Artificial intelligence on diabetic retinopathy diagnosis: an automatic classification method based on grey level co-occurrence matrix and naive Bayesian model

Journal Title: International Journal of Ophthalmology - Year 2019, Vol 12, Issue 7

Abstract

"AIM: To develop an automatic tool on screening diabetic retinopathy (DR) from diabetic patients. METHODS: We extracted textures from eye fundus images of each diabetes subject using grey level co-occurrence matrix method and trained a Bayesian model based on these textures. The receiver operating characteristic (ROC) curve was used to estimate the sensitivity and specificity of the Bayesian model. RESULTS: A total of 1000 eyes fundus images from diabetic patients in which 298 eyes were diagnosed as DR by two ophthalmologists. The Bayesian model was trained using four extracted textures including contrast, entropy, angular second moment and correlation using a training dataset. The Bayesian model achieved a sensitivity of 0.949 and a specificity of 0.928 in the validation dataset. The area under the ROC curve was 0.938, and the 10-fold cross validation method showed that the average accuracy rate is 93.5%. CONCLUSION: Textures extracted by grey level co-occurrence can be useful information for DR diagnosis, and a trained Bayesian model based on these textures can be an effective tool for DR screening among diabetic patients."

Authors and Affiliations

Kai Cao

Keywords

Related Articles

Peptidome profiling of human serum of uveal melanoma patients based on magnetic bead fractionation and mass spectrometry

"AIM: To find new biomarkers for uveal melanoma (UM) by analyzing the serum peptidome profile. METHODS: Proteomic spectra in patients with UM before and after operation were analyzed and compared with those of healthy co...

Differentiation potential of human adipose tissue derived stem cells into photoreceptors through explants culture and enzyme methods

"AIM: To investigate the retinal photoreceptor differentiation potential of human orbital adipose tissue-derived stem cells (ADSCs) generated by enzyme (EN) and explant (EX) culture methods. METHODS: We investigated pote...

Accommodation function comparison following use of contact lens for orthokeratology and spectacle use in myopic children: a prospective controlled trial

"AIM: To study effects of orthokeratology (Ortho-k) on accommodation function in myopic children. METHODS: A prospective, non-randomized, case-control study was performed from September to October 2016. Eighty-three chil...

Roles of the eye care workforce for task sharing in management of diabetic retinopathy in Cambodia

"AIM: To identify the current roles of eye and health care workers in eye care delivery and investigate their potential roles in screening and detection for management of diabetic retinopathy (DR) through task sharing. M...

The efficacy and safety of besifloxacin for acute bacterial conjunctivitis: a Meta-analysis

"AIM: To evaluate the relative efficacy and safety of besifloxacin for treatment of acute bacterial conjunctivitis. METHODS: A comprehensive search in PubMed, EMBASE Web of Science, Cochrane Central Database and CNKI was...

Download PDF file
  • EP ID EP648114
  • DOI 10.18240/ijo.2019.07.17
  • Views 64
  • Downloads 0

How To Cite

Kai Cao (2019). Artificial intelligence on diabetic retinopathy diagnosis: an automatic classification method based on grey level co-occurrence matrix and naive Bayesian model. International Journal of Ophthalmology, 12(7), 1158-1162. https://europub.co.uk/articles/-A-648114