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 201–300 of 426
Radio Frequency Energy Harvesting-Based Self-Powered Dairy Cow Behavior Classification System
Dang, N.-H., Tran, V. T., Dang, T.-H., & Chung, W. (2023) · IEEE Sensors Journal · doi:10.1109/jsen.2023.3253176 · Cited by 18
“an approach for image processing [57], [58], [59] but has also been shown to be suitable for the analysis of time-varying”
MethodPreity, Bhandari, A. K., & Shahnawazuddin, S. (2023) · Archives of Computational Methods in Engineering · doi:10.1007/s11831-023-09998-7 · Cited by 19
“It is perceptible as red dots in the eyes [108].”
Transfer learning for diabetic retinopathy
Benson, J., Maynard, J., Zamora, G., Carrillo, H., Wigdahl, J., Nemeth, S. C., Barriga, S., Estrada, T., & Solíz, P. (2018) · SPIE Medical Imaging: Image Processing · doi:10.1117/12.2293378 · Cited by 19
“and van Grinsven et al.(8) used CNNs to detect specific lesions in patches of digital retinal fundus images.”
MethodUmapathy, L., Winegar, B. A., MacKinnon, L., Hill, M., Altbach, M. I., Miller, J., & Bilgin, A. (2020) · American Journal of Neuroradiology · doi:10.3174/ajnr.a6538 · Cited by 18
Retinopathy grading with deep learning and wavelet hyper-analytic activations
Chandrasekaran, R., & Balaji, L. (2022) · The Visual Computer · doi:10.1007/s00371-022-02489-z · Cited by 20
“Hemorrhages and MAs are the most commonly detected and associated together with a single label [45].”
BackgroundArtificial Intelligence: A Century-Old Story
van Assen, M., Muscogiuri, E., Tessarin, G., & Cecco, C. N. D. (2022) · Contemporary medical imaging · doi:10.1007/978-3-030-92087-6_1 · Cited by 23
Classification for diabetic retinopathy by using staged convolutional neural network
Wang, H., Sun, Y., Cao, Y., Ouyang, G., Wang, X., Wu, S., & Tian, M. (2022) · Asia Conference on Algorithms, Computing and Machine Learning (CACML) · doi:10.1109/cacml55074.2022.00045 · Cited by 17
“For removing the irrelevant information from the fundus images such as various illumination and resolution, a pre-processing method is applied to unify the image quality and improve the image contrast [23][24].”
MethodArtificial intelligence in retinal imaging: current status and future prospects
Heger, K. A., & Waldstein, S. M. (2023) · Expert Review of Medical Devices · doi:10.1080/17434440.2023.2294364 · Cited by 24
“The more advanced proliferative DR is characterized by neo-vascularization and may result in vitreous hemorrhage or trac-tional retinal detachment [7,62–64].”
BackgroundInfluentialExudate Detection in Fundus Images via Convolutional Neural Network
Guo, L., Zheng, S., & Li, X. (2018) · Communications in computer and information science · doi:10.1007/978-981-10-8108-8_18 · Cited by 18
Deep learning model for thorax diseases detection
Shadeed, G. A., Tawfeeq, M. A., & Mahmoud, S. M. (2020) · TELKOMNIKA (Telecommunication Computing Electronics and Control) · doi:10.12928/telkomnika.v18i1.12997 · Cited by 20
Efficient approach for the automatic detection of haemorrhages in colour retinal images
Bhoopalan, R., & Sundaramoorthy, S. (2018) · IET Image Processing · doi:10.1049/iet-ipr.2017.1036 · Cited by 16
Deep Learning and the Future of Biomedical Image Analysis
Jyotiyana, M., & Kesswani, N. (2019) · Studies in big data · doi:10.1007/978-3-030-33966-1_15 · Cited by 18
Feng, L., Zhao, C., Chen, C. L. P., Li, Y., Zhou, M., Qiao, H., & Fu, C. (2020) · Neurocomputing · doi:10.1016/j.neucom.2020.06.100 · Cited by 16
Artificial neural networks as clinical decision support systems
