Fast Convolutional Neural Network Training Using Selective Data Sampling: Application to Hemorrhage Detection in Color Fundus Images
Open accessThe DOI matches its Crossref record; 50 of the 51 references it cites were found in a registry or an index.
Abstract
Convolutional neural networks (CNNs) are deep learning network architectures that have pushed forward the state-of-the-art in a range of computer vision applications and are increasingly popular in medical image analysis. However, training of CNNs is time-consuming and challenging. In medical image analysis tasks, the majority of training examples are easy to classify and therefore contribute little to the CNN learning process. In this paper, we propose a method to improve and speed-up the CNN training for medical image analysis tasks by dynamically selecting misclassified negative samples during training. Training samples are heuristically sampled based on classification by the current status of the CNN. Weights are assigned to the training samples and informative samples are more likely to be included in the next CNN training iteration. We evaluated and compared our proposed method by training a CNN with (SeS) and without (NSeS) the selective sampling method. We focus on the detection of hemorrhages in color fundus images. A decreased training time from 170 epochs to 60 epochs with an increased performance-on par with two human experts-was achieved with areas under the receiver operating characteristics curve of 0.894 and 0.972 on two data sets. The SeS CNN statistically outperformed the NSeS CNN on an independent test set.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“Twenty-five studies17,28,29,35,37,39,40,44–61 compared algorithm performance against healthcare professionals.”
From Diagnostic accuracy of deep learning in medical imaging: a systematic review and meta-analysis (2021) · cites this paper as [44]
Citations
Cites
Very Deep Convolutional Networks for Large-Scale Image Recognition
Simonyan, K., & Zisserman, A. (2014) · arXiv · doi:10.48550/arxiv.1409.1556 · Cited by 114,148 · cited 2×
“Using multiple smaller sized layers with non-linear rectifications makes the CNN more discriminative and less parameters need to be optimized [30].”
BackgroundMethodImageNet classification with deep convolutional neural networks
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017) · Communications of the ACM · doi:10.1145/3065386 · Cited by 112,155 · cited 1×
“The CNN architecture used in this study consists of five convolutional layers followed by rectified linear units (ReLUs) [4] and spatial max-pooling.”
MethodLeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
Fully convolutional networks for semantic segmentation
Long, J., Shelhamer, E., & Darrell, T. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298965 · Cited by 38,834
(2014) · Cited by 33,419
Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
Ioffe, S., & Szegedy, C. (2015) · arXiv · doi:10.48550/arxiv.1502.03167 · Cited by 27,818 · cited 1×
“Additions such as the inclusion of drop-out [42] or batch normalization [43] could potentially further increase performance.”
BackgroundA decision-theoretic generalization of on-line learning and an application to boosting
Freund, Y., & Schapire, R. E. (1997) · Journal of Computer and System Sciences · doi:10.1006/jcss.1997.1504 · Cited by 20,466
Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification
He, K., Zhang, X., Ren, S., & Sun, J. (2015) · 2015 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.123 · Cited by 19,741 · cited 1×
“Pilot experiments using MSRA weight initialization [3] showed an equal number of epochs required to train the SeS with similar end-performance on the test sets.”
MethodDeep learning in neural networks: an overview
Schmidhuber, J. (2015) · Neural Networks · doi:10.1016/j.neunet.2014.09.003 · Cited by 18,570
Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
Ioffe, S. (2024) · arXiv (Cornell University) · doi:10.57702/o9raffed · Cited by 14,214
Cited by
Items 101–200 of 426
Latif, J., Xiao, C., Tu, S., Rehman, S. U., Imran, A., & Bilal, A. (2020) · IEEE Access · doi:10.1109/access.2020.3016782 · Cited by 93
“The retinal images are used for various retinal diseases such as glaucoma [135], diabetic retinopathy [136], [137], hemorrhage [138], hard-exudate [139], retinopathy of prematurity [140], and age-related macular degeneration detection [141] etc.”
BackgroundA Systematic Review on Diabetic Retinopathy Detection Using Deep Learning Techniques
Vij, R., & Arora, S. (2022) · Archives of Computational Methods in Engineering · doi:10.1007/s11831-022-09862-0 · Cited by 81
“[189] use a convolutional neural network to segment the hemorrhages from the retinal images.”
MethodDeep learning-based hemorrhage detection for diabetic retinopathy screening
Aziz, T., Charoenlarpnopparut, C., & Mahapakulchai, S. (2023) · Scientific Reports · doi:10.1038/s41598-023-28680-3 · Cited by 83
“Grinsven 20 proposed a technique for HEs detection to expedite the training process.”
