The Trending Role of Artificial Intelligence in the Pharmaceutical Industry 

Aug 18, 2020  |  7 min read

Over the last several years, the use of artificial intelligence (AI) in the pharma and biomedical industry has gone from science fiction to science fact. Increasingly, pharma and biotech companies are adopting more efficient, automated processes that incorporate data-driven decisions and use predictive analytics tools. The next evolution of this approach to advanced data analytics incorporates artificial intelligence and machine learning.

This article is posted on our Science Snippets Blog.

Unlike the artificial intelligence (AI) of sci-fi movies that takes over the world, the AI being used in pharma and other industries is a narrowly focused type of machine intelligence designed to solve a specific task or set of tasks using automated algorithms.

The goal of this type of AI technology is to find hidden patterns and gather insights from vast amounts of data in ways no human could. Using AI for data mining and analytics is already transforming many industries, including pharma and biotech. Its uses range from drug discovery to production process automation to clinical applications (such as medical imaging and surgical robots).

View case study

Types of Artificial Intelligence

Most artificial intelligence solutions used in healthcare today are based on human-created data science algorithms. This type of AI uses uses multivariate data analytics supported by past experiential evidence. It might combine, for example, population-based treatment outcomes with individual patient’s clinical data and medical history to create treatment alternatives and recommend drug combinations.

Another level of AI is machine learning, which relies on so-called neural networks that mimic the way a human brain works, but can potentially reach decisions much faster and more accurately. Machine learning uses data-driven algorithms that enable software applications to become highly accurate in predicting outcomes without any need for explicit programming.

The next level of AI is deep learning, which is also based on neural networks, but includes a combination of separate layers of calculations along with combined signals. Deep learning has great potential for diagnostic uses, being able to accurately analyze images (such photos of skin conditions or radiology scans) in combination with pathology data and historical treatment outcomes .

Applications for Artificial Intelligence in Pharma

From early stage drug discovery to prescribing treatment options, the use of AI is growing steadily within the biopharma industry, with a projected market volume reaching $10B by 2024 (including AI-based medical imaging, diagnostics, personal AI assistants, drug discovery, and genomics).

Some of the ways AI is being applied in the biopharmaceutical industry today include:

Manufacturing process improvement

In development and production, AI provides numerous opportunities to improve processes. AI can perform quality control, shorten design time, reduce materials waste, improve production reuse, perform predictive maintenance, and more.

AI can be used in many ways to make production more efficient with faster output and less waste. For example, a process that typically relies on human intervention to input or manage process data can be done using CNC (computer numerical control). The AI machine learning algorithms not only ensure tasks are performed very precisely, but also analyze the process to find areas where it can be streamlined. This results in less material waste, faster production, and more consistently meeting the product’s Critical Quality Attributes (CQAs).

Drug discovery and design

From designing new molecules to identifying novel biological targets, AI is playing a role in drug target identification and validation; target-based, phenotypic, as well as multi-target drug discoveries; drug repurposing; and biomarker identification. The key benefit for pharma companies is the potential for AI, especially when implemented during drug trials, to reduce the time it takes a drug to get approval and reach the market. This can result it great cost savings, which could mean lower cost drugs for patients, as well as more treatment choices.

For example, pharma researchers can identify and validate novel cancer drug targets using data such as longitudinal electronic medical records (EMR records), next generation sequencing, and other ‘omic data are used to create representative models of individual patients.

Read more: How to optimize cell culture media to speed biopharma development

Processing biomedical and clinical data

Perhaps the most developed use of AI so far is in algorithms designed to read, group and interpret large volumes of textual data. This can be a big time-saver for researchers in the life sciences industry, providing a more efficient way to examine the enormous amounts of data from the growing volume of research publications in order to validate or discard hypotheses.

Furthermore, many clinical studies still rely on paper diaries in which patients log when they took a drug, what other medications they took, and any adverse reactions they had. Everything from handwritten notes and test results to environmental factors and imaging scans can be collected and interpreted by AI. The benefits of using AI in this way include faster research and cross-referencing of data, as well as combining and extracting data into usable formats for analysis.

A Cognizant study showed that around 80% of clinical trials fail to meet enrollment timelines, and one-third of all Phase III clinical study terminations are due to enrollment difficulties.

Rare diseases and personalized medicine

Combing information from body scans, patient biology and analytics, AI is being used in various ways to detect diseases such as cancer, and even predict health issues people might face based on their genetics. One example is the IBM Watson for Oncology, which uses each patient’s medical information and history to recommend a personalized treatment plan.

AI is also being used to develop personalized drug treatments based on an individual’s test results, reactions to past drugs and historical patient data for drug reactions.

Identifying clinical trial candidates

Besides helping to make sense of clinical trial data, another use of artificial intelligence in the pharmaceutical industry is finding patients to participate in the trials. Using advanced predictive analytics, AI can analyze genetic information to identify the appropriate patient population for a trial, and determine the optimal sample size. Some AI technology can read free-form text that patients enter into clinical trial applications, as well as unstructured data such as doctor’s notes and intake documents.

A staggering 86% of clinical trials fail to recruit sufficient patients. This leads to slower research and delays patients’ access to life-saving drugs.

Predicting treatment results

Among the more time- and cost-saving applications of artificial intelligence, is the ability to match drug interventions with individual patients, reducing work that previously involved trial and error. Machine learning models are capable of predicting a patient’s response to possible drug treatments by inferring potential relationships among factors that might be affecting the results, such as the body’s ability to absorb the compounds, the distribution of those compounds around the body, and a person’s metabolism.

Predictive biomarkers

Development of biomarkers is an important task not only in the context of medical diagnostics, but also for the process of drug discovery and development. For example, predictive biomarkers are used to identify potential responders to a molecular targeted therapy before the drug is tested in humans. In this process, AI uses biomarker models that are “trained” using large datasets.

Drug repurposing

For budget-pressed pharma companies, repurposing drugs promises to be one of the most immediate areas that AI-based technologies can deliver great value. Repurposing previously known drugs or late-stage drug candidates towards new therapeutic areas is a desired strategy for many biopharmaceutical companies as it presents less risk of unexpected toxicity or side effects in human trials, and, likely, less R&D spend.    

Drug adherence and dosage

Ensuring compliance to a drug study protocol by voluntary participants in clinical studies is a huge problem for pharma companies. If patients in a drug study don’t follow the trial rules, they must either be removed from the study or risk corrupting the drug study results. One of the important factors of a successful drug trial is ensuring that participants take the required dosage of the studied drug at the prescribed times. That’s why having a way to ensure drug adherence is so important. Both through remote monitoring and algorithms for evaluating test results, AI can sort the good apples from the bad.

Moving Toward AI in Pharma Development

While the possibilities for using AI in pharma and biotech development are obvious, the actual move toward adopting such technologies can be slow going. Not only do traditional drug development and discovery processes require a more gradual adaption (rather than what some might consider a “disruption” by technology), the process for “training” AI in what works for drug discovery can take longer than in other applications.

For example, when social media tags your photo using AI, it gets immediate feedback from you about whether the results is correct or not, which allows the AI to learn quickly. With drug discovery, the feedback on a new molecule as drug candidate can take months or years to prove.

Yet, it’s undeniable that AI will be the next big thing in the pharma industry, and those companies that adapt and adopt new processes will have a strategic advantage. A good place to start is by using technologies that exist today for data analytics based on multivariate and predictive analytics.

An Example From Amgen

Find out how Amgen is using AI and advanced data analytics as part of their digital transformation process.  Get the case study.

Get Amgen Presentation