Shafi, I., Ansari, S., Din, S., Jeon, G., & Paul, A. (2021) · Concurrency and Computation Practice and Experience · doi:10.1002/cpe.6342 · Cited by 21
Red-lesion extraction in retinal fundus images by directional intensity changes’ analysis
Monemian, M., & Rabbani, H. (2021) · Scientific Reports · doi:10.1038/s41598-021-97649-x · Cited by 16
Comparative Study of Transfer Learning Models for Retinal Disease Diagnosis from Fundus Images
Pin, K., Chang, J. H., & Nam, Y. (2021) · Computers, materials & continua/Computers, materials & continua (Print) · doi:10.32604/cmc.2022.021943 · Cited by 23
Attiku, Y., He, Y., Nittala, M. G., & Sadda, S. R. (2021) · Indian Journal of Ophthalmology · doi:10.4103/ijo.ijo_1212_21 · Cited by 21
“[83-88] Previous studies have demonstrated that some specific disease features such as microaneurysms, exudates, and hemorrhages can be successfully identified using AI technologies.[83,89,90] More importantly, AI based on deep-learning algorithms have shown robust performance in detecting referable DR (defined as mod…”
BackgroundSarhan, M. H., Albarqouni, S., Yiğitsoy, M., Navab, N., & Eslami, A. (2020) · IET Image Processing · doi:10.1049/iet-ipr.2019.0804 · Cited by 17
Deep Learning in IoT: Introduction, Applications, and Perspective in the Big Data Era
Devika, G., & Karegowda, A. G. (2020) · Advances in computational intelligence and robotics book series · doi:10.4018/978-1-7998-2108-3.ch001 · Cited by 16
A Computationally Efficient Red-Lesion Extraction Method for Retinal Fundus Images
Monemian, M., & Rabbani, H. (2022) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2022.3229712 · Cited by 15
“There are a number of research works that focus on the identification of red-lesions from retinal fundus images [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36].”
BackgroundAn Incremental Feature Extraction Framework for Referable Diabetic Retinopathy Detection
Nandy, J., Hsu, W., & Lee, M. L. (2016) · IEEE International Conference on Tools with Artificial Intelligence (ICTAI) · doi:10.1109/ictai.2016.0140 · Cited by 15
“[3] proposed a method to dynamically select image patches to train a CNN model to identify hemorrhages in retina images.”
BackgroundMethodBlockchain-Based Privacy Access Control Mechanism and Collaborative Analysis for Medical Images
Prasad, P. S., Bethel, G. N. B., Singh, N., Gunjan, V. K., Basir, S., & Miah, S. (2022) · Security and Communication Networks · doi:10.1155/2022/9579611 · Cited by 14
RetractedNF-RCNN: Heart localization and right ventricle wall motion abnormality detection in cardiac MRI
Kermani, S., Oghli, M. G., Mohammadzadeh, A., & Kafieh, R. (2020) · Physica Medica · doi:10.1016/j.ejmp.2020.01.011 · Cited by 16
Deep Learning Techniques Dealing with Diabetes Mellitus: A Comprehensive Study
Das, S. K., Roy, P., & Mishra, A. K. (2021) · Studies in computational intelligence · doi:10.1007/978-981-15-9735-0_15 · Cited by 17
Atlas, L. G., Arjun, K. P., Kumar, K. S., Dhanaraj, R. K., & Nayyar, A. (2023) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2023.105692 · Cited by 14
Deep tessellated retinal image detection using Convolutional Neural Networks
Lyu, X., Li, H., Zhen, Y., Ji, X., & Zhang, S. (2017) · Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) · doi:10.1109/embc.2017.8036915 · Cited by 14
“In fundus image analysis, CNN has been applied to varieties of tasks, including anatomic structure segmentation [11], [12], pathologic lesion detection [13] and disease level grading [7].”
