Artificial Intelligence in Pharma: Drug Side-Effect

Ways AI is giving a booster shot to the pharmaceutical industry

Martin Zurek

Last updated on July 12, 2020 | 11 min read

AIBig DataData MiningHealthcareInnovationNLP

Challenge

Enhanced Detection and Monitoring of Adverse Drug Reactions
Although the safety of drugs is tested during clinical trials, many adverse drug reactions (ADR) may only be revealed under certain circumstances, for instance: long-term use, when used in conjunction with other drugs or by people excluded from trials, such as children or pregnant women.
Nowadays consumers often report their experiences with ADR on social media instead of traditional channels, which makes drug safety surveillance systems less efficient.

Solution

Thanks to our solution, drug safety surveillance systems can be easily improved by incorporating the knowledge extracted from social media into them.
Our system enables fast scanning of various data resources, such as: Twitter, Facebook or forums posts.
Then, using state-of-the-art Natural Language Processing techniques it filters out irrelevant information. The remaining, relevant data is processed and certain entities, such as: drug name, ADR, pharmaceutical company name, person age and gender, are extracted and analyzed.
The last step is automatic report generation. The executive summary containing all found ADR and insights on how they occurred is created and could be used in further drug development.

Benefits

• More efficient identification of novel adverse drug reactions
• Reducing cost of drug safety surveillance systems
• Improving drug safety by finding potential ADR faster
• Identified interactions could be used in further study and drug development

Other applications of AI in Pharma

Research and Development

Drug research, discovery, and development cost companies substantial resources. The average cost of bringing a new drug to the market is US$1.3 billion. The cost for development also varies per the disease being researched, and cancer drugs prove to be the most expensive to develop. Nine out of 10 clinical drugs do not make it to clinical trials, and more so do not make it to the FDA approval stage.
This is one of the main reasons many new drugs for major diseases are often so expensive, and why most of these are inaccessible to most people, especially those without insurance or are underinsured.
Pharmaceutical companies investing in artificial intelligence, machine learning, and big data are opening the possibility of making these new drugs more affordable to the end consumer. The technology has great potential in slashing the costs and resources of drug R&D.
An article published in the NY Times describes how AI and deep learning algorithms are rapidly changing the drug discovery science.

Improving manufacturing

When AI is injected into the pharmaceutical manufacturing process, it can help pinpoint opportunities to streamline and boost production processes, be it for future or existing systems.
AI technology can optimize manufacturing through:

• inventory management
• supply chain management
• improved quality control
• warehouse management and control
• optimized (predictive) maintenance
• waste management and reduction
• better production reuse.

Quicker Diagnosis

On the healthcare side, machine learning systems can help doctors collect, process, and analyze great volumes of patient data. Millions of healthcare providers around the world already rely on ML technology for securely storing sensitive patient data, known as electronic medical records (EMRs).
ML systems can go through these EMRs to make fast predictions for patient diagnosis, as well as suggest a suitable treatment protocol for them. Being able to analyze massive amounts of data in a short time, ML systems can help to greatly reduce the diagnosis process, which could help save the lives of millions.

Epidemic Outbreak Prediction

The current global COVID-19 pandemic has highlighted the efficiency and efficacy of machine learning and AI in Pharma. The technology can analyze the history of an epidemic outbreak, including its origin and spread. Tapping into news, official records, social media data, and other sources of information, it can then be used to predict with considerable accuracy when and where a disease outbreak can happen.
In some countries in Europe, AI is being used to crunch pharmacy and hospital data, along with other data sets, to come up with medical and economic decisions in years, instead of the usual years.
Moreover, it is also being used to personalize treatment and help in creating new tools and equipment for patients and healthcare professionals.

A Quick Glance at Big Pharma Investments in AI

Big pharmaceutical companies have also already started tapping into machine learning and AI in Pharma to expand their operations.
European pharma giant F. Hoffmann-La Roche AG has been collaborating with AI startup OWKIN on a project called Socrates, the first data-driven machine learning platform for medical research. Roche has also acquired Alphabet Inc.-backed Flatiron Health, a startup that uses AI for cancer research and patient care improvement. These are just two of the multinational company’s acquisition of healthcare-focused AI startups.
Swiss multinational pharma company Novartis, for their part, has partnered up with IBM Watson, Massachusetts Institute of Technology, Intel, and Quantumblack to give its AI goals a boost. Thanks to these partnerships, Novartis has been leading the way in the use of AI in pharma use cases, with investments in drug trials, drug discovery, patient analytics projects, and other areas of healthcare and pharmacy.
Not one to be left behind, British multinational pharma giant GlaxoSmithKline has also been making moves to embrace artificial intelligence in recent years. One of their most notable moves to invest in machine learning and AI was in 2017 when they partnered with Exscientia, a leading pharma tech company and the first one to automate drug discovery.
As a full-stack AI drug discovery company, Exscientia generates its own data and then combines it with the analytical power of AI and the creativity and expertise of scientists. This allows the company to substantially shorten the preclinical drug discovery stage (by at least three-quarters), thereby accelerating the delivery of new treatments to patients.

Conclusions

The world has been rapidly moving towards developing and embracing new technologies. The healthcare and pharmaceutical industries are not immune to this shift. Artificial Intelligence in Pharma is evolving faster than ever before and we are eyewitnesses of those rapid changes.
Human clinical experts and machines will learn and grow alongside tech innovations. And media drug reviews will greatly help with that. Of course, there will be a number of challenges along the way, but we in DevsData claim that the teaming up of tech and Pharma can only lead to better things.
Who knows, soon, an Apple (or an Android) a day might just be what will keep the doctor away.