Spotfire Copilot for Clinical Study Safety

Overview of Spotfire Copilot

Spotfire Copilot™ is a free, natural language extension to the Spotfire® platform. It leverages large language models (LLMs) to augment business intelligence and artificial intelligence, all in Spotfire. 

This release is custom-built to perform the following tasks in Spotfire: 

• Question-answering for Spotfire questions
• Question-answering for user-loaded documents
• Auto visualization generation and modification
• Data function generation
• Explanation of visualizations
• Data interrogation

For more information on Spotfire Copilot, check out this community article.

Application Overview

In the Spotfire Clinical Study Safety application, a user is able to analyze clinical study data, including demographics data, disposition events, adverse events and relative risk assessment, lab results, and more. This study has 3 patient groups, based on the dose they were given–a high dose group, a low dose group, and a placebo group. For information on this application, check out this article.

Demo Walkthrough

With how much this analysis covers, I’m looking to have Spotfire Copilot help me narrow down what I should be focusing on in this data. 

I also have access to a pharmaceutical manufacturing handbook, where I can ask questions relating to the document and quickly get back a response. In this case, I’ll ask about a particular medication.

Now, let’s look at the demographics page. Here, we have plots to analyze our subjects by gender and age, but I’d like a visual that shows subjects by race as well. Copilot can help me with this. By default, bar charts are created with vertical bars in Spotfire. Because Copilot retains conversational history, I can also modify the visualization by speaking to Copilot, rather than figuring out where in the settings I can find what I’d like to change. I’m also going to quickly update the colors and add some labels to match the rest of the application’s color scheme.

Next, let’s look at the adverse events page. Adverse events are any unfavorable or unintended signs or symptoms that occur during the clinical study. Let’s say I’d like to do some reporting on adverse events. Copilot has been trained to explain visuals, which is a great starting point for a report. 

Something else I’d like to incorporate into my report is the time of resolution for resolved adverse events.

Asking specific questions about the data can also help narrow down what subset of results we should look at first. Across two pages of lab results, we have four categories of labs, each with numerous results, and that gets multiplied by 3 with our 3 treatment groups. That’s a lot to look at! To narrow this information down a bit, I’m going to ask about what results have the highest amount of change from baseline.

We see here that the test for Eosinophils show the largest change from baseline for the high dose and low dose group by a decent margin. Insights like this are valuable in narrowing down what information could be the most important to report on and dig deeper into.

Demo Recording

Check out the video below to watch the full demo!

Resources

• Spotfire Copilot Landing Page + FAQ
• Download Spotfire Copilot
• Spotfire Copilot: Interact with Spotfire in Human Language (history of our Copilot work)
• NLP and LLMs in Spotfire (overview of NLP and LLM tools for Spotfire)
• NLP glossary (understanding common terminology)
• Have questions? Need support? Contact us at datascience@spotfire.com.