MethodDeng, Y., Chen, Y., Xie, L., Wang, L., & Juan, Z. (2022) · Frontiers in Oncology · doi:10.3389/fonc.2022.1001840 · Cited by 15
Retinal Image Segmentation for Diabetic Retinopathy Detection using U-Net Architecture
Deshmukh, S. V., Roy, A., & Agrawal, P. K. (2023) · International Journal of Image Graphics and Signal Processing · doi:10.5815/ijigsp.2023.01.07 · Cited by 13
“Several articles addressing ML and DL algorithms for colorful clinical retinal blood vessel images segmentation have been reported [1, 2], while spatial and spectral segmentation has seen a great amount of progress [3, 4].”
BackgroundClassification of Color Fundus Photographs Using Fusion Extracted Features and Customized CNN Models
Wang, J., Lu, N., Du, W.-C., Liu, K., Hsu, S., Wang, C.-Y., Chen, Y., Chang, L.-C., Twan, W., Chen, T., et al. (2023) · Healthcare · doi:10.3390/healthcare11152228 · Cited by 16
Verma, J., Kansal, I., Popli, R., Khullar, V., Singh, D., Snehi, J., & Kumar, R. (2024) · Communications in computer and information science · doi:10.1007/978-3-031-59091-7_14 · Cited by 17
Mao, J., Ma, X., Bi, Y.-L., & Zhang, R. (2024) · IEEE Transactions on Big Data · doi:10.1109/tbdata.2024.3442548 · Cited by 14
“The rationale for DR grading primarily relies on the manifestations and quantities of different DR lesions [7], [8], [9], which mainly include hard exudates (EX), hemorrhages (HE), microaneurysms (MA), and soft exudates (SE).”
Liu, J., Che, H., Zhao, A., Li, N., Huang, X., Li, H., & Jiang, Z. (2025) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2025.107581 · Cited by 14
An Overview of Medical Imaging Techniques for Knee Osteoarthritis Disease
Vashishtha, A., & Acharya, A. K. (2021) · Biomedical & Pharmacology Journal · doi:10.13005/bpj/2192 · Cited by 19
Selçuk, T., Beyoğlu, A., & Alkan, A. (2022) · Concurrency and Computation Practice and Experience · doi:10.1002/cpe.6768 · Cited by 15
Investigations of CNN for Medical Image Analysis for Illness Prediction
Nirmala, K., Saruladha, K., & Dekeba, K. (2022) · Computational Intelligence and Neuroscience · doi:10.1155/2022/7968200 · Cited by 12
Diverse lesion detection from retinal images by subspace learning over normal samples
Chen, B., Wang, L., Sun, J. Z., Chen, H., Fu, Y., Lan, S., Huang, Y., & Xu, Z. (2018) · Neurocomputing · doi:10.1016/j.neucom.2018.03.023 · Cited by 12
Abnormality detection in retinal image by individualized background learning
Chen, B., Wang, L., Wang, X., Sun, J. Z., Huang, Y., Feng, D. D., & Xu, Z. (2020) · Pattern Recognition · doi:10.1016/j.patcog.2020.107209 · Cited by 12
Application of Deep Learning in Biomedical Informatics and Healthcare
Nayak, D. K., Mishra, P., Das, P., Jamader, A. R., & Acharya, B. (2021) · Intelligent systems reference library · doi:10.1007/978-981-16-5304-9_9 · Cited by 15
Hemanth, S. V., & Alagarsamy, S. (2023) · Journal of Diabetes & Metabolic Disorders · doi:10.1007/s40200-023-01220-6 · Cited by 12
“…10.241%, 3.3042%, 2.3909% and 1.7136% higher than the existing state-of-art classifiers are ANN [28], ADL-CNN [20], CNN VGG16 [22], CNN [23], CNN [24], CNN [25], CNN [26], CNN [29], SeS CNN [30], AlexNet CNN [31], ConvNets CNN [32], Deep CNN [33], Faster RCNN [34] and RNN-LSTM respectively.”
BackgroundInfluential“How much realism is needed?” — the wrong question in silico imagers have been asking
Badano, A. G. (2017) · Medical Physics · doi:10.1002/mp.12187 · Cited by 13
Liu, X., Zhou, Y., Wang, Z., & Chen, X. (2018) · IEEE Access · doi:10.1109/access.2018.2838123 · Cited by 12
“Output signals of output layer neurons are full response to motivation models generated by originating nodes of an input layer [18]–[21].”
