文獻探討_Main_new

###### tags: `陳` # 文獻探討_Main_new ## 格式 - 標題 - 網址 - 年份、期刊 - 作者 - 可用文句 - ==上傳人簡寫== --- ## ==陳== ### V 1. [Localization and diagnosis framework for pediatric cataracts based on slit-lamp images using deep features of a convolutional neural network]( https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0168606#pone.0168606.ref012) - 2017 PLOSONE - Xiyang Liu ,Jiewei Jiang ,Kai Zhang,Erping Long,Jiangtao Cui,Mingmin Zhu,Yingying An,Jia Zhang,Zhenzhen Liu,Zhuoling Lin,Xiaoyan Li,Jingjing Chen,Qianzhong Cao,Jing Li,Xiaohang Wu,Dongni Wang,Haotian Lin ### V 2. [A computer-aided healthcare system for cataract classification and grading based on fundus image analysis](https://www.sciencedirect.com/science/article/pii/S0166361514001754?casa_token=6fkLMq0LLP8AAAAA:LpzBZmDRxvExii2Na6jpdE88Y_2y6PYPiaUjwZjpDJ_pwX06EPni0RkTnlMsAmvNV2DGg644khE) - 2015 - LiyeGuo, Ji-JiangYang, LihuiPeng, JianqiangLi, QingfengLiang ### 3. [Automatic Cataract Classification System Using Neural Network Algorithm Backpropagation](https://ieeexplore-ieee-org.autorpa.lib.ncnu.edu.tw/document/9277441) - 2020 - R. Munarto, M. Ali Setyo Yudono and E. Permata ### 4. [Unified Diagnosis Framework for Automated Nuclear Cataract Grading Based on Smartphone Slit-Lamp Images](https://ieeexplore-ieee-org.autorpa.lib.ncnu.edu.tw/document/9201392) - 2020 - S. Hu et al. ### 5. [Detecting Cataract Using Smartphones](https://ieeexplore-ieee-org.autorpa.lib.ncnu.edu.tw/document/9409132) - 2021 - B. Askarian, P. Ho and J. W. Chong --- ## ==Frank== ### 1. [Cataract Detection and Classifcation Systems Using Computational Intelligence- A Survey](https://link.springer.com/article/10.1007/s11831-020-09440-2) - Year / Journal or Conference Name : 2020 / ACME - Author: Hans Morales‐Lopez, Israel Cruz‐Vega, Jose Rangel‐Magdaleno - Deep learning breaks the complicated mapping of an image into a series of nested simple mappings, each described by a different layer of the model. A series of hid- den layers extract increasingly abstract features from the image. The values of the hidden layers are not given in the data; instead, the model must determine which concepts are useful for explaining the relationships in the observed data. - The strength of CNN’s lies in their **weight sharing**, exploiting the intuition that **similar structures occur in different locations** in an image. When seeing an image as a vectorized image, weights can be shared in such a way that it results in a convolution operation. This **drastically reduces the number of parameters that need to be learned** and renders the network equivariant with respect to translations of the input. Convolutional layers are typically alternated with pooling layers where pixel values of neighborhoods are aggregated using some permutation invariant function which induces a certain amount of translation invariance. At the end of the convolutional stream of the network, **fully-connected layers are usually added to act as classification layers, where weights are no longer shared**. CNN’s are typically trained end-to-end in a completely supervised way. :::info 這篇的chapter3&4個別整理前處理&分類方法的文章，但主要都是以fundus, retinal這類，沒有我們那種的（我們這種方法好像比較少？），所以如果寫到這方面的話可以來這篇參考，他有整理各文章的內容。 ::: ### 2. [Automatic Detection of Eye Cataract using Deep Convolution Neural Networks (DCNNs)](https://ieeexplore.ieee.org/document/9231045/authors#authors) - Year / Journal or Conference Name : 2020 / IEEE - Author: Md. Rajib Hossain, Sadia Afroze, Nazmul Siddique, Mohammed Moshiul Hoque - An automatic cataract detection system using DCNN architecture is proposed in this work. The proposed system obtained an accuracy of **95.77%** on test sets. The proposed DCNN system **overcomes the traditional feature extraction limitations** and this system **has no requirement for image pre-processing**. - trained by **5718 retinal fundus images**. ### V 3. [Enhanced intelligence using collective data augmentation for CNN based cataract detection](https://drive.google.com/drive/u/1/folders/1NpTQ-oHE023X4LBrkDMwMyapvdbIen2d) - 這篇只能從期刊中找到，無法獨立，在檔案中的第148頁 - Year / Journal or Conference Name : 2019 / FC - Author: Azhar Imran, Jianqiang Li, Yan Pei, Fawaz Mahiuob Mokbal, Ji-Jiang Yang, and Qing Wang - [x] To deal