Automatic Feature Learning to Grade Nuclear Cataracts Based on Deep Learning

  title={Automatic Feature Learning to Grade Nuclear Cataracts Based on Deep Learning},
  author={Xinting Gao and Stephen Lin and Tien Yin Wong},
  journal={IEEE Transactions on Biomedical Engineering},
Goal: Cataracts are a clouding of the lens and the leading cause of blindness worldwide. [] Key Method Local filters are first acquired through clustering of image patches from lenses within the same grading class. The learned filters are fed into a convolutional neural network, followed by a set of recursive neural networks, to further extract higher order features. With these features, support vector regression is applied to determine the cataract grade. Results: The proposed system is validated on a large…

Automatic cataract detection and grading using Deep Convolutional Neural Network

It is demonstrated in this paper that the DCNN classifier outperforms state-of-the-art in the performance, and has the potential to be applied to other eye diseases in future.

A Hybrid Global-Local Representation CNN Model for Automatic Cataract Grading

Compared to the global feature set, the detail vascular information, which is lost after multi-layer convolution calculation also plays an important role in cataract grading task, and this finding fits with the morphological definition of fundus image.

CataractNet: An Automated Cataract Detection System Using Deep Learning for Fundus Images

A novel deep neural network, namely CataractNet, is proposed for automatic cataract detection in fundus images with an average accuracy of 99.13% and the computational cost and average running time are significantly reduced compared to other pre-trained Convolutional Neural Network models.

A Novel Deep Learning Method for Nuclear Cataract Classification Based on Anterior Segment Optical Coherence Tomography Images

A novel Convolutional Neural Network model named GraNet is proposed for nuclear cataract classification based on AS-OCT images and a grading block is introduced to learn high-level feature representations based on the pointwise convolution method.

Automatic Cataract Diagnosis by Image-Based Interpretability

  • Jianqiang LiXi Xu Liyang Xie
  • Computer Science
    2018 IEEE International Conference on Systems, Man, and Cybernetics (SMC)
  • 2018
A deep-learning algorithm is developed to intuitively identify cataract attributes to solve limitations of current methods, and achieves state of the art accuracy of proposed method on detection and grading task.

Automatic Classification of Cataract based on Deep Learnig

A novel deep learning based method is proposed to classify cataract using fundus images by resorts to deep transfer learning to reduce the number of parameters that need to be trained.

Fully Deep Learning for Slit-Lamp Photo Based Nuclear Cataract Grading

A fully deep learning based solution for automated nuclear cataract grading based on slit-lamp photos that outperforms the state-of-the-art, and a simple batch balancing strategy is introduced for improving the training of the grading network.

Unified Diagnosis Framework for Automated Nuclear Cataract Grading Based on Smartphone Slit-Lamp Images

The proposed validation method could evaluate a cataract severity in 29 ms and the entire classification process in less than 1s and can improve the accuracy of the examination, reduce misdiagnosis rate and the difficulty of the doctor’s examination.

Employing Deep Learning Architectures for Image-Based Automatic Cataract Diagnosis

A diagnosis system based on color fundus images is addressed for cataract disease, which achieves 97.94% diagnosis rate and dysfunction associated with cataracts could be identified in the early stage.

Machine Learning for Cataract Classification and Grading on Ophthalmic Imaging Modalities: A Survey

This survey provides a comprehensive survey of recent advances in machine learning techniques for cataract classification/grading based on ophthalmic images and summarizes existing literature from two research directions: conventional machine learning methods and deep learning methods.



A Computer Assisted Method for Nuclear Cataract Grading From Slit-Lamp Images Using Ranking

A novel computer-aided diagnosis method via ranking is proposed to facilitate nuclear cataract grading following conventional clinical decision-making process and demonstrates the benefit of grading via ranking by the proposed method.

A Computer-Aided Diagnosis System of Nuclear Cataract

This is the first time that the nucleus region can be detected automatically in slit lamp images and can improve the grading objectivity and potentially be used in clinics and population studies to save the workload of ophthalmologists.

Automatic Grading of Cortical and PSC Cataracts Using Retroillumination Lens Images

An improved cortical cataracts detection system that employs different strategies to address the challenges in cataract detection for lenses with different levels of estimated opacity that simultaneously overcome the over-detection issue for clear lenses and the under-detected issue for lens with high opacity.

Representation Learning: A Unified Deep Learning Framework for Automatic Prostate MR Segmentation

Experimental results show that significant segmentation accuracy improvement can be achieved by the proposed deep learning method compared to other state-of-the-art segmentation approaches.

The Lens Opacities Classification System III

The LOCS III is an improved LOCS system for grading slit-lamp and retroillumination images of age-related cataract and contains an expanded set of standards that were selected from the Longitudinal Study of Cataract slide library.

Deep Feature Learning for Knee Cartilage Segmentation Using a Triplanar Convolutional Neural Network

A novel system for voxel classification integrating three 2D CNNs, which have a one-to-one association with the xy, yz and zx planes of 3D image, respectively, which performs better than a state-of-the-art method using 3D multi-scale features.

Unsupervised Deep Feature Learning for Deformable Registration of MR Brain Images

An unsupervised deep learning approach is proposed to directly learn the basis filters that can effectively represent all observed image patches so that the coefficients by these learnt basis filters in representing the particular image patch can be regarded as the morphological signature for correspondence detection during image registration.

Convolutional-Recursive Deep Learning for 3D Object Classification

This work introduces a model based on a combination of convolutional and recursive neural networks (CNN and RNN) for learning features and classifying RGB-D images, which obtains state of the art performance on a standardRGB-D object dataset while being more accurate and faster during training and testing than comparable architectures such as two-layer CNNs.

Cataract conversion assessment using lens opacity classification system III and Wisconsin cataract grading system.

An approximate conversion algorithm for any two cataract grading systems was proposed and applied to the LOCS III and Wisconsin system and provides general ways to pool and compare cataracts prevalence using different grading systems in epidemiological studies.