From Vaccine to Drug Making, AstraZeneca is Relying on AI for Growth

AstraZeneca is using AI to not just power Covishield vaccine production, but also in other drug makings

Ever since the outbreak of coronavirus, the news is filled with daily record-breaking cases, deaths, and fortunately, vaccine-related improvements. When Pfizer became the first company among the democratic countries to unravel a vaccine, the stock market prices skyrocketed. The achievement shined a ray of light among the hopeless population. Soon, Moderna, AstraZeneca, and Johnson and Johnson followed suit. AstraZeneca, one of the biggest contributors of the Covid-19 vaccine is using artificial intelligence to not just power Covishield vaccine, but also in other drug discoveries.

AstraZeneca is a global, science-led biopharmaceutical business. The company produces major innovative medicines that are being used by millions of patients worldwide. AstraZeneca is organized to deliver its strategic priorities sustainably, supporting continued scientific innovation and commercial success. While the company is already engaged in harnessing data and technology to maximize time for the discovery and delivery of potential new medicines, the pressure to roll out a vaccine at a very short period further streamlined it. The company is embedding data science and artificial intelligence across its research & development to enable its scientists to push the boundaries of science to deliver life-changing medicines. However, all is not good. AstraZeneca ran into trouble multiple times in recent days due to the AstraZeneca vaccine and its data. We'll take you through how the company is using technology to power its medical product production and what troubles they are facing related to data

Using artificial intelligence in drug discovery and development

Drug discovery is a very twisted process. Even though scientists and researchers put out their efforts and minds in coming up with the right drug, like gambling, everything could collapse at the last minute. Generally, developing a new drug takes 10-15 years of undisrupted efforts and hard work, and millions, if not billions of dollars. Unfortunately, even after undergoing such extremes, 90% of candidates fail to make it into the market.

The only way to combat the challenge is to deploy technologies like artificial intelligence to streamline the drug discovery process. AstraZeneca is one of the early adopters of artificial intelligence in the healthcare sphere. The company is incorporating AI in every step of the Research & Development chain to make the drug-making process safer, quicker, and less costly. AstraZeneca is using knowledge graph and image analysis to gather new insights about diseases and identify biomarkers 30% faster than human pathologists. Besides, the company is also availing of machine learning and chemistry automation to shorten the lengthy drug discovery process and cost by 75%. Recently, AstraZeneca engaged machine learning in pathology to speed up the review of tissue samples. Labeling the data is a tedious process. Fortunately, AstraZeneca is making it simple with machine learning. The technology first learns from a large, representative dataset, and later uses its data labeling and annotation service to automate the labor-intensive work.

While drug discovery is already an intellect-intensive process, drug development is a time and labor-consuming job. But AstraZeneca is making it simple with the help of data analytics to mine through electronic health records (EHR) to improve clinical trial through faster patient identification and recruitment. Throughout its success journey, the company was also engaged in gaining a partnership with technology companies that could stand along its product-making process. A couple of those remarkable mergers are listed as follows,

• AstraZeneca has collaborated with Mila, a renewed machine learning company to maximize its potential for drug discovery and development. Mila's leadership in machine learning espoused with AstraZeneca's deep R&D knowledge and data science teams aim to help accelerate the company's research activities.
• AstraZeneca has revealed that its in-house team of engineers using PyTorch to simplify and speed up drug discovery. By combining PyTorch with Microsoft Azure Machine Learning, the company's technology can go through massive amounts of data to gain new insights about the complex links between drugs, diseases, genes, proteins, or molecules.

Where should AstraZeneca work to stop data issues?

Despite becoming one of the initial companies to produce and circulate a Covid-19 vaccine, AstraZeneca has run into a number of troubles in 2020 and 2021. First, it was an unusual blood clot found in vaccinated people and now, it is data concerns. Following the detection of blood clots, many countries across the globe including EU nations and the United States suspended the use of the vaccine for a brief period. But now, the trouble is knocking in form of data concerns and safety worries. In March, the company released the long-awaited phase three trial that it conducted on 32,000 people in United States, Chile, and Peru. The vaccine efficacy was at 79% according to the trial results. The press release was based on the data gathered until 17 February. Unfortunately, Data and Safety Management Board (DSMB), overseeing the study, said it is very concerned about the data that were excluded, pushing AstraZeneca to a critical spot once again.