PhynX Lab, an SK Networks company in Silicon Valley, launched Cheiron, the first-ever generative AI (genAI) search platform specifically for the pharmaceutical industry, in Korea. It was developed to change how medical information is accessed and utilized by professionals for pharmaceutical research and development to become more efficient and accurate.

Company Background

By the founder: it’s co-founded by three students who happen to be at Stanford University. PhynX Lab has quickly established itself to be at the forefront in AI-driven solutions for pharmaceuticals. In early 2024, the firm sealed a deal with SK Networks to develop and market its technology.

It officially established itself as an independent organization in September. With its lean team of seven staff members, PhynX Lab is committed to unlocking Cheiron’s applications away from research and into drug development and other pharmaceutical operations.

Read also: Decart AI Secures $32M funding valuation to over $500M

Cheiron

It is an advanced AI-powered search platform engineered to streamline retrieval and analysis of medical information. Cheiron is more than a traditional keyword-based search engine, as it acts as a comprehensive research assistant that scans academic databases, search engines, and company documents for synthesizing the results into concise, source-cited answers. The design is done with inclusion of MeSH (Medical Subject Headings) systems for ensuring the precision and relevance of findings.

Read Also: Salt AI raises $3 Million and Appoints Aber Whitcomb as CEO

Key Features:

• Modular Retrieval-Augmented Generation (RAG) Framework: Cheiron adopts a modular RAG framework that is the next version of RAG. That means it retrieves information in its databases relevant to producing an answer. The Cheiron modular RAG framework would let it choose certain information for the purpose using NLG, self-reverification of information, and instant high-quality answer generation. Modules in the framework may be connected easily, rather like Lego blocks, facilitating further extension to new technologies or sources of data.
• One-Stop Semantic-Based Search: With natural language processing, Cheiron understands and processes user queries in a human-like manner, delivering precise and contextually relevant information. This semantic understanding reduces the time spent on literature and data searches, enhancing efficiency in an industry where time and margins are tightly linked.
• Comprehensive Data Integration: Cheiron can pull information from different academic search platforms such as PubMed, general search engines, open sources, and internal corporate documents. This integration will enable users to have access to multiple databases in a single platform, which would facilitate the research process.
• Automated Workflow: The platform automates workflows, including organizing data and creating documents, thereby reducing manual effort and the potential for errors. This automation supports research and candidate substance discovery, clinical trial design, production and process development, and market analysis
• Multilingual Support: Beyond Korean, Cheiron supports major languages, making it accessible to a broader range of professionals in the global pharmaceutical industry.

Addressing Industry Challenges

PhynX Lab developed Cheiron to address limitations of existing healthcare large language models (LLMs). Current models are often limited by structure, relying on pre-learned data, which makes them prone to misinterpreting information and providing unfounded answers based on unlearned data. Cheiron’s modular RAG framework allows for real-time retrieval and verification of information, ensuring accuracy and relevance.

Demonstrated Efficiency

The product’s developers during its presentation quoted user testing findings, asserting Cheiron will save a portion of the time for seeking a particular document during their search “up to 80%” in contrast with existing practices. That should help quicken all those critical steps involved in drug or medical discovery from preliminary study design up to and into clinical trials.

PhynX Lab has big plans for future development of Cheiron. The company will extend Cheiron’s applications beyond merely research to drug development as well as other pharmaceutical operations. By continuously improving Cheiron’s capabilities, PhynX Lab aims to revolutionize pharmaceutical workflows, making the processes more efficient and less prone to error.

Industry Reception

Cheiron has received interest among professionals in the pharmaceutical industry for its introduction. In the live demonstration, PhynX Lab compared answers of Cheiron and ChatGPT, emphasizing that the former provides correct, source-verified information. For example, when asked about how intermittent fasting influences hair follicle regeneration, Cheiron referenced a current research paper stating that there is an adverse effect while ChatGPT provided the opposite.