Dataset Information

CNN-based diagnosis models for canine ulcerative keratitis.

ABSTRACT: The purpose of this methodological study was to develop a convolutional neural network (CNN), which is a recently developed deep-learning-based image recognition method, to determine corneal ulcer severity in dogs. The CNN model was trained with images for which corneal ulcer severity (normal, superficial, and deep) were previously classified by veterinary ophthalmologists' diagnostic evaluations of corneal photographs from patients who visited the Veterinary Medical Teaching Hospital (VMTH) at Konkuk University and 3 different veterinary ophthalmology specialty hospitals in Korea. The original images (depicting normal corneas (36) and corneas with superficial (47) ulcers, deep (47) ulcers), flipped images (total 520), rotated images (total 520), and both flipped and rotated images (total 1,040) were labeled, learned and evaluated with GoogLeNet, ResNet, and VGGNet models, and the severity of each corneal ulcer image was determined. To accomplish this task, models based on TensorFlow, an open-source software library developed by Google, were used, and the labeled images were converted into TensorFlow record (TFRecord) format. The models were fine-tuned using a CNN model trained on the ImageNet dataset and then used to predict severity. Most of the models achieved accuracies of over 90% when classifying superficial and deep corneal ulcers, and ResNet and VGGNet achieved accuracies over 90% for classifying normal corneas, corneas with superficial ulcers, and corneas with deep ulcers. This study proposes a method to effectively determine corneal ulcer severity in dogs by using a CNN and concludes that multiple image classification models can be used in the veterinary field.

SUBMITTER: Kim JY

PROVIDER: S-EPMC6775068 | biostudies-literature | 2019 Oct

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

CNN-based diagnosis models for canine ulcerative keratitis.

Kim Joon Young JY Lee Ha Eun HE Choi Yeon Hyung YH Lee Suk Jun SJ Jeon Jong Soo JS

Scientific reports 20191002 1

The purpose of this methodological study was to develop a convolutional neural network (CNN), which is a recently developed deep-learning-based image recognition method, to determine corneal ulcer severity in dogs. The CNN model was trained with images for which corneal ulcer severity (normal, superficial, and deep) were previously classified by veterinary ophthalmologists' diagnostic evaluations of corneal photographs from patients who visited the Veterinary Medical Teaching Hospital (VMTH) at ...[more]

PMID: 31578338

Similar Datasets

Project description:Canine Chronic Ulcerative Stomatitis is a spontaneously occurring inflammatory disease of the oral mucosa. An immune-mediated pathogenesis is suspected though not yet proven. We have recently reported on the clinical and histologic features, and identification of select leukocyte cell populations within the lesion. A clinical and histologic similarity to oral lichen planus of people was proposed. In the present study, these initial observations are extended by examining lesions from 24 dogs with clinical evidence of chronic ulcerative stomatitis. Because dogs with chronic ulcerative stomatitis often have concurrent periodontal disease, we wondered if dental plaque/biofilm may be a common instigator of inflammation in both lesions. We hypothesized that dogs with chronic ulcerative stomatitis would exhibit a spectrum of pathologic changes and phenotype of infiltrating leukocytes that would inform lesion pathogenesis and that these changes would differ from inflammatory phenotypes in periodontitis. Previously we identified chronic ulcerative stomatitis lesions to be rich in FoxP3+ and IL17+ cells. As such, we suspect that these leukocytes play an important role in lesion pathogenesis. The current study confirms the presence of moderate to large numbers of FoxP3+ T cells and IL17+ cells in all ulcerative stomatitis lesions using confocal immunofluorescence. Interestingly, the majority of IL17+ cells were determined to be non-T cells and IL17+ cell frequencies were negatively correlated with severity on the clinical scoring system. Three histologic subtypes of ulcerative stomatitis were determined; lichenoid, deep stomatitis and granulomatous. Periodontitis lesions, like stomatitis lesions, were B cell and plasma cell rich, but otherwise differed from the stomatitis lesions. Direct immunofluorescence results did not support an autoantibody-mediated autoimmune disease process. This investigation contributes to the body of literature regarding leukocyte involvement in canine idiopathic inflammatory disease pathogenesis.

Dataset Information

CNN-based diagnosis models for canine ulcerative keratitis.

Publications

CNN-based diagnosis models for canine ulcerative keratitis.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets