Automatic tool for quantification of nerve fibres in corneal confocal microscopy images

Summary Failure after failure of phase 3 clinical trials in diabetic neuropathy has led to the realisation that potentially effective treatments have failed because of inadequate end‐points in clinical trials. Our pioneering studies showed that corneal confocal microscopy (CCM) had excellent diagnostic potential in diabetic somatic and autonomic neuropathy. We and others subsequently used CCM to diagnose peripheral neuropathies in obesity, impaired glucose tolerance, hereditary neuropathy (CMT1A, Friedreich's ataxia), Fabry's disease, amyloid neuropathy, CIDP, HIV etc. Of significant potential as a surrogate end‐point, we have shown that CCM detects nerve fibre regeneration within 6 months of pancreas transplantation in diabetic patients and after 24 months of an improvement in glycaemic control, blood pressure and lipids, continuous subcutaneous insulin infusion and after bariatric surgery. We have shown nerve repair in several phase 2b trials of ARA290, a peptide derivative of erythropoietin, in patients with sarcoid neuropathy and diabetic neuropathy. We have also shown that CCM detects axonal pathology which relates to the severity of neurological deficits in patients with Parkinson's disease, multiple sclerosis and dementia.