The Nitro model significantly reduces the inefficiencies, time and effort involved to manually manipulate data from multiple sources into actionable insights:

  • Efficiency:
    • By automating the workflow in Dataiku, the team at Moderna has achieved significant time savings. Previously, the manual process consumed approximately ten hours per month. However, with the implementation of automation, the team is now able to accomplish the same tasks in a fraction of the time, resulting in a opportunity cost savings of approximately forty hours per month.
  • Speed:
    • Running the model on a weekly basis is highly advantageous for Moderna. By doing so, the team is able to identify and uncover impactful insights and knowledge gaps up to three weeks ahead of time. This early detection and awareness of insights and gaps in knowledge are crucial for improving decision-making processes, alignment with scientific objectives while addressing HCP gaps in knowledge.
  • New insights:
    • Through the implementation of sentiment analysis, Moderna is able to explore previously unexplored raw feedback data, resulting in the discovery of new insights. This has been particularly valuable during the COVID-19 pandemic, where the analysis of hundreds of feedback pieces on vaccine hesitancy has provided concise and significant insights for informed decision-making and strategic planning by senior leadership.
  • Reusability:
    • The implementation of Nitro has led to the emergence of numerous new NLP use cases from various departments within Moderna. This project holds great potential for exponential long-term value, as it can be easily applied to handle larger volumes of structured and unstructured data across different domains. The versatility of Nitro allows for efficient management and analysis of diverse data types, paving the way for ongoing advancements and insights in the future.

The improved efficiency achieved through the collaboration of the pharmacist within Medical Affairs and data engineers from the Advanced Analytics & Data Automation team is anticipated to have a positive impact on multiple departments, extending beyond Medical Affairs. This includes departments such as clinical operations and the commercial team, both within the US and on a global scale. The enhanced generative medical insight analysis facilitated by this collaboration will enable these departments to make more informed decisions, optimize operations, and drive better outcomes in their respective areas of focus.

Value Brought by Dataiku:

By leveraging Dataiku's built-in recipes and plug-ins, an efficient operational model was swiftly developed. Different aspects of the Dataiku platform helped:

  • Transparency, collaboration, centralization: Dataiku brings together all stakeholders within a simple yet powerful visual interface. This enables business users to understand the capabilities and opportunities brought up by data analysis — a milestone to ensure that the most impactful projects are built, in the most efficient manner.
  • Faster development: The built-in data preparation steps were pivotal to progress faster with data cleaning and wrangling.
  • Project and workflow reusability Pivotal to grow the impact exponential to use cases across departments and domains.
  • Cutting-edge technologies: The new Generative AI plugin made it very simple to leverage ChatGPT and train our model to summarize text and perform advanced sentiment analysis to enhance the value of the project.

This approach enables the generation of targeted and relevant insights that align with Moderna's medical strategy. Dataiku's platform plays a crucial role in optimizing the analysis process, ensuring that the insights generated are aligned with Moderna's goals and reflect the sentiments expressed by healthcare providers. This alignment ensures that Moderna can make informed decisions and develop strategies that effectively address the needs and concerns of HCPs, ultimately driving the success of their medical initiatives.

Value Type:

  • Increase revenue
  • Reduce cost
  • Reduce risk
  • Save time
  • Increase trust

Comments

Setup Info
    Tags
      Help me…