MethodAutomatic Segmentation of Retinal Layer in OCT Images With Choroidal Neovascularization
Xiang, D., Tian, H., Yang, X., Shi, F., Zhu, W., Chen, H., & Chen, X. (2018) · IEEE Transactions on Image Processing · doi:10.1109/tip.2018.2860255 · Cited by 67
“[25] dynamically selected misclassified negative samples during training to speed-up deep learning network training in order to detect hemorrhages in color fundus images.”
Yang, Y., Shang, F., Wu, B., Yang, D., Wang, L., Xu, Y., Zhang, W., & Zhang, T. (2021) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2021.3062638 · Cited by 70
“[35] proposed a selective sampling method”
Dense pooling layers in fully convolutional network for skin lesion segmentation
Nasr-Esfahani, E., Rafiei, S., Jafari, M. H., Karimi, N., Wrobel, J. S., Samavi, S., & Soroushmehr, S. M. R. (2019) · Computerized Medical Imaging and Graphics · doi:10.1016/j.compmedimag.2019.101658 · Cited by 84
Explainable end-to-end deep learning for diabetic retinopathy detection across multiple datasets
Chetoui, M., & Akhloufi, M. A. (2020) · Journal of medical imaging · doi:10.1117/1.jmi.7.4.044503 · Cited by 63
“…datasets to detect the presence of DR (Ex, HM, and MA) using DR labeling as in the work of Chalakkal et al., 15 Wang et al., 28 and Grinsven et al. 30 The detection of diabetic macula edema is not considered in this work since the EyePACS used for training does not have grades for macular edema.”
MethodMa, H., & Pang, X. (2019) · IEEE Access · doi:10.1109/access.2019.2936945 · Cited by 67
“As an important part of deep learning algorithm, convolutional neural network algorithm has been applied in large data processing and image correlation processing more than traditional machine learning [26]–[28].”
MethodReview of Machine Learning Applications Using Retinal Fundus Images
Jeong, Y., Hong, Y.-J., & Han, J.-H. (2022) · Diagnostics · doi:10.3390/diagnostics12010134 · Cited by 74
“Specifically, CNN299 and CNN512 consist of four and six convolution layers with different dimensions of input layers, followed by two fully connected layers.”
MethodInfluentialA Novel Machine Learning Model Based on Exudate Localization to Detect Diabetic Macular Edema
Perdómo, O., Otálora, S., Rodríguez, F., Arévalo, J., & Gonzalez, F. A. (2016) · Ophthalmic Medical Image Analysis International Workshop (OMIA) · doi:10.17077/omia.1057 · Cited by 64
“Computer-aided systems (CADx) are an interesting alternative to tackel these problems CADx systems perform an automatic assesment of the disease grade, they may increase the number of patients diagnosed and may reduce the time to detect diabetic eye diseases [10].”
BackgroundErciyas, A., & Barışçı, N. (2021) · Computational and Mathematical Methods in Medicine · doi:10.1155/2021/9928899 · Cited by 71
“In [53], HM is detected, and a 41-pixel square image containing HM was extracted from the original image.”
MethodA Review on Progress in Semantic Image Segmentation and Its Application to Medical Images
Kar, M. K., Nath, M. K., & Neog, D. R. (2021) · SN Computer Science · doi:10.1007/s42979-021-00784-5 · Cited by 71
“From left to right (top row): mammographic mass classification [7], segmentation of lesions in the brain [8], leak detection in airway tree segmentation [9]; from left to right (middle row): diabetic retinopathy classification [10], prostate segmentation [11], nodule classification [12]; from left to right (bottom row…”
BackgroundMethodKooi, T., & Karssemeijer, N. (2017) · Journal of medical imaging · doi:10.1117/1.jmi.4.4.044501 · Cited by 67
“43 The vanilla CNN architecture is a generic problem solver for many signal processing tasks but is still limited by the constraint that a single tensor needs to be fed to the front-end layer, if no further adaptations to the network are made.”
