Making AI a possibility for life sciences

A pharma success story: How a generative AI chatbot helped in driving swift query resolution in clinical trial

On average, it takes over 10 years and costs upwards of $2.6 billion to get a new drug from initial discovery to approval by regulatory bodies like the FDA. Moreover, processes like clinical trials are long, complex, and expensive that pharmaceutical companies must undertake to test the safety and efficacy of new medical treatments. A major factor behind these daunting timelines and costs is the organizational challenge of coordinating and managing thousands of human participants across multiple research sites during trials that can last months or years. Manual and repetitive administrative/training tasks become immensely tedious to cope with. This inevitably leads to inefficiencies that inflate costs and delay results. Let us bring you a real-time example of how a pharma company embraced the power of generative AI to streamline their clinical trial operations. With over 500 active clinical studies spread across the country, the Research Pharmacists (RPs) were spending at least 50% of their time handling queries from the study teams. Most of these queries were repetitive and hindered the RPs productivity in the other crucial areas of the study. The project leader opted to re-engineer the clinical trial operations by including robotic process automation, and bots to not only make the process efficient, but also bridge the knowledge gaps between the study team training and patient enrollment. Administrative tasks which were repetitive and taking up a lot of time were identified and were first in line to either get automated or handed over to a bot. Initially, the bot was challenged by the RPs on how it could effectively answer the study team’s questions. Also, they were skeptical regarding the process of redirecting any out-of-the-manual queries or escalations which required immediate attention. Generative AI Bot Intervention in Clinical Trial Operations Interactions between study teams and research pharmacists are a crucial aspect of ensuring the smooth progress of trials. Study teams and RPs often establish clear communication channels early in the trial. They mutually exchange information which includes details such as the drug's mechanism of action, formulation, dosage, administration protocols, and any unique considerations. The study team may enquire about proper storage, preparation, dispensing if methods are unclear, appropriate documentation practices. With handling over 500 clinical sites, the RPs were constantly getting tons of questions from different study teams. Though most of these questions were repetitive, timely responses were required for the smooth functioning of the clinical trial process. i2e Consulting was engaged to design a solution using which repetitive queries could be answered on behalf of the RPs. The Pharma client also wanted the solution to be capable of redirecting escalations and out-of-the-manual questions to the respective RP for immediate attention. Implementing the Gen AI Chatbot: Challenges & Accomplishments There were two challenges in front of team i2e, (1) to identify the repetitive questions directed towards the RPs, and (2) to design a process to handle any out-of-the-manual queries. Team i2e developed an advanced chatbot leveraging Amazon SageMaker’s generative AI capabilities along with a wrapper around Kore.ai to build a user friendly chatbot interface capable of answering queries pertaining to Investigational Product (IP). Utilizing sophisticated prompt engineering capabilities resulted in enhanced performance and efficiency in delivering precise responses. The backend web platform was built to store and process the vast IP manual documents and direct user’s inputs to relevant responses and escalations. The team also built a notification system which triggers emails to the CRPs in case of out-of-the-manual questions or escalations. Result? A significant reduction in repetitive queries, and 2X expedited query response, most importantly, the CRPs were able to invest their time in other crucial areas of the clinical trials, and the study team was able to clear off their questions instantly which contributed to the smooth progression of the clinical trial. The gen AI chatbot was a critical piece of the puzzle the client was trying to solve. The Clinical Research Coordinator wanted to plug the knowledge gaps resulting in repetitive questions directed towards the RPs. With the chatbot recording the questions and their answers, it became a central repository to identify knowledge gaps and bring about changes in the IP manual and the training modules. Expanding the Chatbot Footprint After successful implementation in two therapeutic areas, the project leader then expanded the chatbot to clinical trials in other areas. The solution designed was easy-to-deploy and had the scalability to include more clinical trials teams, and still delivered a similar output. The Future of Generative AI in Clinical Trails The possibilities of AI and ML do not end with chatbots, it can be extended to monitoring patient health. Herer are a few areas where AI can increase precision and productivity in clinical trials. Monitoring patient health and compliance Keeping human trial participants closely engaged and supported is crucial for productive trials. However, regular check-ins and counseling on a global scale has traditionally strained staff bandwidth. AI chatbots present a scalable solution, acting as always-available virtual health assistants. Integrating seamlessly into messaging apps people already use daily, bots can automatically check symptoms, deliver health information, manage medication intake alerts, collect progress self-reports, offer motivating behavioral interventions, and more. Working around the clock at marginal cost, they can provide responsive support with more reliability than overworked nurses across scattered trial sites. Bots further enable continuous remote patient monitoring to improve compliance rates that directly impact trial integrity. By establishing ongoing dialogue at scale, they also facilitate early detection of adverse reactions or mental health issues so coordinators can rapidly respond and keep participants engaged. Extracting powerful insights from trial data The data collected across a multi-year clinical trial is vast and complex, usually amounting to terabytes of medical records, genomic sequences, imaging scans, biomarker assays, questionnaire responses, and more. Manually combing through such immense datasets using legacy analytics tools is just not feasible. This is where generative AI truly shines, capable of intelligently parsing mountains of structured and unstructured data to uncover subtle patterns that lead to actionable insights. As natural language models, chatbots can further analyze doctors' notes, patient messages, and open-ended survey responses. Generative AI promises to transform every facet of clinical trials, from participant engagement to research analytics and everything in between. Chatbots and other models like GPT-3 mark an exciting shift toward cloud-based software that keeps getting smarter, allowing pharmaceutical experts to focus on innovation and strategy rather than getting bogged down in manual processes.

