Abstract:Background: Corneal fluorescein staining is a key biomarker in evaluating dry eye disease. However, subjective scales of corneal fluorescein staining are lacking in consistency and increase the difficulties of an accurate diagnosis for clinicians. This study aimed to propose an automatic machine learning-based method for corneal fluorescein staining evaluation by utilizing prior information about the spatial connection and distribution of the staining region. Methods: We proposed an end-to-end automatic machine learning-based classification model that consists of staining region identification, feature signature construction, and machine learning-based classification, which fully scrutinizes the multiscale topological features together with conventional texture and morphological features. The proposed model was evaluated using retrospective data from Beijing Tongren Hospital. Two masked ophthalmologists scored images independently using the Sjögren's International Collaborative Clinical Alliance Ocular Staining Score scale. Results: A total of 382 images were enrolled in the study. A signature with six topological features, two textural features, and two morphological features was constructed after feature extraction and selection. Support vector machines showed the best classification performance (accuracy: 82.67%, area under the curve: 96.59%) with the designed signature. Meanwhile, topological features contributed more to the classification, compared with other features. According to the distribution and correlation with features and scores, topological features performed better than others. Conclusions: An automatic machine learning-based method was advanced for corneal fluorescein staining evaluation. The topological features in presenting the spatial connectivity and distribution of staining regions are essential for an efficient corneal fluorescein staining evaluation. This result implies the clinical application of topological features in dry-eye diagnosis and therapeutic effect evaluation.

The Application of Special Stain in the Pathologic Diagnosis of Corneal Dystrophy

A Comparative Study on Four Special Stains for the Amyloid of Cutaneous Amyloidosis

Application of Special Staining Techniques in Pathological Diagnosis of Fungi Infections in HIV/AIDS Patients

Deep learning-based transformation of the H&E stain into special stains

Automatic segmentation of corneal deposits from corneal stromal dystrophy images via deep learning

Morphological Differentiation and Between the Bulbar Conjunctival Lymphatics and Blood Capillaries with Enzyme Double Staining

An Automated Grading System Based on Topological Features for the Evaluation of Corneal Fluorescein Staining in Dry Eye Disease

Fluorescence Staining Vs. Routine Koh Smear For Rapid Diagnosis Of Oral Candidiasis-A Diagnostic Test

Application of an electron microscopic histochemica with ruthenium red technique

COMPARATIVE STUDY FOR THE IMMUNOHISTOCHEMICAL METHODS OF C4D IN RENAL ALLOGRAFT BIOPSIES

Accuracy of special histochemical staining methods in diagnosis of oral pathology: A systematic review and meta-analysis

Corneal histomorphology and electron microscopic observation of R124L mutated corneal dystrophy in a relapsed pedigree

Virtual stain transfer in histology via cascaded deep neural networks

[Morphological characteristics in corneal smear of acanthamoeba keratitis]

Macular Corneal Dystrophy in A Chinese Family Related with Novel Mutations of Chst6

Cornea Classic: Bron, Evans, and Smith, "Grading of Corneal and Conjunctival Staining in the Context of Other Dry Eye Tests" (2003)

[Application of direct fluorescence technique in the diagnosis of superficial mycosis]

Dual Contrastive Learning Guided Pathological Image Re-Staining.

A new staining procedure for electron microscopical cytology

Application of in vivo confocal microscopy to investigate the effect of dendritic cells in non infectious peripheral keratitis

Fluorescent periodic acid Schiff-like staining combined with standard cytologic staining of the same cytologic specimen may facilitate cytopathologic diagnosis