Regional choroidal thickness estimation from color fundus images based on convolutional neural networks.

Purpose: This study aims to estimate the regional choroidal thickness from color fundus images from convolutional neural networks in different network structures and task learning models. Method: 1276 color fundus photos and their corresponding choroidal thickness values from healthy subjects were obtained from the Topcon DRI Triton optical coherence tomography machine. Initially, ten commonly used convolutional neural networks were deployed to identify the most accurate model, which was subsequently selected for further training. This selected model was then employed in combination with single-, multiple-, and auxiliary-task training models to predict the average and sub-region choroidal thickness in both ETDRS (Early Treatment Diabetic Retinopathy Study) grids and 100-grid subregions. The values of mean absolute error and coefficient of determination (R2) were involved to evaluate the models' performance. Results: Efficientnet-b0 network outperformed other networks with the lowest mean absolute error value (25.61 µm) and highest R2 (0.7817) in average choroidal thickness. Incorporating diopter spherical, anterior chamber depth, and lens thickness as auxiliary tasks improved predicted accuracy (p-value = 6.39×10-44, 2.72×10-38, 1.15×10-36 respectively). For ETDRS regional choroidal thickness estimation, multi-task model achieved better results than single task model (lowest mean absolute error = 31.10 µm vs. 33.20 µm). The multi-task training also can simultaneously predict the choroidal thickness of 100 grids with a minimum mean absolute error of 33.86 µm. Conclusions: Efficientnet-b0, in combination with multi-task and auxiliary task models, achieve high accuracy in estimating average and regional macular choroidal thickness directly from color fundus photographs.

Segmentation of retinal vessels in fundus images based on U-Net with self-calibrated convolutions and spatial attention modules.

Automated and accurate segmentation of retinal vessels in fundus images is an important step for screening and diagnosing various ophthalmologic diseases. However, many factors, including the variations of vessels in color, shape and size, make this task become an intricate challenge. One kind of the most popular methods for vessel segmentation is U-Net based methods. However, in the U-Net based methods, the size of the convolution kernels is generally fixed. As a result, the receptive field for an individual convolution operation is single, which is not conducive to the segmentation of retinal vessels with various thicknesses. To overcome this problem, in this paper, we employed self-calibrated convolutions to replace the traditional convolutions for the U-Net, which can make the U-Net learn discriminative representations from different receptive fields. Besides, we proposed an improved spatial attention module, instead of using traditional convolutions, to connect the encoding part and decoding part of the U-Net, which can improve the ability of the U-Net to detect thin vessels. The proposed method has been tested on Digital Retinal Images for Vessel Extraction (DRIVE) database and Child Heart and Health Study in England Database (CHASE DB1). The metrics used to evaluate the performance of the proposed method are accuracy (ACC), sensitivity (SE), specificity (SP), F1-score (F1) and the area under the receiver operating characteristic curve (AUC). The ACC, SE, SP, F1 and AUC obtained by the proposed method are 0.9680, 0.8036, 0.9840, 0.8138 and 0.9840 respectively on DRIVE database, and 0.9756, 0.8118, 0.9867, 0.8068 and 0.9888 respectively on CHASE DB1, which are better than those obtained by the traditional U-Net (the ACC, SE, SP, F1 and AUC obtained by U-Net are 0.9646, 0.7895, 0.9814, 0.7963 and 0.9791 respectively on DRIVE database, and 0.9733, 0.7817, 0.9862, 0.7870 and 0.9810 respectively on CHASE DB1). The experimental results indicate that the proposed modifications in the U-Net are effective for vessel segmentation. The structure of the proposed network.

Direct Estimation of Choroidal Thickness in Optical Coherence Tomography Images with Convolutional Neural Networks.

Automatic and accurate estimation of choroidal thickness plays a very important role in a computer-aided system for eye diseases. One of the most common methods for automatic estimation of choroidal thickness is segmentation-based methods, in which the boundaries of the choroid are first detected from optical coherence tomography (OCT) images. The choroidal thickness is then computed based on the detected boundaries. A shortcoming in the segmentation-based methods is that the estimating precision greatly depends on the segmentation results. To avoid the dependence on the segmentation step, in this paper, we propose a direct method based on convolutional neural networks (CNNs) for estimating choroidal thickness without segmentation. Concretely, a B-scan image is first cropped into several patches. A trained CNN model is then used to estimate the choroidal thickness for each patch. The mean thickness of the choroid in the B-scan is obtained by taking the average of the choroidal thickness on each patch. Then, 150 OCT volumes are collected to evaluate the proposed method. The experiments show that the results obtained by the proposed method are very competitive with those obtained by segmentation-based methods, which indicates that direct estimation of choroidal thickness is very promising.