What is generative AI & how can life sciences organizations benefit from it

Life sciences industry is experiencing its own gen AI revolution resulting in efficient clinical trials, accelerated dug discovery, quicker document summarizations, chatbots taking over repetitive manual tasks, and we are just getting started. When used effectively, gen AI can assist in almost all the departments and stakeholders right from R&D to commercialization of a drug. In this blog we bring you a few basics about Generative AI, its applications throughout the drug development life cycle, and some potential use cases. How Does Generative AI Work? Generative AI operates through intricate neural network mechanisms, mimicking the complexities of the human brain. At its core, these networks process vast datasets, learning patterns and structures to produce original content autonomously. Discoveries in in-depth learning have driven generative models to elevated heights, allowing them to tackle various data categories such as text, images, and music. Now, let’s look at the gen AI models closely. Generative adversarial networks (GANs) and Variational autoencoders (VAEs) are the innovators in this area. They are revolutionizing various processes with their ability to generate more authentic new data from a given training dataset. These foundational models serve as the backbone for subsequent advancements, reshaping language and image processing. This is useful in creating artificial data sets (to train machine languages) where obtaining real-world data is tedious or violates patient privacy. For example, using a combination of GAN and VAE networks, the Centre for Computational Science and Mathematical Modelling, Coventry University, proposed a framework to generate artificial brain tumor MR images. These helped in training machine learning models used in classification of brain tumors which resulted in increased efficiency of these models from 72.63% to 96.25% (1). Generative AI in the Life Sciences Industry Generative AI is emerging as a new powerful tool in the world of life sciences. By using its power to analyze massive medical datasets, Generative AI is transforming business processes across the drug development life cycle. Leading to decreasing the time to market and providing superior and patient centric healthcare. With less than 10% of drug discovery efforts yielding breakthroughs, Generative AI appears as a game changer. It enhances compound screening, suggests feasible therapies, and simulates molecular dynamics for deeper drug behavior understanding. Plus, it also helps in making the clinical trial process more efficient and accurate, resulting in decreased costs and improved trial results. Applications of Gen AI Throughout the Drug Development Lifecycle Gen AI's ability to use diverse data sources and create customized content can help in many ways. For example, it can accelerate drug development processes, compress asset lifecycles, speed up therapy development, FDA approvals, and finally commercialization. Let us discover in detail the impactful role of generative AI in driving innovation in the drug development life cycle. Research & development (R&D) Gen AI is streamlining R&D in the life sciences industry by systematizing content extraction from different sources, including patents and scientific papers. Unlike traditional NLP, AI tools provide deeper insights into medical context and intent, simplifying open-ended questions and seamless integration of evidence. This streamlines research, requiring minimal additional training for tailored results. Clinical trials Gen AI optimizes patient selection through biomarker-driven precision medicine. By leveraging genetic, phenotypic, and real-world data, AI models identify patient subgroups for tailored treatments, enhancing trial diversity and efficiency. Moreover, AI analyzes medical images to uncover hidden biomarkers, enabling shorter trials and unforeseen treatment discoveries with higher success rates. Commercialization AI models can expedite life sciences marketing as they enable customized content creation and real-time campaign adjustments. They even leverage brands to refine strategies promptly, for smarter decisions. This promises efficiency gains and enhanced campaign effectiveness. Gen AI also assists in personalizing campaigns for more precision in targeting. Let us look at some of the potential real-world use cases of Gen AI in Life sciences. Use Case 1: Leveraging smart data management With tons of data pouring in constantly, data management is a labor-intensive and expensive affair for the life sciences companies; However, with a combination of traditional and Gen AI techniques, it can be automated smartly. Routine manual tasks like reviewing and configuring the data are streamlined, improving overall efficiency. A few clicks and the vast data are segregated as per the organizational protocols. Case report forms can be auto generated based on the protocols, and patient profiles. The data quality can also be enhanced by real-time cleaning and bridging the critical gaps through intelligent query generation. This smart approach accelerates clinical trials, utilizes resources optimally, and speeds up the outcomes. Use Case 2: Optimizing indication selection Indication selection for asset strategy