Analyze KOL communities using PharmaSignals(Analytics platform)

Pharmasignals - Analytics platform for KOL conversation

How PharmaSignals uses AI to capture Voice of Hematology KOL Communities

KOL Communities

It is widely known that knowledge thrives and grows in communities where specialists can discuss and exchange the latest in research, discoveries, publications, and opinions. Research requires refined expertise and long-term dedication within the medical community, often for years. Bringing these experts together creates synergies enabling a broader network reach and exchange of information, speeding up what otherwise may not have reached the public eye. As a result, discoveries are made, and new therapies are developed bettering our communities.

Hematology KOL Communities

At SetuServ, we believe strongly in supporting medical research and using our technologies to advance medicine for the greater community. As a result, SetuServ developed PharmaSignals, an AI text analytics tool specifically dedicated to the pharmaceutical industry that captures real-time customer intelligence from Voice of Physicians and Patients connecting pharma companies to digitally active Hematology KOL Communities.

With PharmaSignals, SetuServ analyzes KOL communities related to Hematology including the American Society of Hematology (ASH), the American Society of Clinical Oncology (ASCO), Columbus Oncology, Southern Oncology Hematology Associates, and the Columbus Oncology and the Hematology Associates. These organizations host a variety of medical conferences, online forums, and discussions where many of the world’s greatest KOLs in oncology and hematology convene to exchange knowledge and opinions on the latest in these therapeutic areas. (More on the Digital Presence of Key Opinion Leaders in Hematology can be found here.) The information exchanged digitally is quite valuable yet capturing and aggregating relevant data across multiple channels has proven challenging.

 

Capturing Data from Hematology KOL Communities: Process and Methodology

PharmaSignals seamlessly collects, cleans, and classifies real-time customer intelligence from Voice of Physicians and Patients using the following methodology:

Collect

  • Opinions expressed by physicians on studies and clinical trials are scattered across Twitter, media posts, conference transcripts, and other digital channels. 
  • PharmaSignals mines these data across all channels and integrates sources into one platform.

Clean

  • Data is cleaned, noise is removed, and subject-relevant information is passed through to be analyzed

Classify

  • Cleaned data is organized by drug, study name, disease area, mechanism of action, etc.

Filters include sentiment and key insight themes information is disseminated to members via an online portal/email digest. Filters include media type, source, author type, creator name, handle name, post type, and drug. Users are also able to drill down by study and disease area. Similarly, quantitative metrics allow for seamless measurement, tracking, and advanced analytics, including share of mentions (SOM), the share of interactions (SOI), and time-series analytics.

Use Case Example: Voice of the KOL from the ASH Conference

Such data mining and engineering analytics was performed for the ASH conference. Online conversations from physicians were mined and analyzed. The chart below shows the number of mentions by day during, and after the conference.

Use of PharmaSignals - Trends of Twitter/Media Sites

Data was analyzed using machine learning. Major themes revolved around a variety of topics including Chronic Lymphocytic Leukemia (CLL), Acute Myeloid Leukemia (AML), and Mantel Cell Lymphoma (MCL). Each therapeutic area can be viewed by various filters, including media type, source, author type, creator name, study name, product, and sentiment.

Use of Pharmasignals in analyzing voice of KOLs

PharmaSignals:

SetuServ’s Pharma Signals platform is specifically designed to analyze KOL communities’ large amounts of data using contemporary technologies like AI and NLP thus achieving speed, accuracy, and scale. The platform can aggregate data from various social and media sources, separate noise from insightful signals, and extract entities such as study name, disease area, drug, mechanism of action, and mine insights of interest. Various teams use the derived insights in different ways to provide insights on KOL, PCP, and patient conversations.

The output represents the quality of KOL opinions. Identifying and mining such insights provide actionable intelligence to pharma companies, especially on competitors’ drug pipelines.

For more information on our Hematology KOL Community database and applied analytics, don’t hesitate to contact [email protected].