BackgroundAn Approach Towards Automatic Detection of Toxoplasmosis using Fundus Images
Chakravarthy, A. D., Abeyrathna, D. L. B., Subramaniam, M., Chundi, P., Halim, M. S., Hasanreisoğlu, M., Sepah, Y. J., & Nguyen, Q. D. (2019) · IEEE International Conference on Bioinformatics and Bioengineering (BIBE) · doi:10.1109/bibe.2019.00134 · Cited by 12
“Many of these approaches are designed to address tasks specific to diseases such as whole image classification for screening exams [13], or detection of exudate, hemorrhage and microaneurysms [11], [12], [14].”
Background(2022) · European Journal of Medical and Health Sciences · doi:10.34104/ejmhs.022.024040 · Cited by 15
Salehi, H. S., Mahdian, M., Murshid, M. M., Judex, S., & Tadinada, A. (2019) · Lasers in Dentistry · doi:10.1117/12.2510076 · Cited by 13
Automated Screening for Diabetic Retinopathy Using Compact Deep Networks
Lunscher, N., Chen, M. L., Jiang, N., & Zelek, J. (2017) · Journal of Computational Vision and Imaging Systems · doi:10.15353/vsnl.v3i1.182 · Cited by 11
On Deep Learning based algorithms for Detection of Diabetic Retinopathy
Thanati, H., Chalakkal, R. J., & Abdulla, W. H. (2019) · International Conference on Electronics, Information, and Communication (ICEIC) · doi:10.23919/elinfocom.2019.8706431 · Cited by 11
“Van Grinsven et al . in 2016 [27] also suggested speeding up the training of CNN by selective sampling.”
BackgroundMao, J., Luo, Y., Chen, K., Lao, J., Chen, L., Shao, Y., Zhang, C., Sun, M., & Shen, L. (2019) · Clinical and Experimental Ophthalmology · doi:10.1111/ceo.13670 · Cited by 12
“Studies of retinal BV segmentation of adults with CNN methods showed sensitivity and specificity of 0.7-0.9 and 0.8-0.9, 33-36 Van Grinsven et al 37 presented an algo-rithm based on DCNN for evaluating RHs and applied to the adult datasets of Kaggle 38 and Messidor.”
MethodInfluentialNi, J., Chen, Q., Liu, C., HongHao, W., Cao, Y., & Liu, B. (2019) · International Conference on Machine Learning and Applications · doi:10.1109/icmla.2019.00260 · Cited by 11
“To address this issue, we use the selective data sampling (SeS) technique [10] that dynamically selects misclassified negative samples during the training process based on classification by the”
MethodBohórquez, J. A. C., Granados, Y. C., & Fajardo, C. A. (2019) · Ciencia e Ingeniería Neogranadina · doi:10.18359/rcin.4156 · Cited by 13
“Recently some works have used the CNNs as a diagnostic tool of diseases such as heart attacks [20], colon cancer [18], melanoma [21], Alzheimer [22], cardiac arrhythmia’s [3] and hemorrhage detection [19].”
MethodPaul, A. J. (2021) · IEEE Madras Section Conference (MASCON) · doi:10.1109/mascon51689.2021.9563584 · Cited by 14
Rani, N. S., Bipin, N. B. J., & Yadhu, C. R. (2019) · IEEE Global Conference for Advancement in Technology (GCAT) · doi:10.1109/gcat47503.2019.8978422 · Cited by 11
“Van et al [11] had evaluated the performance of Convolutional neural networks with (SeS) and without (NSeS) the selective sampling method for detection of hemorrhages in color fundus images.”
MethodLabib, K., Ghumman, H., Jain, S., & Jarstad, J. S. (2024) · Cureus · doi:10.7759/cureus.71063 · Cited by 18
“This is demonstrated by the finding that AI generally detects more overt pathologies with higher accuracy, such as moderate-sized hemorrhages, compared to more subtle pathological findings, such as microhemorrhages [4].”
