Artificial Intelligence in the Biopharma and Healthcare Sector

By Ronald Dorenbos, Associate Director Materials & Innovation, Takeda

Ronald Dorenbos, Associate Director Materials & Innovation, Takeda

Although the financial sector, the automotive industry and the marketing and retail branch are taking the lead in implementing the latest artificial intelligence (AI) and machine learning (ML) technologies, today there is hardly any field that is not being touched by the progress that has been made since the term AI was first coined by John McCarthy in 1956. The biopharma and healthcare sectors are no exception and although acceptance and implementation of AI-driven technologies in these fields have been sluggish, more and more companies come to the realization that it is critical to embrace AI and related technologies to stay ahead of the competition or at least stay current.

Natural Language Processing

Progress in natural language processing (NLP) has led to sophisticated algorithms able to recognize spoken and written language which allows machines not only to recognize words and sentences, but as well to understand context and make connections between text in a variety of different formats from spoken words and written documents to scientific publications and electronic medical records (EMRs). Alexa, Cortana, Siri, and Google Assistant all are examples of powerful, AI-driven search engines made possible,in a large part,because of improvements in NLP. Without NLP the biopharma industry would not have seen today’s advances in AI-assisted drug discovery, design of and decision making in clinical trialsand complex data mining and analysis of big data sets.

"A variety of AI applications, especially ones that have narrow, well-defined tasks,already surpass human capabilities and they will only get better​"

Learning Machines

In parallel with the developments in NLP, ML has evolved to a level where artificial systems can learn from experience. NVIDEA has developed a self-driving car that learned how to drive by ‘observing’ human drivers andGoogle’s DeepMindtrained its algorithms to spot early signs of age-related macular degeneration based on a million Optical Coherence Tomography scans. Similarly, Babylon Health learns through millions of consultations it provides to people about issues with their health and is on par with experienced doctors. A variety of AI applications, especially ones that have narrow, well-defined tasks,already surpass human capabilities and they will only get better.

AI in Pharma

In the pharmaceutical industry,a variety of sectors across the value chain are implementing AI approaches or consider doing so soon. For drug discovery and drug development companies like Berg, Benevolent AI, Atomwise, Exscientia, Insilico Medicine, and Numerate are among the leaders in the field. In supply chain and manufacturing Neural shows how AI, in combination with automation and virtual andaugmented reality (AR/VR),provides entirely new ways to control processes, manage production lines, and do quality control and troubleshooting. Implementation of these novel technologies increases efficiency and productivity and leads to a safer working environment. 

AI for Patients and Diagnostics

Combining genomics, biotechnology, wearable sensors, and AIleads to unprecedented levels of personalized and precision medicine and patient centricity and patient involvement in creating and providing treatments are taken to new heights. AI-based platforms provide interactive, personalized, relevant, and inexpensive communication solutions for patients and are being used to positively modify patients’ behaviors. These platforms can drive adherence, educate and motivate patients, encourage and help with sharing of data and improve overall outcomes. In diagnostics, AI helps with analysis ofa variety of formats like images, voice, keystroke performance, EEGs, breathing, heart rate, and movement and sleeping behavior.Inclinical trials, AI increases enrollment efficiency, improves patient matching, and helps with design and management of the trial and monitoring of patients. 

Quantum Computing and Big Data

In parallel to advancements in the field of AI and ML, breakthroughs in other fields will have a major impact on future developments.In March 2018, Google announced Bristlecone, a quantum computer with 72 qubits. Earlier, in 2015, they announced a quantum computer 100 million times faster than a laptop, which translates to the same number of calculations in a day as it would take a laptop 274 thousand years. Although a practical quantum computeris not yet reality, it shows the possibilities. Furthermore, it is predicted that the world’s data will hit 163 Zettabytes by 2025, equivalent to watching the Netflix catalog 489 million times or 40 trillion DVDs,reaching the moon and back over 100 million times.A significant part of this will be healthcare data and global big data and analytics spending is predicted to reach $187 billion in 2019, up from $122 billion in 2015.Many agencies that try to forecast the future limit themselves to two to three years, since it becomes increasingly difficult to have a clear picture further ahead. Despite all these uncertainties, one thing is for sure: the future will be exciting and if you don’t want to miss the boat, better jump on the bandwagon soon.

Views and opinions expressed in this article are those of the author and do not necessarily reflect the policy or position of Takeda Pharmaceuticals.

Read Also

AI-Bridging the Human Element

AI-Bridging the Human Element

Bridget Frey, CTO, Redfin
Emotion AI: The Heart of Marketing Innovation

Emotion AI: The Heart of Marketing Innovation

Natalie Monbiot, SVP/ Futures, Starcom USA
The Importance of Humans in Machine-Learning-Based Fraud Systems

The Importance of Humans in Machine-Learning-Based Fraud Systems

Nitesh Kumar, Head of Data Science, Affirm, Inc.

New Editions