A pioneering AI tool developed at the University of Sheffield could transform care for people living with Motor Neurone Disease (MND) by accurately predicting when a patient will need a feeding tube.
The breakthrough model helps clinicians pinpoint the optimal time for a gastrostomy – a critical procedure that can preserve nutrition, protect quality of life and even extend survival.
Why Timing Matters in MND
Motor Neurone Disease – also known as Amyotrophic Lateral Sclerosis – is a progressive and fatal condition that attacks the nerve cells controlling muscles.
As the disease advances, many patients lose the ability to swallow safely. This can lead to severe weight loss and malnutrition. A gastrostomy involves inserting a feeding tube directly into the stomach to maintain nutrition.
However, the timing is crucial.
If carried out too early, the procedure can negatively impact quality of life. If done too late, risks increase significantly. Patients may enter a “refractory” stage of malnutrition, or the surgery may become impossible due to weakened breathing muscles.
Until now, clinicians have had no reliable way to predict when the intervention will be needed. For some patients it may be eight months after diagnosis. For others, it could be 20 years.
AI Model Reduces Uncertainty
Researchers from across Europe, led by Professor Johnathan Cooper-Knock at the University of Sheffield’s Institute for Translational Neuroscience (SITraN), developed a sophisticated machine learning model to address MND’s unpredictable progression.
The AI tool uses routine clinical data collected at diagnosis to estimate how quickly the disease will progress in each individual. It then predicts when significant weight loss is likely to occur — a key indicator that a feeding tube is required.
Using data from more than 20,000 MND patients, the model achieved:
A median prediction error of just 3.7 months at diagnosis
Improved accuracy to 2.6 months when patients were reassessed six months later
Less than one month error in the largest subgroup of patients
The tool has also demonstrated stable, reliable performance across diverse patient populations in the US, Europe and Sweden.
“From Reactive to Proactive Care”
Professor Cooper-Knock said:
“One of the hardest aspects of living with MND is the uncertainty. It is a cruel and devastating disease.
Until now it has been impossible for clinicians to predict when someone living with MND may need a feeding tube. By pinpointing the optimal window for a gastrostomy to within three months, doctors and patients can better plan for the surgery.
This is not just about a surgical procedure; it’s about preserving a patient’s dignity and ability to maintain nutrition safely. Knowing this critical window allows us to move from reacting to the disease’s progression to proactively managing it.”
Published in Leading Medical Journal
The promising findings have been published in eBioMedicine.
Researchers are now planning a prospective clinical trial to formally validate the tool before it can become a standard part of MND care.
If successful, the AI system could ensure patients receive the right care at the right time – maximising both quality of life and survival.
