ActiveDigestion, Kidneys & Other OrgansNIHR-supported projectComputing & AI
Development and Feasibility Evaluation of a New Standard of Care for Investigation of Patients with Kidney Stone Disease: A Multi-Omic Machine Learning Approach
Recipient organisationNIHR Newcastle Biomedical Research Centre
NIHR supportRecorded as supported by this research centre
PeriodFeb 2025 — Jan 2027
In plain English
AI plain-English summary
Half of all kidney stone patients will develop another stone within five years, yet the standard NHS workup rarely uncovers why. Current investigations focus on the stone itself, not the underlying causes. Recurrence is common, and when a stone blocks the ureter, infection can trigger sepsis—a life-threatening emergency requiring urgent surgery. Patients themselves rank avoiding these emergencies and elective operations as their top priority. Previous attempts to predict recurrence using artificial intelligence have failed, largely because they relied on small datasets and omitted two key factors: genetics and diet. This project will build a machine-learning model that integrates a patient’s genetic profile, dietary data, and clinical history to identify who is at highest risk of recurrence and why. If successful, the model could form the basis of a new standard of care—a personalised investigation pathway that flags treatable underlying conditions before the next stone forms. The immediate impact would be fewer emergency admissions, fewer operations, and better long-term management of a painful, recurrent disease. The approach is diagnostic, not fundamental science: it aims to change what happens in the clinic, not to discover new biology.
View original technical description
Kidney stone disease (KSD) is a very painful, and potentially life threatening condition that results in high anxiety and reduced quality of life for patients. Kidney stones are hard ‘stones’ that form within the urine in the kidneys. When they fall into the water pipe (ureter) between the kidney and bladder they can cause severe pain. Infections can develop behind a stone that is blocking the water pipe, which often leads to sepsis, a life threatening condition. This requires emergency surgery to relieve the infection. Our PPI (Patient and Public Involvement) group have stated that avoidance of emergencies, as well as elective operations are their primary interests. Up to half of patients will experience a recurrence of their disease after 5 years. There are a large number of conditions which lead to higher risk of further (recurrent) kidney stones. Unfortunately, the current standard of care does not adequately assess for these conditions. There is therefore a need to develop a new standard of care for people with kidney stones to investigate whether there is an underlying cause for their disease. This is backed up by our PPI group. Artificial intelligence (AI - also known as Machine Learning) has been used in recent years to predict outcomes of kidney stone operations. However, previous AI studies aiming to predict recurrence have been unsuccessful. This is likely due to small datasets and lack of relevant information. There is increasing evidence that kidney stone disease can be due to genetics and diet, amongst other factors. No studies that have built AI models for recurrence have included genetics or diet.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know