BackgroundDerin Öğrenme Modelleri ve Uygulama Alanlarına İlişkin Bir Derleme
Doğan, F., & Türkoğlu, İ. (2019) · DÜMF Mühendislik Dergisi · doi:10.24012/dumf.411130 · Cited by 103
Comparison of different methods for tissue segmentation in histopathological whole-slide images
Bándi, P., van de Loo, R., Intezar, M., Geijs, D. J., Ciompi, F., van Ginneken, B., van der Laak, J. A. W. M., & Litjens, G. J. S. (2017) · IEEE International Symposium on Biomedical Imaging (ISBI) · doi:10.1109/isbi.2017.7950590 · Cited by 56
A survey of deep learning models in medical therapeutic areas
Nogales, A., García-Tejedor, Á. J., Martín, D. M., Vara, J. S., & Rodríguez, C. A. (2021) · Artificial Intelligence in Medicine · doi:10.1016/j.artmed.2021.102020 · Cited by 63
“Ophthalmology is another therapeutic area where deep learning has developed many models with ocular fundus images to detect retinal diseases like age-related macular degeneration [48, 63], hemorrhages [64], microaneurysms [49], diabetic retinopathy [59, 77, 101] and rhegmatogenous retinal detachment [97] and glaucoma…”
BackgroundDeep Convolution Features in Non-linear Embedding Space for Fundus Image Classification
Dondeti, V., Bodapati, J. D., Shaik, N. S., & Veeranjaneyulu, N. (2020) · Revue d intelligence artificielle · doi:10.18280/ria.340308 · Cited by 71
“Van Grinsven et al. [8] suggest another deep learning approach to hemorrhage detection.”
BackgroundDeep convolutional neural network in medical image processing
Mohapatra, S., Swarnkar, T., & Das, J. (2020) · Elsevier eBooks · doi:10.1016/b978-0-12-823014-5.00006-5 · Cited by 62
CoT-XNet: contextual transformer with Xception network for diabetic retinopathy grading
Zhao, S., Wu, Y., Tong, M., Yao, Y., Qian, W., & Qi, S. (2022) · Physics in Medicine and Biology · doi:10.1088/1361-6560/ac9fa0 · Cited by 54
Shahid, A. H., & Singh, M. P. (2019) · Journal of Applied Biomedicine · doi:10.1016/j.bbe.2019.05.010 · Cited by 53
Intelligent Analysis of Medical Big Data Based on Deep Learning
Sun, H., Liu, Z., Wang, G., Lian, W., & Ma, J. (2019) · IEEE Access · doi:10.1109/access.2019.2942937 · Cited by 53
“[33] tried to dynamically select the wrong false negative samples during the training process.”
BackgroundChalakkal, R. J., Abdulla, W. H., & Hong, S. C. (2020) · Elsevier eBooks · doi:10.1016/b978-0-12-817440-1.00003-6 · Cited by 56
Wong, N., Meshkinfamfard, S., Turbé, V., Whitaker, M. D., Moshe, M., Bardanzellu, A., Dai, T., Pignatelli, E., Barclay, W., Darzi, A. W., et al. (2022) · Communications Medicine · doi:10.1038/s43856-022-00146-z · Cited by 62
A Systematic Literature Review on Diabetic Retinopathy Using an Artificial Intelligence Approach
Bidwai, P., Gite, S., Pahuja, K., & Kotecha, K. V. (2022) · Big Data and Cognitive Computing · doi:10.3390/bdcc6040152 · Cited by 64
Cunefare, D., Langlo, C. S., Patterson, E. J., Blau, S., Dubra, A., Carroll, J., & Farsiu, S. (2018) · Biomedical Optics Express · doi:10.1364/boe.9.003740 · Cited by 49
Retina blood vessel segmentation using a U-net based Convolutional neural network
Xiancheng, W., Wei, L., Bingyi, M., He, J., Jiang, Z., Xu, W., Ji, Z., Hong, G., & Zhaomeng, S. (2018) · Swinburne figshare (Swinburne University of Technology) · doi:10.25916/sut.26248415 · Cited by 70
Diabetic retinopathy detection and classification using hybrid feature set
Amin, J., Sharif, M., Rehman, A. U., Raza, M., & Mufti, M. R. (2018) · Microscopy Research and Technique · doi:10.1002/jemt.23063 · Cited by 51
Deep learning in ultrasound elastography imaging: A review
Li, H., Bhatt, M., Qu, Z., Zhang, S., Hartel, M. C., Khademhosseini, A., & Cloutier, G. (2022) · Medical Physics · doi:10.1002/mp.15856 · Cited by 47
Fast Deep Learning Training through Intelligently Freezing Layers
Xiao, X., Mudiyanselage, T. B., Ji, C., Hu, J., & Pan, Y. F. (2019) · IEEE International Conference on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData) · doi:10.1109/ithings/greencom/cpscom/smartdata.2019.00205 · Cited by 59
“[27] decreased”
Deep Learning for Histopathological Image Analysis: Towards Computerized Diagnosis on Cancers
Xu, J., Zhou, C., Lang, B., & Liu, Q. (2017) · Advances in computer vision and pattern recognition · doi:10.1007/978-3-319-42999-1_6 · Cited by 56
Playout, C., Duval, R., & Cheriet, F. (2018) · Lecture notes in computer science · doi:10.1007/978-3-030-00934-2_12 · Cited by 51
Ikram, A., Imran, A., Li, J., Alzubaidi, A. A., Fahim, S., Yasin, A., & Fathi, H. (2024) · IEEE Access · doi:10.1109/access.2024.3427394 · Cited by 51
“For the retinal lesion, Grinsven et al. [33] and Lam et al. [39] have used the Kaggle dataset.”
