How generative AI could change the life sciences landscape - an interview with Indegene

How-generative-AI-could-change-the-life-sciences-landscape.jpg
© Getty Images (Getty Images)

With generative AI continuing to create a buzz, OSP took the opportunity to speak to Indegene, a digital-first, life sciences commercialization company that helps biopharmaceutical, emerging biotech and medical device companies develop products. Talking to Tarun Mathur and Sameer Lal, we found out how they thought AI could change the life sciences landscape.

OSP: Can you explain what Generative AI is and how it can transform the life sciences industry?

Mathur: Generative AI is a type of artificial intelligence that uses large language models such as GPT-4 to generate new content in various forms such as audio, code, images, text, simulations, and videos. It has the potential to transform the life sciences industry by delivering targeted and customized information to specific stakeholders, enabling swift decision-making, improving the productivity of skilled resources such as medical writers, addressing skill gaps, converting data to intelligence, and delivering insights.

Lal: In addition, Generative AI can also transform content forms and be used for content analysis & synthetic data creation, dialog and response generation, summarization and content translation, and classification and entity identification. Its impact on the life sciences industry can be profound.

OSP: What are the broad application areas of Generative AI in this industry?

Mathur: Large language models, such as GPT-3 and GPT-4, can identify and extract information from unstructured text and accurately classify it into suitable domain taxonomies. This capacity for classification and entity identification could have far-reaching applications in various domains, including Safety, Regulatory and Labeling, MLR review, and the marketing and sales of pharmaceutical companies. New generation GPT models are multi-modal, capable of processing and generating text, images, and other media. It has significant implications for scientific communication in life sciences companies, allowing for streamlined processes across medical and commercial teams. Clinical teams can use tables-to-text and text-to-table conversion while authoring clinical documents; medical affairs and commercial teams can create a variety of infographics for medical training and promotional materials.

Lal: Generative AI models can accelerate tasks such as summarizing and synthesizing literature articles and large clinical/regulatory documents when combined with embedding technology and vector databases. It could improve content authoring efficiency throughout the drug lifecycle. Generative AI solutions can also automate the initial draft of translated content, ensuring content availability across local affiliates. In addition, Generative AI models can aid content analysis and synthetic data creation, providing insights and recommendations for various medical and sales-related interactions. Imagine developing and deploying chatbots trained on a corpus of data for multiple applications, including medical information, patient engagement, and MSL training.

OSP: As we delve into the impact of Generative AI on the life sciences industry, I suppose we must also consider the potential pitfalls that come with it. Can you shed some light on the challenges and limitations we should know when implementing Generative AI?

Lal: There are inherent risks associated with Generative AI and life sciences organizations need to be aware of these pitfalls. The outputs may not always be accurate or appropriate, biased and manipulated, and involve reputational and legal risks. The model's accuracy depends on the training data quality and the match between the model and the use case. Large GPT models have been shown to produce the "hallucinating effect," where the model creates facts or information that does not appear credible, accurate, or based on the user's request. Data privacy and security concerns and ownership and liability for such solutions make adopting Generative AI platforms in life sciences companies challenging.

Mathur: To leverage this technology while being mindful of these risks, life sciences organizations need to ensure that we augment Generative AI platforms with current and accurate life sciences-specific data, using best practices with prompt engineering and human oversight. Organizations need to be aware of the quality of the query being asked, and the model and sources of truth used for machine learning training.

OSP: That's interesting. Can you provide specific examples of how Generative AI can be used in the life sciences industry?

Mathur: Sure. One of the use cases is in safety and pharmacovigilance. Adverse events must be reported to regulatory authorities, which can be time-consuming and error-prone. Generative AI can help identify and extract relevant information from safety reports, generate a first draft of the adverse event report, and allow the safety professional to review and refine the output. It can save time and increase the efficiency of the safety reporting process.

Lal: Another example is the possibility of deriving insights and strategies from large data or content. Moreover, Generative AI models could also be used to cleanse data and create synthetic data to augment datasets while providing recommendations on the next-best action across various medical and sales-related interactions. 

OSP: What are some of the key considerations to focus on while using Generative AI solutions?

Lal: To ensure accurate and contextualized output of Generative AI, one needs to focus on: (1) Usage of validated, domain-specific data, (2) Validation and testing of Generative AI systems, (3) Prompt engineering to trigger more accurate responses.

Mathur: One should also not forget to continuously review outcomes and improve the model fine-tuning process and cadence. Focus on use cases where content is provided to Generative AI engines instead of asking them to fetch/create content to avoid hallucinations.