Implementing generative AI is essential for pharma

The emergence of generative artificial intelligence (AI) is causing a significant shift in content automation practices across various industries. In particular, the pharmaceutical sector stands to be profoundly impacted by this technology, potentially leading to a transformation in their traditional operational strategies. The implementation of automation within a heavily regulated document ecosystem is a multifaceted undertaking that warrants careful consideration. Although the pharmaceutical industry stands to benefit greatly from this approach, it is important to exercise prudence. Incorporating LLM components into pharmaceutical workflows requires a provider that adopts a dual approach, considering both data and human factors. This is crucial for the success of pharmaceutical companies. The utilization of LLMs and symbolic AI in model training is imperative to guarantee the dependability and legitimacy of the outcomes. It is imperative to consider a diverse array of fundamental customer prerequisites, including but not limited to data segregation and protection, lucidity and comprehensibility of AI models, impartiality and equity, and the ability to scrutinise discoveries. Due to the confidential nature of patient information, it is imperative for medical writers to utilise a secure and reliable solution. 

As per the statement of Brian Burke, who holds the position of Research Vice President for Technology Innovation at Gartner, it is anticipated that generative AI solutions will be accountable for the identification of more than thirty percent of new pharmaceuticals and materials by the year 2025. This percentage is expected to rise from the current zero. The potential impact of generative AI on the pharmaceutical industry and other heavily regulated businesses is significant, although the future of this technology remains uncertain.

Recent Innovations using Generative AI in Pharma

The renewed interest and ethical considerations regarding Generative AI may potentially overshadow the numerous beneficial outcomes that AI technological tools can achieve, especially in sectors like healthcare and pharmaceuticals.

Nvidia has introduced the BioNeMo Cloud Service, aimed at facilitating drug discovery by offering pre-trained AI models to drug researchers. This service aims to streamline the drug discovery cycle and enhance its efficacy.

According to a publication on the Nvidia blog, generative artificial intelligence (AI) models possess the capability to swiftly detect potential drug molecules, and additionally have the ability to create compounds or protein-based therapeutics from the ground up. The aforementioned models possess the capability to forecast the protein's three-dimensional configuration and the efficacy of a molecule's docking with a target protein. This is due to their training on extensive datasets comprising small molecules, proteins, DNA, and RNA sequences.

Amgen and other companies are currently utilizing the BioNeMo service.

BioNeMo offers a unified interface that provides optimized models and model hosting, enabling drug discovery teams to efficiently deploy and expand Generative AI workloads. Additionally, enterprises can train and refine custom models using their proprietary data, leveraging their extensive research and data resources. The platform's advanced features and streamlined interface make it a valuable tool for drug discovery and related fields. Nvidia has formed partnerships with enterprises that express a keen interest in augmenting medical devices and procedures through the utilization of artificial intelligence technologies. At the NVIDIA GTC, a prominent international conference on artificial intelligence, NVIDIA disclosed its collaboration with Medtronic, a leading medical device technology enterprise and an enthusiastic adopter of AI, to incorporate edge AI functionalities into Medtronic's GI Genius intelligent endoscopy module, which is created and produced by Cosmo Pharmaceuticals.

The FDA has granted clearance for the use of GI Genius, an AI-assisted colonoscopy tool, to aid healthcare professionals in identifying polyps that could potentially lead to colorectal cancer. This marks the first instance of such a tool being approved for clinical use. According to Nvidia, the integration of the Nvidia Clara healthcare platform has the potential to facilitate the scaling up of AI algorithm development for real-time procedures by Medtronic, which could lead to an acceleration of AI innovation for enhanced patient care. The GI Genius platform was developed with the purpose of accommodating a collection of artificial intelligence algorithms.

Furthermore, the integration of Nvidia Holoscan, a software platform that utilizes real-time AI computing for the development of medical devices, and Nvidia IGX, an industrial-grade edge AI hardware platform, will be utilized in conjunction with GI Genius to provide medical practitioners with AI-enhanced diagnostic images. Both of these platforms have been specifically developed to be utilized in tandem with GI Genius. As per Nvidia, Holoscan provides users with complete access to the infrastructure stack essential for the scalable and software-defined processing of streaming data at the edge. It is expected that the integrated solution will become available later this year. This development is expected to contribute to the emergence of a new era marked by a higher incidence of software-defined medical equipment and procedures.

Challenges in Implementing

The showcased advancements in LLM technology by ChatGPT have the potential to significantly augment the quantity of applications. However, there exist several noteworthy impediments that must be addressed before LLMs can be established as a customary practices in the pharmaceutical industry.

Accessing data sets in the pharmaceutical industry poses a challenge. Pharmaceutical marketers encounter challenges with proprietary data sources that often lack interoperability, while a significant proportion of generative artificial intelligence models have not undergone training on datasets specific to the life sciences industry.

Ethical considerations. Marketers exhibit a sense of prudence in embracing generative AI technology, owing to their apprehensions regarding the potential for the content generated to be misleading.

The regulations pertaining to the protection of data privacy. It is incumbent upon pharmaceutical marketers to ensure that their utilization of Generative Artificial Intelligence aligns with relevant data privacy regulations, such as the Health Insurance Portability and Accountability Act (HIPAA) in the United States and the General Data Protection Regulation (GDPR) in the European Union.