BackgroundMethodInfluentialCunefare, D., Huckenpahler, A. L., Patterson, E. J., Dubra, A., Carroll, J., & Farsiu, S. (2019) · Biomedical Optics Express · doi:10.1364/boe.10.003815 · Cited by 47
DeepBranch: Deep Neural Networks for Branch Point Detection in Biomedical Images
Tan, Y., Liu, M., Chen, W., Wang, X., Peng, H., & Wang, Y. (2019) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2019.2945980 · Cited by 46
“Actually, CNNs have achieved great success with hierarchical feature representations in 2D medical images, such as hemorrhage detection in blood vessel images [28], retinal vessel segmentation [29] and lung disease detection [30].”
BackgroundDeep learning for diabetic retinopathy assessments: a literature review
Skouta, A., Elmoufidi, A., Andaloussi, S. J., & Ouchetto, O. (2023) · Multimedia Tools and Applications · doi:10.1007/s11042-023-15110-9 · Cited by 52
Classification and Segmentation of Diabetic Retinopathy: A Systemic Review
Shaukat, N., Amin, J., Sharif, M. I., Sharif, M. I., Kadry, S., & Sevcik, L. (2023) · Applied Sciences · doi:10.3390/app13053108 · Cited by 54
Review of Deep Learning Methods in Mammography, Cardiovascular, and Microscopy Image Analysis
Carneiro, G., Zheng, Y., Xing, F., & Yang, L. (2017) · Advances in computer vision and pattern recognition · doi:10.1007/978-3-319-42999-1_2 · Cited by 51
BHCNet: Neural Network-Based Brain Hemorrhage Classification Using Head CT Scan
Mushtaq, M. F., Shahroz, M., Aseere, A. M., Shah, H., Majeed, R., Shehzad, D., & Samad, A. (2021) · IEEE Access · doi:10.1109/access.2021.3102740 · Cited by 46
“There are two sets of layers implemented consists of convolutional 2D and Max pooling 2D layer [53], [67] as shown in Figure 5 which extracts the features and down-samples the features map to 4 × 4 × 64.”
MethodAlgorithms for red lesion detection in Diabetic Retinopathy: A review
Biyani, R., & Patre, B. M. (2018) · Biomedicine & Pharmacotherapy · doi:10.1016/j.biopha.2018.07.175 · Cited by 42
Weakly Supervised Lesion Detection From Fundus Images
Wang, R., Chen, B., Meng, D., & Wang, L. (2018) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2018.2885376 · Cited by 43
“Compared to two supervised methods SeS CNN [27] and U-net [28], the ROC and PR curves of our method are both above the former’s, and the ROC curve is above and the PR curve is partially above corresponding curves of U-net [28].”
MethodResultInfluentialWang, J., Horlacher, M., Cheng, L., & Winther, O. (2023) · Briefings in Bioinformatics · doi:10.1093/bib/bbad249 · Cited by 49
Li, P., Li, Z., Gao, F., Wan, L., & Yu, J. (2017) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme.2017.8019415 · Cited by 44
“[10] utilize a 9-layer CNN for bleeding detection in fundus images, and speed up the CNN training by dynamically selecting misclassified negative samples during training.”
BackgroundMethodMassive datasets and machine learning for computational biomedicine: trends and challenges
Андреевич, К. А., Pardalos, P. M., & Karakitsiou, A. (2018) · Annals of Operations Research · doi:10.1007/s10479-018-2891-2 · Cited by 58
“Another approach is to dynamically utilize misclassified samples as in (van Grinsven et al. 2016) that was reported to have both good speed-up and performance.”
BackgroundMulti-scale Microaneurysms Segmentation Using Embedding Triplet Loss
Sarhan, M. H., Albarqouni, S., Yiğitsoy, M., Navab, N., & Eslami, A. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-32239-7_20 · Cited by 41
Jekyll: Attacking Medical Image Diagnostics using Deep Generative Models
Mangaokar, N., Pu, J., Bhattacharya, P., Reddy, C. K., & Viswanath, B. (2020) · IEEE European Symposium on Security and Privacy · doi:10.1109/eurosp48549.2020.00017 · Cited by 57
“Recent examples also include ML schemes that perform well on breast cancer detection [32], skin cancer classification [33], arrhythmia detection [34], hemorrhage identification [35], and diabetic retinopathy detection [36].”
