A team at the University of Edinburgh’s School of Engineering, led by Maria Grazia De Angelis (University of Edinburgh, Edinburgh, UK), are working on creating a wearable haemodialysis machine with the use of an alternative ‘smart’ filtration system, which could make life easier for dialysis patients.
The team received funding from Kidney Research UK to find a replacement for the current filtration process, which requires many litres of ultrapure water. The hope is that, with a new system that allows for water to be filtered and recycled, the amount of water needed can be reduced, allowing for a miniaturised, wearable haemodialysis device to be created.
De Angelis et al plan to use machine learning to work through the thousands of filter material combinations to find the one that works best for the purification process prior to testing the filter membrane in a laboratory setting.
“Machine learning accelerates work which would traditionally take years into just a matter of months,” said De Angelis. “By feeding the computer programme as much information as possible about the characteristics of the materials and the toxins that it will need to filter, we can get a quite accurate understanding of whether the chosen substance would perform well in a clinical setting. Then, testing in a lab will be used to confirm the information from the Artificial Intelligence programme by using fluids from real patients, ensuring that we can be certain that the material will be effective when introduced in a clinical trial setting.”
Whilst a new filtration system could allow for patients to be less restricted in their daily lives, there is also the possibility that this new device could reduce some of the side-effects that can come from haemodialysis, such as sickness and fatigue, by reducing the time between treatments.
The issue, according to De Angelis, is that “dialysis is designed to replicate the blood-filtering function of the kidneys, but a key difference is that a healthy kidney is working all the time while haemodialysis patients only receive treatment three times a week. By the time of their next scheduled treatment not only is their body full of toxins but their blood pressure is very high. Four hours is not long enough to be on dialysis in terms of cleaning the blood, but it is impossible for patients to have haemodialysis 24 hours a day. Having a device that is portable would allow more frequent dialysis over longer periods of time; this could drastically reduce the side-effects of the treatment. It could also enable patients to have a degree of control over their own treatment, administering it at a time that suits them.”
The hope of the team is that they can have a shortlist of materials that are the best fit for the filtration system within the next couple of years, after which they will test the most effective materials for use in a portable dialysis device.
