Region of interest-based predictive algorithm for subretinal hemorrhage detection using faster R-CNN

Abstract

Macular edema (ME) is an essential sort of macular issue caused due to the storing of fluid underneath the macula. Age-related Macular Degeneration (AMD) and diabetic macular edema (DME) are the two customary visual contaminations that can lead to fragmentary or complete vision loss. This paper proposes a deep learning-based predictive algorithm that can be used to detect the presence of a Subretinal hemorrhage. Region Convolutional Neural Network (R-CNN) and faster R-CNN are used to develop the predictive algorithm that can improve the classification accuracy. This method initially detects the presence of Subretinal hemorrhage, and it then segments the Region of Interest (ROI) by a semantic segmentation process. The segmented ROI is applied to a predictive algorithm which is derived from the Fast Region Convolutional Neural Network algorithm, that can categorize the Subretinal hemorrhage as responsive or non-responsive. The dataset, provided by a medical institution, comprised of optical coherence tomography (OCT) images of both pre- and post-treatment images, was used for training the proposed Faster Region Convolutional Neural Network (Faster R-CNN). We also used the Kaggle dataset for performance comparison with the traditional methods that are derived from the convolutional neural network (CNN) algorithm. The evaluation results using the Kaggle dataset and the hospital images provide an average sensitivity, selectivity, and accuracy of 85.3%, 89.64%, and 93.48% respectively. Further, the proposed method provides a time complexity in testing as 2.64s, which is less than the traditional schemes like CNN, R-CNN, and Fast R-CNN.

Pipelined Structure in the Classification of Skin Lesions Based on Alexnet CNN and SVM Model With Bi-Sectional Texture Features

“…Artificial intelligence plays an ultimate role in the automatic classification of medical images of different modalities [7], [8]. The scheme [9] first divided the image through distinct component analysis into pigment and blood elements before segmenting the vascular architecture of the lesion.…”

Robust periocular recognition in training of CNN models using HOG- based gradient approach

“…Proenca et al, [5] made use of the contour of the eyelid, whereas Le et al [6] made use of the form of the brow. Zhao et al [7] convolutional . 's neural net technique trains the CNN model with extra information such as race and gender.…”

“…Bakshi et al used mobile biometrics with lower face intensity localized patterns. Zhao and Kumar [7] proposed a ROI detection network and a sustained attention module for trading and matching that prioritises key components. Chen et al [8] used five separate local characteristics, including those taken from the regions between the eye brow and the upper eyelid, the distances between it, intensities in the upper portion of the inner corner, textural characteristics from the tears duct regions, and iris colour.…”

Big Data Classification of Ultrasound Doppler Scan Images Using a Decision Tree Classifier Based on Maximally Stable Region Feature Points

“…Deep learning algorithms are preferred over machine learning algorithms where it is difficult to identify the ROI or in extracting the features from the ROI. Convolutional Neural Network (CNN)plays a crucial role in deep learning whichis used in different applications such as MRI imaging [9], fundus image [10], dermatology [11], Optical Coherence Tomography (OCT) [12], radiology [13], etc. The CNN will estimate the complex patterns when the network is trained.…”