ResultDetecting red-lesions from retinal fundus images using unique morphological features
Monemian, M., & Rabbani, H. (2023) · Scientific Reports · doi:10.1038/s41598-023-30459-5 · Cited by 10
Damrawi, G. E., Zahran, M. A., Amin, E., & Abdelsalam, M. M. (2020) · Journal of Taibah University for Science · doi:10.1080/16583655.2020.1796244 · Cited by 11
“Improving and speed-up the Convolutional Neural Network (CNN) training algorithm for medical image analysis has been done in [16].”
BackgroundVij, R., & Arora, S. (2024) · Wireless Personal Communications · doi:10.1007/s11277-024-10968-w · Cited by 9
Islam, S., Deo, R. C., Barua, P. D., Soar, J., Yu, P., & Acharya, U. R. (2024) · IEEE Access · doi:10.1109/access.2024.3477420 · Cited by 17
Guo, S. (2024) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2024.107078 · Cited by 9
Hamer, A., Simms, D. M., & Waine, T. (2021) · International Journal of Remote Sensing · doi:10.1080/01431161.2020.1864059 · Cited by 9
“By directing the training to areas of known confusion in the background class the CNN was better able to separate edge cases, which supports other studies reporting a decrease in CNN performance with random sampling compared to a selective strategy (Van Grinsven et al. 2016).”
ResultOrdóñez, P. F., Cepeda, C. M., Garrido, J., & Chakravarty, S. (2017) · Journal of medical imaging · doi:10.1117/1.jmi.4.4.041309 · Cited by 10
Deep Red Lesion Classification for Early Screening of Diabetic Retinopathy
Ashraf, M. N., Hussain, M., & Habib, Z. (2022) · Mathematics · doi:10.3390/math10050686 · Cited by 9
Dense Fully Convolutional Network for Skin Lesion Segmentation.Preprint
Nasr-Esfahani, E., Rafiee, S., Jafari, M. H., Karimi, N., Wrobel, J. S., Soroushmehr, S. M. R., Samavi, S., & Najarian, K. (2017) · arXiv (Cornell University) · Cited by 9
Cross-modality transfer learning with knowledge infusion for diabetic retinopathy grading
Chen, T., Bai, Y., Mao, H., Liu, S., Xu, K., Xiong, Z., Ma, S., Yang, F., & Zhao, Y. (2024) · Frontiers in Medicine · doi:10.3389/fmed.2024.1400137 · Cited by 11
Zhao, M., Hou, C., Cao, L., & Zhang, J. (2025) · Applied Sciences · doi:10.3390/app15116085 · Cited by 10
Measuring, Quantifying, and Predicting the Cost-Accuracy Tradeoff
Baughman, M., Chakubaji, N., Truong, H., Kreics, K., Chard, K., & Foster, I. T. (2019) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata47090.2019.9006370 · Cited by 8
“The most common approach for reducing costs when training models is via reduction of training data, for example by applying selection criteria to obtain representative and diverse training data [13], [14].”
BackgroundBai, Y., Hao, J., Fu, H., Hu, Y. Z., Ge, X., Liu, J., Zhao, Y., & Zhang, J. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-16434-7_54 · Cited by 8
Detection of microaneurysms and hemorrhages based on improved Hessian matrix
Yang, L., Yan, S., & Xie, Y. (2021) · International Journal of Computer Assisted Radiology and Surgery · doi:10.1007/s11548-021-02358-5 · Cited by 8
“The resolution of Idark was adjusted from 4288 × 2848 to 512 × 512 [20], to halve computational times of eigenvalues of the Hessian matrix (Fig.”
MethodCalculation of ophthalmic diagnostic parameters on a single eye image based on deep neural network
Zhu, X., Song, X., Min, X., Zhou, H., Sun, W., Wang, J., & Zhai, G. (2021) · Multimedia Tools and Applications · doi:10.1007/s11042-021-11047-z · Cited by 8
“One important research field is retinal fundus photographs, for example, the automatic detection of diabetic macular edema and diabetic retinopathy [14] and the detection of hemorrhages on color fundus images [23].”