Genetic U-Net: Automatically Designed Deep Networks for Retinal Vessel Segmentation Using a Genetic Algorithm.

Recently, many methods based on hand-designed convolutional neural networks (CNNs) have achieved promising results in automatic retinal vessel segmentation. However, these CNNs remain constrained in capturing retinal vessels in complex fundus images. To improve their segmentation performance, these CNNs tend to have many parameters, which may lead to overfitting and high computational complexity. Moreover, the manual design of competitive CNNs is time-consuming and requires extensive empirical knowledge. Herein, a novel automated design method, called Genetic U-Net, is proposed to generate a U-shaped CNN that can achieve better retinal vessel segmentation but with fewer architecture-based parameters, thereby addressing the above issues. First, we devised a condensed but flexible search space based on a U-shaped encoder-decoder. Then, we used an improved genetic algorithm to identify better-performing architectures in the search space and investigated the possibility of finding a superior network architecture with fewer parameters. The experimental results show that the architecture obtained using the proposed method offered a superior performance with less than 1% of the number of the original U-Net parameters in particular and with significantly fewer parameters than other state-of-the-art models. Furthermore, through in-depth investigation of the experimental results, several effective operations and patterns of networks to generate superior retinal vessel segmentations were identified. The codes of this work are available at https://github.com/96jhwei/Genetic-U-Net.

Modified Hilbert Curve for Rectangles and Cuboids and Its Application in Entropy Coding for Image and Video Compression.

In our previous work, by combining the Hilbert scan with the symbol grouping method, efficient run-length-based entropy coding was developed, and high-efficiency image compression algorithms based on the entropy coding were obtained. However, the 2-D Hilbert curves, which are a critical part of the above-mentioned entropy coding, are defined on squares with the side length being the powers of 2, i.e., 2n, while a subband is normally a rectangle of arbitrary sizes. It is not straightforward to modify the Hilbert curve from squares of side lengths of 2n to an arbitrary rectangle. In this short article, we provide the details of constructing the modified 2-D Hilbert curve of arbitrary rectangle sizes. Furthermore, we extend the method from a 2-D rectangle to a 3-D cuboid. The 3-D modified Hilbert curves are used in a novel 3-D transform video compression algorithm that employs the run-length-based entropy coding. Additionally, the modified 2-D and 3-D Hilbert curves introduced in this short article could be useful for some unknown applications in the future.

Real-time Prediction of the Daily Incidence of COVID-19 in 215 Countries and Territories Using Machine Learning: Model Development and Validation.

BACKGROUND: Advanced prediction of the daily incidence of COVID-19 can aid policy making on the prevention of disease spread, which can profoundly affect people's livelihood. In previous studies, predictions were investigated for single or several countries and territories. OBJECTIVE: We aimed to develop models that can be applied for real-time prediction of COVID-19 activity in all individual countries and territories worldwide. METHODS: Data of the previous daily incidence and infoveillance data (search volume data via Google Trends) from 215 individual countries and territories were collected. A random forest regression algorithm was used to train models to predict the daily new confirmed cases 7 days ahead. Several methods were used to optimize the models, including clustering the countries and territories, selecting features according to the importance scores, performing multiple-step forecasting, and upgrading the models at regular intervals. The performance of the models was assessed using the mean absolute error (MAE), root mean square error (RMSE), Pearson correlation coefficient, and Spearman correlation coefficient. RESULTS: Our models can accurately predict the daily new confirmed cases of COVID-19 in most countries and territories. Of the 215 countries and territories under study, 198 (92.1%) had MAEs <10 and 187 (87.0%) had Pearson correlation coefficients >0.8. For the 215 countries and territories, the mean MAE was 5.42 (range 0.26-15.32), the mean RMSE was 9.27 (range 1.81-24.40), the mean Pearson correlation coefficient was 0.89 (range 0.08-0.99), and the mean Spearman correlation coefficient was 0.84 (range 0.2-1.00). CONCLUSIONS: By integrating previous incidence and Google Trends data, our machine learning algorithm was able to predict the incidence of COVID-19 in most individual countries and territories accurately 7 days ahead.

