Subscribe Become an Author Sign Up Log In

Machine Learning

Section: Biopharma Insights     View all sections


Current State of AI in Pharma: Key Achievements Beyond Hype

   by Andrii Buvailo    2935
Current State of AI in Pharma: Key Achievements Beyond Hype

/Last update -- 24 Dec 2019/

A background context -- opportunities and challenges

Current widespread interest towards artificial intelligence (AI) and its numerous research and commercial successes was largely catalyzed by several landmark breakthroughs in 2012, when researchers at the University of Toronto achieved unprecedented improvement in the image classification challenge ImageNet, using their deep neural network “AlexNet” running on graphics processing units (GPUs), and when that same year Google’s deep neural network managed to identify a cat from millions of unlabeled Youtube videos, representing a conceptual step in unsupervised machine learning.

[Interview] Applying AI To Shape Business Strategies At European Pharma Organizations

   by Andrii Buvailo    381
[Interview] Applying AI To Shape Business Strategies At European Pharma Organizations

The application of artificial intelligence (AI) in the pharmaceutical industry has become a long-term strategic priority for most companies. However, the efficiency of this endeavor depends greatly on the availability of large volumes of properly curated quality data, which is not always the case.

While pharma organizations generate huge volumes of data across all stages of drug discovery, development, and commercialization, not all types of data are equally useful for building efficient machine learning (ML) pipelines. For instance, it is relatively easier to apply AI-tech to consumer-related business processes, where lots of well-understood and properly labeled data is available, than it is for basic research tasks, where data is complex, often poorly labeled and extremely domain-specific.

The above situation leads to a faster pace of progress with AI application in such areas as financial analysis, consumer-behavior prediction, patient classification, marketing, and so on.

One of the important hurdles that pharma companies are trying to solve using AI tech is brand management. Indeed, understanding peculiar features of various patient categories, their purchasing behaviors, reactions to different products, revealing possible risks and side-effects for each class -- those things become essential for pharma companies to be able to develop and implement truly patient-centric brand management strategies. Luckily, this is one of the most fruitful areas for the application of machine learning (especially deep learning) models.

To get a better understanding of how it can be done, I have asked several questions to Agnieszka Wolk, Senior Director, Data Science, IQVIA, who recently presented this topic at the PMSA 2019 European Summit in Basel, Switzerland. . 

 

[Interview] Demystifying the Role of Artificial Intelligence in The Life Sciences

   by Andrii Buvailo    550
[Interview]  Demystifying the Role of Artificial Intelligence in The Life Sciences

In this interview, Rasim Shah, Director at OKRA Technologies provided a glimpse into how the company applies state of the art machine learning technologies to solve real world challanges in the life sciences. Rasim also agreed to answer several questions about a more general context of AI in pharma, its current challanges and future perspective, as well as describe the current efforts the European Union puts into supporting the AI ecosystem in the region.

 

Rasim Shah, Director at OKRA Technologies:

OKRA Technologies is a leading European artificial intelligence (AI) company for life sciences. Our goal is to empower life science executives at their desks or whilst on the move, with explainable AI outputs. OKRA’s solutions deliver suggestions, predictions and explanations to enable life sciences executives and operational teams to drive the right drug to the right patient with humanised and understandable AI outputs. The OKRA engine learns from real-world data, structured, unstructured, clinical, commercial and scientific literature to drive the right insight to the different teams in life sciences. Our deep expertise in AI, combined with in-depth medical and product knowledge from life science leaders, has allowed us to develop and co-create products that can transform the way life sciences approach traditional industry challenges. We focus on operationalising AI in an ethical way by placing users of these systems at the centre.

The Overview of AI in Drug Discovery in 2019: The “Proof-of-concept Year”

   by Andrii Buvailo    2675
The Overview of AI in Drug Discovery in 2019: The “Proof-of-concept Year”

The race for adopting new machine learning (ML), deep learning (DL) and related technologies (for simplicity -- “artificial intelligence”/”AI”) keeps rapidly unfolding in the pharmaceutical industry, albeit with varying rate of progress across different use cases

Let’s review retrospectively some of the key developments in the drug discovery area in 2019 and see how they characterize the current state of AI in the pharmaceutical industry (“hype vs reality”). Note, that I do not cover the healthcare sector in this post (diagnostics, medical applications of AI, digital health etc) -- those will be discussed in one of the future posts.

 

[Interview] The Rise of Quantum Physics in Drug Discovery

   by Andrii Buvailo    1527
[Interview] The Rise of Quantum Physics in Drug Discovery

Computer-aided drug design (CADD) is a central part of so-called “rational drug design”, pioneered in the last century by companies like Vertex. Over the last decades, CADD had great influence on the way new therapeutics are discovered, however, it also showed limitations due to modest accuracy of computational tools.  

The majority of software tools used for computational chemistry and biology rely on molecular mechanics -- a simplified representation of molecules, essentially reducing them down to “balls and sticks”: atoms and bonds between them. In this way it is easier to compute, but accuracy suffers greatly.

In order to gain adequate accuracy, one has to account for the electronic behavior of atoms and molecules, i.e. consider subatomic particles -- electrons and protons. This is what quantum mechanical (QM) methods are all about -- and the theory is not new, dating back to the early decades of the 20th century.