in biopharma involves making critical decisions based on conditions to target a specific molecule. Despite the plethora of available information from various sources like opinion leaders, literature reviews, and trial data, the traditional approach often overlooks crucial evidence. On the other hand, Gen AI addresses this challenge by analyzing structured and unstructured datasets comprehensively. It leverages Real-World Data (RWD) and molecular knowledge graphs to expose semantic similarities between events and estimate biological proximity. By tapping into overlooked RWD and identifying new connections, Gen AI facilitates the discovery of novel indications, accelerating validation processes and minimizing opportunity costs. This advanced approach enhances decision-making by integrating diverse data sources and uncovering previously missed correlations, thereby improving the overall success of drug development endeavors. Use Case 3: Accelerating clinical trial outcomes Gen AI also serves as an effective assistant in clinical trial management by swiftly processing huge data to offer actionable understandings and boost trial outcomes. Leading pharmaceutical firms are embracing the Gen AI advancements to make operational decisions and accelerate trials. It is also offering custom insights, smart alerts for discussions, and automated communication. This can streamline cross-functional collaboration. These tools also update enrollment by automating analyses, addressing enrollment hurdles, and fostering teamwork. Gen AI's multifaceted support transforms clinical trials, promising faster, more efficient processes, and improved outcomes. Use Case 4: Streamlining patient engagement and recruitment through AI chatbots AI Chatbots is reforming patient engagement in clinical trials by providing quick responses to patient queries, helping them through trial information, and conducting pre-screening questionnaires. The chatbots also update the recruitment procedure, thereby cutting down time and resources while offering effective and precise patient selection. In the pharmaceutical industry, these chatbots improve customer interactions through customized information about treatments, medications, and even their side effects. This is not all, timely support is also offered so patients can make sound decisions in time. AI chatbots are indeed offering advanced patient engagement solutions along with all-around support through clinical trials for life sciences organizations. Generative AI is proving to be a driving force in life sciences organizations. It is not only about creating but going beyond and changing the game. From researching comprehensive data to transforming pharma methodologies, the influence of AI is profound. Any new guesses for what may happen in the world of Generative AI further? Well, hang tight as many more innovations are waiting to be uncovered! Future Directions of Generative AI  Looking ahead, the future of generative AI in life sciences is full of promise and excitement. It's all about staying informed, collaborating, and promoting ethical practices. Going forward, AI can be helpful in accelerating drug discovery while ensuring fair and reasonable patient care. From continuous learning to sharing valuable insights, every step counts. Generative AI is shaping a future where innovation and patient well-being go parallel. With visionary leadership and a commitment to excellence, the possibilities are endless. Whether seeking Gen AI services or bespoke custom solutions, i2e Consulting can elevate your drug development journey. Talk to us and learn how we can help you achieve success in the pharmaceutical industry. References: Bilal Ahmad et al; Brain Tumor Classification Using a Combination of Variational Autoencoders and Generative Adversarial Networks; Biomedicines; 2022Chaitanya Adabala Viswa et al; Generative AI in the pharmaceutical industry: Moving from hype to reality; McKinsey & Company; 2024

Is your data ready for generative AI: a guide for life sciences organizations  

Generative AI (Artificial Intelligence) comes with a promise of offering unparalleled opportunities to life sciences organizations. Yet, the success of the journey grips on how data ready is your company for gen AI. From improving drug discovery to enhancing trials and devising marketing strategies, there is a vast potential to take benefit from the use of gen AI applications. However, the hurdle most Chief Data Officers (CDOs) and data leaders in the life sciences domain are facing is managing data and scaling AI use cases. Now, they need to focus on making changes within the data and the architecture for gen AI to produce meaningful results for the business. In this blog, we explore the importance of making data ready for generative AI and actionable insights for life science companies to navigate the generative AI data with confidence. Importance Of Data Quality for Gen AI Applications Data quality affects the accuracy, dependability, and consistency of algorithmic patterns and results of gen AI applications. To ensure its standards, organizations should build strategies comprising of data validation and data cleansing methods. Data validation refers to authenticating the accuracy of information through different facets. It includes verifying the data for