BackgroundCao, P., Hou, Q., Song, R., Wang, H., & Zai͏̈ane, O. R. (2022) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2022.105341 · Cited by 45
Hu, K., Shen, B., Zhang, Y., Cao, C., Xiao, F., & Gao, X. (2019) · Neurocomputing · doi:10.1016/j.neucom.2019.07.079 · Cited by 40
Sharma, Y., & Singh, B. K. (2021) · Computer Methods and Programs in Biomedicine · doi:10.1016/j.cmpb.2021.106487 · Cited by 46
Wu, J., Gou, F., & Tan, Y. (2021) · Computational and Mathematical Methods in Medicine · doi:10.1155/2021/6654946 · Cited by 38
“With the development of the Internet of Things and 5G technology, machine-assisted systems have played an increasingly important role in the mining, classification, and decision-making of medical information [28].”
BackgroundDiabetic retinopathy lesion segmentation using deep multi-scale framework
Guo, T., Yang, J., & Yu, Q. (2023) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2023.105050 · Cited by 38
Mamani, G. E. H., Setio, A. A. A., van Ginneken, B., & Jacobs, C. (2018) · Physics in Medicine and Biology · doi:10.1088/1361-6560/aab4b3 · Cited by 46
“Furthermore, we used selective sampling (van Grinsven et al 2016).”
Towards pixel-to-pixel deep nucleus detection in microscopy images
Xing, F., Xie, Y., Shi, X., Chen, P., Zhang, Z., & Yang, L. (2019) · BMC Bioinformatics · doi:10.1186/s12859-019-3037-5 · Cited by 41
Fang, F., Liu, Y.-M., Park, J. H., & Liu, Y.-J. (2023) · IEEE transactions on fuzzy systems · doi:10.1109/tfuzz.2023.3239732 · Cited by 30
“INTRODUCTION IN RECENT years, as a nonlinear system with rich dynamic behavior, neural networks (NNs) have much attention due to their benefits in text recognition, security assessment, image detection, associative memory, and other fields [1], [2], [3], [4], [5].”
BackgroundAutomated Hemorrhage Detection from Coarsely Annotated Fundus Images in Diabetic Retinopathy
Huang, Y., Lin, L., Li, M., Wu, J., Cheng, P., Wang, K., Yuan, J., & Tang, X. (2020) · IEEE International Symposium on Biomedical Imaging (ISBI) · doi:10.1109/isbi45749.2020.9098319 · Cited by 42
“For example, Grinsven et al. presented a CNN architecture for detecting hemorrhages with nine layers trained by selective samples [7].”
BackgroundIMPROVED PREDICTIVE SPARSE DECOMPOSITION METHOD WITH DENSENET FOR PREDICTION OF LUNG CANCER
Mienye, I. D., Sun, Y., & Wang, Z. (2020) · International Journal of Computing · doi:10.47839/ijc.19.4.1986 · Cited by 40
Dar, J. A., Srivastava, K. K., & Lone, S. A. (2022) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2022.103905 · Cited by 41
Automatic detection of the foveal center in optical coherence tomography
Liefers, B., Venhuizen, F. G., Schreur, V., van Ginneken, B., Hoyng, C., Fauser, S., Theelen, T., & Sánchez, C. I. (2017) · Biomedical Optics Express · doi:10.1364/boe.8.005160 · Cited by 36
Self-citationDiabetic retinopathy detection using supervised and unsupervised deep learning: a review study
Naz, H. U., Ahuja, N. J., & Nijhawan, R. (2024) · Artificial Intelligence Review · doi:10.1007/s10462-024-10770-x · Cited by 45
Reddy, V. P. C., & Gurrala, K. K. (2021) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2021.103439 · Cited by 35
Lung anomaly detection from respiratory sound database (sound signals)
Dar, J. A., Srivastava, K. K., & Mishra, A. (2023) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2023.107311 · Cited by 38
Zhao, C., Han, J., Jia, Y., Fan, L., & Gou, F. (2018) · Journal of Electrical and Computer Engineering · doi:10.1155/2018/2187247 · Cited by 34
Sağlık Alanında Kullanılan Derin Öğrenme Yöntemleri
Kaya, U., Yılmaz, A., & Dikmen, Y. (2019) · European Journal of Science and Technology · doi:10.31590/ejosat.573248 · Cited by 44
“SES ESA, bağımsız bir test setinde NSES ESA ‘den istatistiksel olarak daha iyi performans gösterdiğini ortaya koymuşlardır [72].”
