Sex judgment using color fundus parameters in elementary school students

Noma, Saki; Yamashita, Takehiro; Asaoka, Ryo; Terasaki, Hiroto; Yoshihara, Naoya; Kakiuchi, Naoko; Sakamoto, Taiji

doi:10.1007/s00417-020-04969-1

Cited by 6 publications

(12 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We observed in Visual explanations for DOVS-ii classifiers that mean GGCAM behaviour for a matching class label ( e.g .,“Female” image, “Female” class label) tends to have activation in image regions that have physiologically salient components ( e.g ., vasculature and/or optic disk), some of which have previously been identified as potentially relevant markers for sex determination from fundus images [ 32 ]. Moreover, in contrast, mean GGCAM activation for an opposite class label ( e.g ., “Female” image, “Male” class label) tends not to have activation in physiologically salient image regions ( e.g ., at the border of the background image and the fundus).…”

Section: Discussionmentioning

confidence: 99%

Learning from small data: Classifying sex from retinal images via deep learning

et al. 2023

View full text Add to dashboard Cite

Deep learning (DL) techniques have seen tremendous interest in medical imaging, particularly in the use of convolutional neural networks (CNNs) for the development of automated diagnostic tools. The facility of its non-invasive acquisition makes retinal fundus imaging particularly amenable to such automated approaches. Recent work in the analysis of fundus images using CNNs relies on access to massive datasets for training and validation, composed of hundreds of thousands of images. However, data residency and data privacy restrictions stymie the applicability of this approach in medical settings where patient confidentiality is a mandate. Here, we showcase results for the performance of DL on small datasets to classify patient sex from fundus images—a trait thought not to be present or quantifiable in fundus images until recently. Specifically, we fine-tune a Resnet-152 model whose last layer has been modified to a fully-connected layer for binary classification. We carried out several experiments to assess performance in the small dataset context using one private (DOVS) and one public (ODIR) data source. Our models, developed using approximately 2500 fundus images, achieved test AUC scores of up to 0.72 (95% CI: [0.67, 0.77]). This corresponds to a mere 25% decrease in performance despite a nearly 1000-fold decrease in the dataset size compared to prior results in the literature. Our results show that binary classification, even with a hard task such as sex categorization from retinal fundus images, is possible with very small datasets. Our domain adaptation results show that models trained with one distribution of images may generalize well to an independent external source, as in the case of models trained on DOVS and tested on ODIR. Our results also show that eliminating poor quality images may hamper training of the CNN due to reducing the already small dataset size even further. Nevertheless, using high quality images may be an important factor as evidenced by superior generalizability of results in the domain adaptation experiments. Finally, our work shows that ensembling is an important tool in maximizing performance of deep CNNs in the context of small development datasets.

show abstract

Section: Discussionmentioning

confidence: 99%

Learning from small data: Classifying sex from retinal images via deep learning

et al. 2023

View full text Add to dashboard Cite

show abstract

“…The reason for the difference between the sexes in the temperature at the mid-vitreous was not determined. Sex differences in the location of the retinal vessels have been reported [ 17 ], and the heat retention effect of blood circulation may be different, but further research on the sex differences in the intraocular temperatures is needed.…”

Section: Discussionmentioning

confidence: 99%

Intraocular Temperature at Different Sites in Eye Measured at the Beginning of Vitreous Surgery

Shinoda

Yagura

Matsumoto

et al. 2021

JCM

View full text Add to dashboard Cite

The temperature of the vitreous has been reported to vary during cataract and vitreous surgery. We measured intraocular temperature at four intraocular sites; the anterior chamber (AC), just behind the crystalline lens, mid-vitreous, and just anterior to the optic disc (OD) at the beginning of vitrectomy with a thermoprobe in 48 eyes. The temperatures were compared in three groups; eyes that underwent vitrectomy for the first time (Group V, n = 30), eyes that had previous vitrectomy and the vitreous cavity had been filled with balanced salt solution (BSS; Group A, n = 12), and eyes that had previous vitrectomy and the vitreous cavity was filled with silicone oil (Group S, n = 6). There was a gradient in the temperature in all groups, i.e., it was lowest in the AC, and it increased at points closer to the retina. The intraocular temperature was significantly correlated with the type of fluid in the vitreous cavity. The mean intraocular temperatures were not significantly different in Groups V and A, but they were significantly higher in Group S. Clinicians should be aware of the differences in the temperature at the different intraocular sites because the temperatures may affect the physiology of the retina and the recovery process.

show abstract

“…This study was part of a longitudinal, prospective, cross-sectional, observational study of third grade students who were 8–9 years old at the first examination [ 23 , 34 ]. The students attended the Elementary School of the Faculty of Education of Kagoshima University.…”

Section: Methodsmentioning

confidence: 99%

“…All these measurements were conducted using CFP images and the ImageJ software (ImageJ version 1.47, National Institutes of Health, Bethesda, MD; available at: http://imagej.nih.gov/ij/ ). The macro function of ImageJ enabled semiautomated calculation of the above-described CFP parameters; all the 54 parameters were automatically calculated when the locations of the fovea, center of the optic nerve head, and crossing points of supralateral or inferolateral RA or RV were decided [ 34 ].…”

Section: Methodsmentioning

confidence: 99%

“…We also reported a sex discrimination accuracy in children (8.5 years old), where the AUROC was 63.2% when same ridge binomial logistic regression approach was used [ 34 ]. It was found that the fundus of female children had a more oval-shaped disc and higher green and blue intensities on the nasal side of optic disc than the fundus of male children.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Three-year changes in sex judgment using color fundus parameters in elementary school students

Yamashita,

Asaoka,

Terasaki

et al. 2023

PLoS ONE

Self Cite

View full text Add to dashboard Cite

Purpose In a previous cross-sectional study, we reported that the sexes can be distinguished using known factors obtained from color fundus photography (CFP). However, it is not clear how sex differences in fundus parameters appear across the human lifespan. Therefore, we conducted a cohort study to investigate sex determination based on fundus parameters in elementary school students. Methods This prospective observational longitudinal study investigated 109 right eyes of elementary school students over 4 years (age, 8.5 to 11.5 years). From each CFP, the tessellation fundus index was calculated as red/red + green + blue (R/[R+G+B]) using the mean value of red-green-blue intensity in eight locations around the optic disc and macular region. Optic disc area, ovality ratio, papillomacular angle, and retinal vessel angles and distances were quantified according to the data in our previous report. Using 54 fundus parameters, sex was predicted by L2 regularized binomial logistic regression for each grade. Results The right eyes of 53 boys and 56 girls were analyzed. The discrimination accuracy rate significantly increased with age: 56.3% at 8.5 years, 46.1% at 9.5 years, 65.5% at 10.5 years and 73.1% at 11.5 years. Conclusions The accuracy of sex discrimination by fundus photography improved during a 3-year cohort study of elementary school students.

show abstract

Sex judgment using color fundus parameters in elementary school students

Cited by 6 publications

References 49 publications

Learning from small data: Classifying sex from retinal images via deep learning

Learning from small data: Classifying sex from retinal images via deep learning

Intraocular Temperature at Different Sites in Eye Measured at the Beginning of Vitreous Surgery

Three-year changes in sex judgment using color fundus parameters in elementary school students

Contact Info

Product

Resources

About