Retrospective analysis of the accuracy of predicting the alert level of COVID-19 in 202 countries using Google Trends and machine learning.

BACKGROUND: Internet search engine data, such as Google Trends, was shown to be correlated with the incidence of COVID-19, but only in several countries. We aim to develop a model from a small number of countries to predict the epidemic alert level in all the countries worldwide. METHODS: The "interest over time" and "interest by region" Google Trends data of Coronavirus, pneumonia, and six COVID symptom-related terms were searched. The daily incidence of COVID-19 from 10 January to 23 April 2020 of 202 countries was retrieved from the World Health Organization. Three alert levels were defined. Ten weeks' data from 20 countries were used for training with machine learning algorithms. The features were selected according to the correlation and importance. The model was then tested on 2830 samples of 202 countries. RESULTS: Our model performed well in 154 (76.2%) countries, of which each had no more than four misclassified samples. In these 154 countries, the accuracy was 0.8133, and the kappa coefficient was 0.6828. While in all 202 countries, the accuracy was 0.7527, and the kappa coefficient was 0.5841. The proposed algorithm based on Random Forest Classification and nine features performed better compared to other machine learning methods and the models with different numbers of features. CONCLUSIONS: Our result suggested that the model developed from 20 countries with Google Trends data and Random Forest Classification can be applied to predict the epidemic alert levels of most countries worldwide.

Deriving external forces via convolutional neural networks for biomedical image segmentation.

Active contours, or snakes, are widely applied on biomedical image segmentation. They are curves defined within an image domain that can move to object boundaries under the influence of internal forces and external forces, in which the internal forces are generally computed from curves themselves and external forces from image data. Designing external forces properly is a key point with active contour algorithms since the external forces play a leading role in the evolution of active contours. One of most popular external forces for active contour models is gradient vector flow (GVF). However, GVF is sensitive to noise and false edges, which limits its application area. To handle this problem, in this paper, we propose using GVF as reference to train a convolutional neural network to derive an external force. The derived external force is then integrated into the active contour models for curve evolution. Three clinical applications, segmentation of optic disk in fundus images, fluid in retinal optical coherence tomography images and fetal head in ultrasound images, are employed to evaluate the proposed method. The results show that the proposed method is very promising since it achieves competitive performance for different tasks compared to the state-of-the-art algorithms.

Surrogate-Assisted Retinal OCT Image Classification Based on Convolutional Neural Networks.

Optical Coherence Tomography (OCT) is beco-ming one of the most important modalities for the noninvasive assessment of retinal eye diseases. As the number of acquired OCT volumes increases, automating the OCT image analysis is becoming increasingly relevant. In this paper, we propose a surrogate-assisted classification method to classify retinal OCT images automatically based on convolutional neural networks (CNNs). Image denoising is first performed to reduce the noise. Thresholding and morphological dilation are applied to extract the masks. The denoised images and the masks are then employed to generate a lot of surrogate images, which are used to train the CNN model. Finally, the prediction for a test image is determined by the average of the outputs from the trained CNN model on the surrogate images. The proposed method has been evaluated on different databases. The results (AUC of 0.9783 in the local database and AUC of 0.9856 in the Duke database) show that the proposed method is a very promising tool for classifying the retinal OCT images automatically.

Optic Disk Detection in Fundus Image Based on Structured Learning.

Automated optic disk (OD) detection plays an important role in developing a computer aided system for eye diseases. In this paper, we propose an algorithm for the OD detection based on structured learning. A classifier model is trained based on structured learning. Then, we use the model to achieve the edge map of OD. Thresholding is performed on the edge map, thus a binary image of the OD is obtained. Finally, circle Hough transform is carried out to approximate the boundary of OD by a circle. The proposed algorithm has been evaluated on three public datasets and obtained promising results. The results (an area overlap and Dices coefficients of 0.8605 and 0.9181, respectively, an accuracy of 0.9777, and a true positive and false positive fraction of 0.9183 and 0.0102) show that the proposed method is very competitive with the state-of-the-art methods and is a reliable tool for the segmentation of OD.