Optimizing Data Augmentation for Semantic Segmentation on Small-Scale Dataset
Ma, R., Tao, P., & Tang, H. (2019) · International Conference on Control and Computer Vision (ICCCV) · doi:10.1145/3341016.3341020 · Cited by 36
“Random cropping [17], [18] is one of the most commonly used data augmentation methods.”
MethodBrain Tumor Prediction by Binary Classification Using VGG‐16
Singh, V., Sharma, S., Goel, S., Lamba, S., & Garg, N. (2021) · Wiley eBooks · doi:10.1002/9781119752134.ch9 · Cited by 34
Deep Learning in Retinal Image Segmentation and Feature Extraction: A Review
Hoque, M. E., & Kipli, K. (2021) · International Journal of Online and Biomedical Engineering (iJOE) · doi:10.3991/ijoe.v17i14.24819 · Cited by 37
“This newly developed model had AUC 0.991 and 0.990 for both EyePACS-1 and Messidor-2.”
MethodInfluentialMedical Image Analysis Using Deep Learning: A Systematic Literature Review
Kumar, E. S., & Bindu, C. S. (2019) · Communications in computer and information science · doi:10.1007/978-981-13-8300-7_8 · Cited by 37
Recent Advances in Deep Learning Techniques and Its Applications: An Overview
Hazra, A., Choudhary, P., & Singh, M. S. (2020) · Lecture notes in bioengineering · doi:10.1007/978-981-15-6329-4_10 · Cited by 35
Shaaf, Z. F., Jamil, M. M. A., Ambar, R., Alattab, A. A., Yahya, A. A., & Asiri, Y. (2022) · Diagnostics · doi:10.3390/diagnostics12020414 · Cited by 37
“Oversampling [46] and data augmentation [47] are data-level methods that work with training datasets, whereas classifier-level methods such as cost-sensitive learning [48], hard mining [49], and loss function work with model training options.”
MethodA review on optimization techniques for medical image analysis
Kaur, P., & Singh, R. K. (2022) · Concurrency and Computation Practice and Experience · doi:10.1002/cpe.7443 · Cited by 33
Didier, A. J., Nigro, A., Noori, Z., Omballi, M., Pappada, S. M., & Hamouda, D. (2024) · Frontiers in Artificial Intelligence · doi:10.3389/frai.2024.1365777 · Cited by 32
Hierarchical support vector machine for facial micro-expression recognition
Pan, H., Xie, L., Lv, Z., Li, J., & Wang, Z. (2020) · Multimedia Tools and Applications · doi:10.1007/s11042-020-09475-4 · Cited by 28
“Because the distribution of category imbalances in data-driven learning algorithms will tend to be more unfavorable for recognition results [6, 8, 10].”
BackgroundKundu, S., Karale, V., Ghorai, G. K., Sarkar, G., Ghosh, S., & Dhara, A. K. (2022) · Journal of Imaging Informatics in Medicine · doi:10.1007/s10278-022-00629-4 · Cited by 28
“Fully convolutional networks (FCN) were introduced for pixel-level segmentation [21] and were applied for medical image segmentation [6, 22, 23].”
MethodKhojasteh, P., Aliahmad, B., Arjunan, S. P., & Kumar, D. (2018) · Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) · doi:10.1109/embc.2018.8513606 · Cited by 27
“By applying the CE method on the original image , ICE is obtained as follows [7]:”
BackgroundMethodInfluentialDecision tree and deep learning based probabilistic model for character recognition
Sampath, A., & Gomathi, N. (2017) · Journal of Central South University · doi:10.1007/s11771-017-3701-8 · Cited by 27
“The classifiers that can be used for the recognition are support vector machines (SVM) [8], neural network (NN) [9] classifier, etc.”
Grading of Knee Osteoarthritis Using Convolutional Neural Networks
Sarvamangala, D. R., & Kulkarni, R. V. (2021) · Neural Processing Letters · doi:10.1007/s11063-021-10529-3 · Cited by 29
“CNNs have been applied extensively in medical imaging applications including brain segmentation [13,27], lung disease classification [2], colon cancer detection [30], breast cancer diagnosis [33,38] and hemorrhage detection in color fundus images [12].”
