Delivering on a decade of ALS data – the role of AI in drug discovery

The use of artificial intelligence and big data in healthcare has advanced at remarkable speed over the past decade, rapidly reshaping how we diagnose, treat, and understand disease. Once experimental, AI is now becoming a trusted tool in clinical decision-making and medical innovation – playing an increasingly vital role in keeping people healthy.  

In July, a robot trained on videos of previous surgeries was able to perform surgery independently (on an animal model) with 100% accuracy. There were 17 tasks to be completed during the surgery. The robot had to identify specific ducts and arteries; grab hold of them; strategically place clips; and sever parts with scissors. It did so with total precision – an extraordinary achievement.

AI is being used to read chest X-rays to identify cancers far before the human eye notices them, helping oncologists catch and treat lung cancer earlier than ever. AI has been used to help a couple conceive a baby after 18 years of trying – the method uses AI to identify and recover hidden sperm in samples from men who were thought to be infertile.

The role of AI in drug discovery

While AI is transforming healthcare in hospitals, its potential expands to the entire health landscape, not least the discovery of new drugs.

This is a critical step for diseases where meaningful treatment options are currently limited, as is the case with many neurodegenerative diseases where cells of the central nervous system stop working or die, like Alzheimer’s Disease, Parkinson’s Disease, and amyotrophic lateral sclerosis (ALS) – the most common form of motor neurone disease (MND).

ALS is a progressive neurodegenerative disease that damages the motor neurones in the brain and spinal cord. Signals from the brain stop reaching muscles, leading to severe muscle degeneration. Eventually this affects the muscles that are used to swallow food and drink, and those used to breathe.

There are approximately 140,000 new cases of ALS diagnosed worldwide each year – 384 new cases every day. It’s predicted that ALS cases will increase from around 223,000 in 2014 to 377,000 in 2040.

Neurodegenerative diseases are extraordinarily complex. In conditions like ALS, the question of why motor neurones die in some people but not others remains one of the most urgent and difficult to answer. This complexity has long hindered the development of effective treatments, leaving patients with few therapeutic options. ALS, in particular, has no long-term treatments, and no cure.

ALS is a heterogeneous disease – meaning it has multiple genetic and environmental causes. Moreover, it involves a cascade of biological disruptions: from faulty RNA processing and protein misfolding to inflammation and cellular transport failure. These overlapping malfunctions make it nearly impossible to isolate a single cause or intervention point.

AI thrives on complexity. By analysing vast datasets across genetics, cell biology, and clinical records, AI can detect patterns and predict disease mechanisms in ways that were previously out of reach.