with the issues of **unbalanced data**, this study proposed two data augmentation techniques, i.e. Gaussian Scale-Space Theory, and general data augmentation settings. - ==Data Augmentation== - Using the augmented dataset to train a convolutional network, the proposed method achieved a significant performance in all measurements. - original data 8030 - augmented data 9743 - trained by 7793 retinal images, testing by 1949 retinal images - The proposed method achieved an accuracy of 96.91%, and 93.79% for cataract detection on the augmented dataset, and original dataset respectively. :::info 這篇主要在講有用因資料不平衡，所以使用擴增的方式來處理，而其結果比較原始沒有擴增的資料有更好的結果。 ::: ### 4. [Analysis of different automatic cataract detection and classification methods](https://ieeexplore.ieee.org/document/7154796) - Year / Journal or Conference Name : 2015 / IACC - Author: Niya C P, Jayakumar T.V :::warning 這篇不好，他是在分析過往幾篇paper並比較結果，但比較的幾篇已經很舊且使用的那些方法我們之前有看過的都比這幾篇好。 ::: ### V 5. [Artificial Intelligence for Cataract Detection and Management](https://journals.lww.com/apjoo/fulltext/2020/04000/artificial_intelligence_for_cataract_detection_and.6.aspx) --- ## ==王== ### 1. [Automatic cataract detection and grading using Deep Convolutional Neural Network](https://ieeexplore.ieee.org/abstract/document/8000068) - Year : 2017 - Conference papaer , ICNSC - Author : Linglin Zhang, Jianqiang Li, i Zhang, He Han, Bo Liu, Jijiang Yang, Qing Wang ### 2. [Computer-aided diagnosis of cataract using deep transfer learning](https://www.sciencedirect.com/science/article/abs/pii/S1746809419301077) - Year : 2019 - Journal , BSPC - Author : Turimerla Pratap∗, Priyanka Kokil ### 3. [Web-Based Cataract Detection System Using Deep Convolutional Neural Network](https://ieeexplore.ieee.org/abstract/document/8949636) - Year : 2019 - Conference paper , NigeriaComputConf - Author : Musa Yusuf, Samuel Theophilous, Jadesola Adejoke, Annah B. Hassan ### 4. [Automatic Cataract Classification Using Deep Neural Network With Discrete State Transition](https://ieeexplore.ieee.org/abstract/document/8759939) - Year : 2020 - Journal , IEEE T-MI - Author : Yue Zhou, Guoqi Li, Huiqi Li ## ==洪== ### 1. [Cataract Detection Using Single Layer Perceptron Based on Smartphone](https://ieeexplore.ieee.org/abstract/document/8982445) - Year : 2019 - Conference : ICICoS - Author : Riyanto Sigit、Elvi Triyana、Mochammad Rochmad - 本篇重點著重在Preprocessing( Cropping Image、 GrayscaleMedian Filter)、Segmentation(Canny Edge Detection、Hough Circle Transform)和Feature Extraction(gray histogram )上，沒有提及CNN ### 2. [Cataract Detection and Grading Based on Combination of Deep Convolutional Neural Network and Random Forests](https://ieeexplore.ieee.org/document/8525852) - Year : 2018 - Conference : IC-NIDC - Author : Jing Ran、Kai Niu、Zhiqiang He、Hongyan Zhang、Hongxin Song - The method DCNN combined with RF is proposed in this paper for six-level cataract grading. (DCNN-RF) - DCNN extracts features from fundus images automatically. RF completes cataract grading on the feature datasets produced by DCNN. The grading is six-level which is more difficult than four-level one. ### 3. [Analysis and study of cataract detection techniques](https://ieeexplore.ieee.org/document/7955355) - Year : 2016 - Conference : ICGTSPICC - Author : Darshana Patil、Arvind Nair、Niranjan Bhat、Rohit Chavan、Dheeraj Jadhav ### 4. [Cataract Detection Using Convolutional Neural Network with VGG-19 Model](https://ieeexplore.ieee.org/abstract/document/8759939) - Year : 2021 - Conference : AIIoT - Author : Md. Sajjad Mahmud Khan、Mahiuddin Ahmed、Raseduz Zaman Rasel、Mohammad Monirujjaman Khan - An early detection and proper guided diagnosis could tremendously impact in minimizing the rates of cataracts as automated cataract systems would free up ophthalmologist's time. - [x] Because of its feature extraction ability, deep learning is highly used in medical image analysis. To be more specific Convolutional Neural Networks (CNN) are used for medical image analysis. Without the need for human interference, the CNN model extracts local features directly from the fundus images. - ==Advantage (CNN)== - To be more specific VGG-19 were used for this study which achieved an accuracy of 97.47% on the test dataset. - This result