MethodRecent developments on computer aided systems for diagnosis of diabetic retinopathy: a review
Dubey, S., & Dixit, M. (2022) · Multimedia Tools and Applications · doi:10.1007/s11042-022-13841-9 · Cited by 30
Fine-grained attention & knowledge-based collaborative network for diabetic retinopathy grading
Tian, M., Wang, H., Sun, Y., Wu, S., Tang, Q., & Zhang, M. (2023) · Heliyon · doi:10.1016/j.heliyon.2023.e17217 · Cited by 28
Estimating Biological Gender from Panoramic Dental X-Ray Images
Milošević, D., Vodanović, M., Galić, I., & Subašić, M. (2019) · International symposium on image and signal processing and analysis/ISPA ... · doi:10.1109/ispa.2019.8868804 · Cited by 26
“For example, in [5] authors use multi-scale convolutional neural networks to classify lung nodules on thoracic Computed Tomography screening, in [6] authors use convolutional neural networks for hemorrhage detection in color fundus images, and in [7] authos use 3D convolutional neural networks to find sexual dimorphis…”
BackgroundMachine learning in nephrology: scratching the surface
Li, Q., Fan, Q., Han, Q., Geng, W., Zhao, H., Ding, X., Yan, J.-Y., & Zhu, H. (2020) · Chinese Medical Journal · doi:10.1097/cm9.0000000000000694 · Cited by 27
Dual-Branch U-Net Architecture for Retinal Lesions Segmentation on Fundus Image
Yin, M., Soomro, T. A., Jandan, F. A., Fatihi, A., Ubaid, F. B., Irfan, M., Afifi, A. J., Rahman, S., Telenyk, S., & Nowakowski, G. (2023) · IEEE Access · doi:10.1109/access.2023.3333364 · Cited by 25
“Theelen et al. [23] present a method aimed at improving CNN training for medical image analysis, with a focus on hemorrhage detection in colored fundus images.”
MethodPrathibha, S., & Siddappaji (2024) · Global Transitions · doi:10.1016/j.glt.2024.04.001 · Cited by 29
Yang, D., Jia, X., Xiao, Y., Wang, X., Wang, Z., & Yang, Z. (2019) · BioMed Research International · doi:10.1155/2019/9783106 · Cited by 25
Deep Learning Based Analysis of Ophthalmology: A Systematic Review
Kumar, P., Kumar, R., & Gupta, M. (2021) · EAI Endorsed Transactions on Pervasive Health and Technology · doi:10.4108/eai.10-9-2021.170950 · Cited by 34
“[61] used this concept to detect hemorrhages from color fundus images.”
MethodReview on diabetic retinopathy with deep learning methods.
Shekar, S. C., Satpute, N., & Gupta, A. (2021) · Journal of medical imaging · doi:10.1117/1.jmi.8.6.060901 · Cited by 14
Vahadane, A., Joshi, A., Madan, K., & Dastidar, T. R. (2018) · IEEE International Symposium on Biomedical Imaging (ISBI) · doi:10.1109/isbi.2018.8363840 · Cited by 23
“Deep learning techniques have been shown to outperform image processing and machine learning methods for natural image object as well as biomedical object classification [10].”
BackgroundTesting Segmentation Popular Loss and Variations in Three Multiclass Medical Imaging Problems
Furtado, P. (2021) · Journal of Imaging · doi:10.3390/jimaging7020016 · Cited by 30
Wang, C., Calle, P., Ton, N. B. T., Zhang, Z., Yan, F., Donaldson, A. M., Bradley, N. A., Yu, Z., Fung, K.-M. A., Pan, C., et al. (2021) · Biomedical Optics Express · doi:10.1364/boe.421299 · Cited by 25
de Simone, B. D., Deeken, G., & Catena, F. (2025) · Journal of Clinical Medicine · doi:10.3390/jcm14093111 · Cited by 28
Mitral Valve Leaflet Tracking in Echocardiography using Custom Yolo3
Chandra, V., Sarkar, P. G., & Singh, V. K. (2020) · Procedia Computer Science · doi:10.1016/j.procs.2020.04.089 · Cited by 21
Wang, C., Jin, Y., Chen, X., & Liu, Z. (2020) · Sensing and Imaging · doi:10.1007/s11220-020-00299-y · Cited by 22
Application of Neural Networks in the Medical Field
Mansouri, S. (2023) · Journal of Wireless Mobile Networks Ubiquitous Computing and Dependable Applications · doi:10.58346/jowua.2023.i1.006 · Cited by 45
Shareef, O., Soleimani, M., Tu, E. Y., Jacobs, D. S., Ciolino, J. B., Rahdar, A., Cheraqpour, K., Ashraf, M., Habib, N. B., Greenfield, J. A., et al. (2024) · The Ocular Surface · doi:10.1016/j.jtos.2024.07.010 · Cited by 27
Retinal Image Synthesis for CAD Development
Pujitha, A. K., & Sivaswamy, J. (2018) · Lecture notes in computer science · doi:10.1007/978-3-319-93000-8_70 · Cited by 21
“method [5] for binary (HE/no HE) classification used local markings by experts and report a SN/SP on the MESSIDOR dataset of 91.”