errors, patterns, and inconsistencies and ensuring it runs parallel to the organizations' standards. While the data cleansing process is implemented to fix the errors found during validation, it involves eliminating duplication, correcting errors, and standardizing the data for overall consistency. Data validation is decisive for Gen AI applications as it makes sure the data presented to AI models is reliable, consistent, and precise. Without validation, the input data could have inconsistencies, biases, and errors, leading to variable and unreliable AI-led output. These make sure that AI models are trained to offer reliable and high-quality data for organizations to lay their problem-solving decisions. What Constitutes Data Readiness for Generative AI Data readiness for gen AI involves multi-layered tactics with a few components that are critical for organizations. Next, let us look at the steps involved in preparing data for gen AI usage. Steps to Prepare Data for Gen AI To leverage the power of Gen AI, the data should be prepared well. Here are the four critical steps to prepare life sciences data for gen AI. Data acquisition and creation The fundamental practice of preparing data for gen AI starts with acquiring data from diverse datasets and curating relevant data. The data should consist of all the critical components which are essential to generate the right response. For example, while acquiring data for drug development care should be taken to include chemical structures, target proteins, biological assays, drug reactions, and trials. Data can be obtained from academic literature, internal records, public repositories, and proprietary databases. The next focus should be on the creation of data by cleaning the acquired data and standardizing it to maintain quality and consistency. At the same time, the steps should involve correcting errors, eliminating duplication, and regulating data formats. Additional data including patient demographics, assay conditions, and molecule identifiers should also be analyzed and cleared for further data interpretation and training models. Data cleansing and preprocessing This step involves improving the available data, particularly when it is disorganized or limited. For generating superior results from gen AI cleansing and preprocessing methods must be applied. Data synthesis is the method implemented, which involves creating new data samples based on the available data. A few generative AI techniques at this stage include interpolation and extrapolation, which means creating synthetic data as per the statistic models. Data synthesis is a broad concept that constitutes methods to create new data and is not limited to merely resampling. Gen AI models like generative adversarial networks and variational autoencoders can synthesize data samples from the curated data. Nonetheless, it should be ensured that the data reflects the real-world annotations. Feature engineering and selection This is a critical stage as the data collected must go through sifting, where the raw data transforms into a standard format appropriate for training gen AI models and contribute to visionary performance. For example, the data for drug development should undergo changing biological sequences for numerical embeddings, encode chemical structures, and extract information from clinical data. Some of the techniques involved at this stage are normalization, dimension reduction, and selection for computational efficiency. Moderation and model building Life sciences data for gen AI should be validated and facilitated for model training. This step involves adhering to quality based on accuracy and reliability for AI models. Conducting experiments, validating datasets, and checking for model robustness are a few more steps to assess the performance of gen AI models. The approach begins with a base model and then passes through layers of SFT (Supervised Dine Tuning), RLHF (Reinforcement Learning from Human Feedback), and Proximal Policy Optimizations. Another crucial aspect of model building is moderation, which helps to generate relevant data by eliminating socially irresponsible answers. SME verification Finally, Subject Matter Experts (SMEs) are required to verify the final data samples and ensure it aligns with drug discovery and biological plausibility. Adding a human element is necessary to validate the gen AI responses and test the data quality. Some other measures like implementing control mechanisms and data governance are critical to maintain reliability and integrity. Conclusion In the era of AI-driven world, the potential of data readiness in leveraging pharma organizations should not be overlooked. From enhancing drug discovery processes to clinical trials, and coming up with unparalleled marketing strategies, gen AI applications have the potential to energize the pharma organizations. Adhering to meticulous data preparation through advanced practices and accelerating pharma organizations to use gen AI’s full potential and result in breakthroughs and innovation in the healthcare and drug development industry. The future is gen AI and i2e Consulting can help you prepare for it. Our data scientists are experts in preparing data for gen AI models. We can also advise on implementing control mechanisms and data governance practices.