shows that presented method achieve high accuracy even on unfiltered and image quality unassessed fundus photographs without ophthalmologist's intervention. - Convolutional Neural Network has become a hot research area in the field of image recognition. CNN has a weight sharing network and this network reduces the complexity of neural networks. This weight sharing network is similar to biological neural networks. ### V 5. [Classification of Cataract Slit-lamp Image Based on Machine Learning](https://ieeexplore-ieee-org.autorpa.lib.ncnu.edu.tw/document/8549701) --- # Advantage of CNN ### [Localization and diagnosis framework for pediatric cataracts based on slit-lamp images using deep features of a convolutional neural network](https://pubmed.ncbi.nlm.nih.gov/28306716/) Recently, deep learning convolutional neural network (CNN) methods have gained consid- erable popularity since they offer superior performance in the field of image recognition tasks[18–22]. The CNN is an end-to-end learning model that avoids image pre-processing, requires no expert knowledge and extracts relevant high-level features directly from the raw image. The CNN architecture is inspired by the visual cortex of cats in Hubel’s and Wiesel’s early work[23]. ### [Diagnosis of Ophthalmic Diseases in Fundus Image Using various Machine Learning Techniques](https://ieeexplore.ieee.org/document/9488928) **table 1** - looks like CNN accuracy better than the others ### [Cataract Detection Using Convolutional Neural Network with VGG-19 Model](https://www.researchgate.net/publication/352624583_Cataract_Detection_Using_Convolutional_Neural_Network_with_VGG-19_Model) Deep learning (DL) is a subfield of machine learning. DL uses multiple layers to extract higher levels of features from raw input. Because of its feature extraction ability, deep learning is highly used in medical image analysis. To be more specific Convolutional Neural Networks (CNN) are used for medical image analysis. Without the need for human interference, the CNN model extracts local features directly from the fundus images. ### [Cataract Detection and Classification Systems Using Computational Intelligence: A Survey](https://link.springer.com/article/10.1007/s11831-020-09440-2) The strength of CNN’s lies in their weight sharing, exploiting the intuition that similar structures occur in different locations in an image. When seeing an image as a vectorized image, weights can be shared in such a way that it results in a convolution opera- tion. This drastically reduces the number of parameters that need to be learned and renders the network equivariant with respect to translations of the input. Convolutional layers are typically alternated with pool- ing layers where pixel values of neighborhoods are aggre- gated using some permutation invariant function which induces a certain amount of translation invariance. At the end of the convolutional stream of the network, **fully-connected layers are usually added to act as classifi- cation layers, where weights are no longer shared.** CNN’s are typically trained end-to-end in a completely supervised way. ### [A Hybrid Global-Local Representation CNN Model for Automatic Cataract Grading](https://pubmed.ncbi.nlm.nih.gov/31059460/) This paper uses convolutional neural networks (CNN) to learn useful features directly from input data, and deconvolution network method is employed to investigate how CNN characterizes cataract layer-by-layer. We found that 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. ### [A survey on deep learning in medical image analysis](https://www.sciencedirect.com/science/article/abs/pii/S1361841517301135) ![](https://i.imgur.com/ecaMKiv.png) ### [Detection and Classification of Lung Abnormalities by Use of Convolutional Neural Network (CNN) and Regions with CNN Features (R-CNN)](https://ieeexplore.ieee.org/document/8369798) Moreover, convolutional neural network (CNN) has brought about breakthrough in pattern recognition of images including medical images. In usual CAD algorithms, designing an image-feature extractor is important. However, this task is difficult. On the other hand, a CAD algorithm by use of CNN does not necessarily require the image-feature extractor.