BackgroundMethodAplicaciones de las redes neuronales y el deep learning a la ingeniería biomédica
Sarmiento-Ramos, J. L. (2020) · Revista UIS Ingenierías · doi:10.18273/revuin.v19n4-2020001 · Cited by 45
Diabetic Retinopathy Grading Using ResNet Convolutional Neural Network
Sallam, M. S., Asnawi, A. L., & Olanrewaju, R. F. (2020) · IEEE Conference on Big Data and Analytics (ICBDA) · doi:10.1109/icbda50157.2020.9289822 · Cited by 20
“[12] has advocated the same direction of the previous research, but they have formulated the problem as patch classification problem instead of pixel classification problem.”
MethodLocal Structure Awareness-Based Retinal Microaneurysm Detection with Multi-Feature Combination
Deng, J., Tang, P., Zhao, X., Pu, T., Qu, C., & Peng, Z. (2022) · Biomedicines · doi:10.3390/biomedicines10010124 · Cited by 24
“These parameters were set following Grinsven et al. [19], i.e., α = 4, τ = − 4, γ = 128 and σ = χ /30.”
MethodDeepHealth: Review and challenges of artificial intelligence in health informaticsPreprint
Kwak, G. H., & Hui, P. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1909.00384 · Cited by 51
“To deal with this data complexity, traditional machine learning or deep learning approaches using hand-designed feature extraction were used [11, 80, 186, 219, 223, 261, 285].”
MethodAlDahoul, N., Karim, H. B. A., Tan, M. J. T., Momo, M. A., & Fermin, J. L. (2021) · F1000Research · doi:10.12688/f1000research.73082.1 · Cited by 20
“These methods involve CNNs developed from scratch to grade the disease using images of the retinal fundus(22); transfer learning based on Inception-v3 neural network to perform multiple binary classification (moderate versusworse DR, and severe or worse DR)(23); segmentation prior to detection by pixel classification(…”
MethodUrsuleanu, T. F., Luca, A. R., Gheorghe, L., Grigorovici, R., Iancu, S., Hlusneac, M., Preda, C., & Grigorovici, A. (2021) · Diagnostics · doi:10.3390/diagnostics11081373 · Cited by 26
Gudhe, N. R., Behravan, H., Sudah, M., Okuma, H., Vanninen, R. L., Kosma, V., & Mannermaa, A. (2022) · Scientific Reports · doi:10.1038/s41598-022-16141-2 · Cited by 20
Metadata
- Authors
- van Grinsven, M. J. J. P., van Ginneken, B., Hoyng, C. B., Theelen, T., & Sanchez, C. I.
- Affiliations
- Radboud University Medical Center (NL)
- Venue
- IEEE Transactions on Medical Imaging
- Published
- 2016-05
- Volume
- 35
- Issue
- 5
- Pages
- 1273-1284
- ISSN
- 0278-0062 (ISSN-L)
- 1558-254X
- OpenAlex
- W2344912502
- Cited by
- 426
- References
- 51
- Citations by year
- 2026: 7 · 2025: 13 · 2024: 27 · 2023: 41 · 2022: 49 · 2021: 70 · 2020: 57 · 2019: 68 · 2018: 50 · 2017: 38 · 2016: 6
- FWCI
- 46.88
- Type
- journal article
- Language
- en
- Open access
- green
- hdl.handle.net
- Locations
- IEEE Transactions on Medical Imaging (publishedVersion)
- PubMed (publishedVersion)
- Radboud Repository (Radboud University) (submittedVersion)
- Radboud Repository (Radboud University) (submittedVersion, other-oa)
- Topics
- Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Retinal Imaging and Analysis
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning and Data Classification
- Keywords
- convolutional neural network training, training efficiency, sample weighting, hemorrhage detection, color fundus images, medical image analysis, training time reduction
- MeSH terms
- Machine Learning, Fundus Oculi, Humans, Image Interpretation, Computer-Assisted, Retinal Hemorrhage, Databases, Factual, Neural Networks, Computer
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.