Method for diagnosing keratoconus cases based on machine learning

Abstract

The invention relates to a method for diagnosing keratoconus cases based on machine learning. According to the method, a support vector machines-recursive feature elimination (SVM-RFE) algorithm and agradient boosting regression tree (ABDT) algorithm are applied to accurate diagnosis of the keratoconus cases. Furthermore for aiming at specific application cases, effective overall plan designing,process designing and algorithm parameter setting are performed. Through test of a large number of clinic instances, the diagnosis accuracy of the method is effectively improved and basically satisfies the requirement for clinic applications.

Description

technical field [0001] The invention belongs to the field of ophthalmic medical diagnosis and relates to machine learning technology, in particular to a method for diagnosing keratoconus cases based on machine learning. Background technique [0002] Keratoconus refers to the progressive thinning of the cornea in the central or paracentral areas of the cornea, and corneal ectasia with a conical bulge. It usually occurs in young people around the age of 20, especially in young male patients. It often causes high irregularities. Astigmatism and varying degrees of visual impairment have a poor prognosis and eventually require treatment such as corneal transplantation. Because keratoconus has a great impact on vision and visual function, early screening and early treatment intervention are the key; however, earlier intervention brings greater diagnostic challenges. How to accurately identify early corneal dilated changes is more important than determining advanced disease challe...

AI Technical Summary

Problems solved by technology

Because keratoconus has a great impact on vision and visual function, early screening and early treatment intervention are the key; however, earlier intervention brings greater diagnostic challenges. How to accurately identify early corneal dilated changes is more important than determining advanced disease challenge

At present, there is no unified standard for early screening of keratoconus, and there is a lot of controversy. The difference in different parameters has caused great confusion to clinicians. A relatively accurate diagnosis can only be given after detailed consultation and analysis by experienced experts; At the same time, a large number of cases, limited experts, and complicated corneal parameters have added great difficulty to the early diagnosis of keratoconus.

Images