Industry: Healthcare

Remodeling AI-powered Key Opinion Leader (KOL) Search Function

0 Engineers
0 Months
0% Efficiency in Search Function

Services

Rapid Application Development, Product Engineering, Data Engineering

Technology

Python, Javascript, Flask, React, GraphQL, MySQL, Neo4j, AWS Elastic Bean Stalk

Location

USA

Interested?

Estimate a project

Overview

A Seattle-based tech startup is shaking up the global healthcare ecosystem by innovating an AI-powered Key Opinion Leader (KOL) search function that will incorporate knowledge graphs. This novel way of retrieving and presenting information would transform product development and market research for emerging medical technologies.

This KOL Search will create a single view of aggregated data to end users, helping them discover and identify relevant KOLs in their areas of interest. The tool conducts searches across publications and clinical trials, and ultimately reduces the time spent manually scouring and aggregating data. It will enable enterprises to find trends and gaps in the healthcare market and fine-tune their offerings to meet emerging demands.

Teamed up with TechVariable to design the UI/UX, front-end, and backend modules for the new feature.

Challenges

There were many challenges facing the team of four. In addition to working together in time zones, the group also learned new coding languages and techniques, creating a brand new tool in a record one-month timeframe.

  • The project involved quickly incorporating a large number of JSON files into Neo4J, a graphing database. These files included healthcare practitioners, healthcare organizations, open payments, and taxonomy data.
  • Another challenge to overcome was performing real-time aggregation queries on millions of data sets in Neo4j for the analytical graphs on the dashboard.
  • The team was asked to implement a ‘Global Search AutoComplete Feature. This would allow the user to input any character, and if any property of any node contained those keywords, the search would quickly return that data.
  • A variety of knowledge graphs with different custom designs were part of the required outcomes. The team successfully delivered those results within the given timeframe, despite challenging time constraints on the project.

Solutions

  • The team closely collaborated with the client while creating the new software. They created new approaches and learned new skills to complete the project.
  • They first converted the JSON files into CSV format. They had to create separate CSV files for nodes and relationship data. Then, they implemented a Neo4j admin tool to import the new CSV files. It allowed them to insert around 70 million nodes and 200 million pieces of relationship data in around five to six minutes, instead of hours.
  • Since the queries for generating the graphs were mostly OLAP queries, they
    denormalized the data and precalculated expensive aggregations. They then stored the data and aggregations on new nodes and relationships.
  • The team implemented a full-text search on the nodes using Lucene indices, and using a distance-similarity calculation, they were able to fetch the results in little time.
  • To streamline the knowledge graph generation, the team used only one component each for tables, line charts, and bar charts. By ignoring the designs attached to each graph type, the team used a general approach, dynamically passing those to the desired components.

Need a custom software application for your business?

We at TechVarible do acknowledge that one size will not fit all. Hence, we work in collaboration with you to identidy, analyze & then develop a solution that fulfils your needs.

Either we will define the functional scope of your project to estimate the timeline and budget or you can create your own agile team from among our recources.

Book a Call

The Result

  • The new Global Search Autocomplete feature allows users to search millions of pieces of KOL data, publications, clinical trials, and related industry information all in a single search box. These advanced aggregations occur in real time, constructing relevant and actionable knowledge graphs for users.
  • The team also created a special module linking healthcare practitioners’ Twitter accounts to search results. This enables end users to see relevant KOL tweets and hashtags that are currently trending. Additionally, the results show similar Twitter accounts for the pertinent therapeutic field.
  • The new search tool was less expensive and used newer modules than its competitors, giving them an edge in the industry. It has already contracted TechVariable to work on a second project.

Modules Implemented

Enquire now

Give us a call or fill in the form below and we will contact you. We endeavor to answer all inquiries within 24 hours on business days.

Help us understand your requirement

Staff Augmentation

"*" indicates required fields

Which developer are you looking to hire?*
How many years of experience do you need?*
How soon are you looking to fill this position?*
Upload your JD
Max. file size: 25 MB.
Or