Neurologist uses AI to better predict how ALS progresses in different people

SINGAPORE – After the birth of her third daughter, former food technologist Huang Jing Han noticed that she was slurring her speech and had difficulty walking down the stairs.

The year was 2015, and she was 30 years old.

“It was in 2017 that I was officially diagnosed with ALS. I was in denial and tried acupuncture, other traditional Chinese medicine treatments and different types of therapy for about two years before accepting the diagnosis,” Huang, now 41, said.

ALS, or amyotrophic lateral sclerosis, is a progressive and fatal neurodegenerative disease that destroys nerve cells in the brain and spinal cord, causing a loss of muscle control that worsens over time.

In Singapore, between 250 and 400 people are living with ALS, and specialist centres like the National Neuroscience Institute (NNI) diagnose 30 to 50 new cases each year.

Also called Lou Gehrig’s disease, after the famous New York Yankees baseball player who was diagnosed with ALS in 1939, the condition is difficult to identify because there is no single definitive test and its early symptoms often mimic those of other conditions.

The course of ALS varies widely from person to person, and there is currently no treatment to prevent it, reverse it or stop it from progressing. This unpredictability makes it difficult for doctors to anticipate how the disease will progress in individual patients.

This is why neurologist Crystal Yeo from the NNI is using artificial intelligence to analyse large volumes of patient data to better predict how ALS develops in different people.

The AI-driven approach aims to ensure that patients receive care that is most relevant to their needs, allowing them and their families to focus on what matters.

Calling the main cause of ALS “sporadic”, Yeo said it is likely due to a number of factors – “genetic factors, environmental factors and certain toxin exposures”.

“We are beginning to understand that what we call ‘ALS’ is not actually just one disease (but a group of diseases with diverse underlying) biology – how it starts, develops and changes living cells, tissues and body systems.

“The field is therefore moving towards precision medicine, which targets therapies to that biology so that the right patients get the right treatments,” she said.

While ALS is becoming more prevalent, especially among Asia’s rapidly ageing population, Yeo said research carried out in Singapore found that the onset of the disease is generally five to 10 years earlier than in the Caucasian population.

“That was not the only difference. Patients in Singapore and Asia tend to live a lot longer – up to twice as long as those in Western countries. Also, the type of ALS is different, such as where it starts – the hands, feet and tongue,” she added.

She said the reason her team is using AI to predict the course of the disease is that no two patients with ALS progress at the same pace.

“Their disease may look the same when they come to see me in the clinic, but one of them would progress really quickly, while the other would not. We do not know what causes the difference between the two, nor can we predict the progress.

“Yet it is very important for us to predict (how the disease will progress) because it would help us monitor patients and help them better plan their lives,” she said.

To address this, Yeo’s team uses machine learning models to look at clinical data pulled from international sources to identify the top predictors of ALS progression.

These include blood test results, ALS clinical monitoring scores that measure functions like muscle strength and breathing, nerve conduction study findings, and proteins and genes found in damaged motor neurons grown from patients’ stem cells.

“After that, we made sure these models and their predictions were accurate not just in the US population, for instance, but also in the Singapore population. So, whatever conclusions we drew about how we predict the future of these patients to better treat or help them plan their lives are potentially applicable to ALS patients worldwide,” she said.

Her team’s clinical research and findings were published in 2025 in Muscle & Nerve, a monthly peer-reviewed medical journal covering neuromuscular medicine.

AI is also being used to predict which drugs or combinations of drugs are likely to work best for different groups of patients by analysing signals in motor neurons grown from their stem cells.

This helps doctors to tailor treatments to the right patients, allowing for earlier intervention and greater precision.

“It is about finding the right treatment for the right patient at the right time,” she added.

After nine years of living with the condition, Huang said she is well-adjusted and focused on making the most of each day.

Despite her initial fears that her three daughters would grow up without her, Huang said: “I am now able to be a role model for them and show them that we can overcome any challenge as long as we hold on to hope and faith while doing our best.”

She said she has “embraced the wheelchair and learnt to use an eye tracker to do technical analysis for day trading”.

“I don’t let what I can’t do stop me from doing what I can do. I still run a small business, and I am able to answer customer inquiries on my phone by using WhatsApp and meet customers with the help of caregivers.

“After ALS, I learnt that every day is a gift. Stay in the present and cherish each day... focus on what we have left and let go of what was lost,” Huang added.