BackgroundRetinal multi-lesion segmentation by reinforcing single-lesion guidance with multi-view learning
Zhang, L., Fang, Z., Li, T., Xiao, Y., Zhou, J. T., & Yang, F. (2023) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2023.105349 · Cited by 8
Standvoss, K., Goerke, L., Crijns, T., van Niedek, T., Burgos, N. A., Janssen, D., van Vugt, J., Gerritse, E. J., Mol, J. H. Q., van de Vooren, D., et al. (2018) · SPIE Medical Imaging: Computer-Aided Diagnosis · doi:10.1117/12.2294016 · Cited by 8
Automating Visual Inspection with Convolutional Neural Networks
Das, S., Hollander, C. D., & Suliman, S. (2019) · Annual Conference of the PHM Society · doi:10.36001/phmconf.2019.v11i1.868 · Cited by 11
Detection of Proliferative Diabetic Retinopathy in Fundus Images Using Convolution Neural Network
Hassan, H. A., Yaakob, M., Ismail, S., Rahman, J. A., Rusni, I. M., Zabidi, A., Yassin, A. I. M., Tahir, N. M., & Shafie, S. M. (2020) · IOP Conference Series Materials Science and Engineering · doi:10.1088/1757-899x/769/1/012029 · Cited by 8
Li, P., Tao, J., Yuan, Q., Zhang, R., & Gao, P. (2025) · Frontiers in Cell and Developmental Biology · doi:10.3389/fcell.2025.1608994 · Cited by 12
A Survey on Intelligent Screening for Diabetic Retinopathy
Yulan, D., Chengzhang, Z., Xi, S., Zhenzhen, C., & Beiji, Z. (2019) · Chinese Medical Sciences Journal · doi:10.24920/003587 · Cited by 7
“Grinsven et al.(45) realized HE detection using a nine-layer CNN.”
MethodDu, M., Tao, M., Jian, H., Zhou, D., & Wang, S. (2020) · IEEE Access · doi:10.1109/access.2020.3008000 · Cited by 8
“…of the metachronous liver metastasis group are not statistically different from the other two groups: the ROC is established based on the texture parameter entropy value and uniformity, and the AUC value of the area under the curve of the diagnosis of liver metastasis is 0.73 to 0.78 [18].”
BackgroundInfluentialPriyanka, & Kumar, D. (2020) · Recent Advances in Computer Science and Communications · doi:10.2174/2666255813999200915141534 · Cited by 7
Raja, C., Santhosh Krishna, B., V, Balaji, L., Suman, S. K., Bhagyalakshmi, L., Alrashoud, M. R., Giri, J. P., & Sathish, T. (2024) · Automatika · doi:10.1080/00051144.2024.2409552 · Cited by 7
“The authors [54] claim that without the heat maps of the “Weights”, the CNNs are commonly perceived as black-boxes and that shortly CNN heat maps shall disclose the regions of significance for discrimination [55].”
Ophthalmic Diagnosis and Deep Learning - A Survey.
Sengupta, S., Singh, A., Leopold, H. A., & Lakshminarayanan, V. (2018) · Cited by 7
Character Recognition Algorithm Based on Fusion Probability Model and Deep Learning
Liu, Z., Pan, X., & Yuan, P. (2020) · The Computer Journal · doi:10.1093/comjnl/bxaa025 · Cited by 7
Wang, X., Wang, F., & Niu, Y. (2021) · Sensors · doi:10.3390/s21113693 · Cited by 7
“Anatomical or tissue object detection has receiving increasing attention in medical image analysis field since it facilitates to isolate the area of interest from the background and determine the relevant position and category in medical images [1–3].”
BackgroundUrsuleanu, T. F., Luca, A. R., Gheorghe, L., Grigorovici, R., Iancu, S., Hlusneac, M., Preda, C., & Grigorovici, A. (2021) · E-Health Telecommunication Systems and Networks · doi:10.4236/etsn.2021.102003 · Cited by 8
“BM models, introduced by [94] [95], are a family of RNNs that are easy to implement and that reproduce many probability distributions, BMs are used in image classification [4].”
A Cascade U‐Net With Transformer for Retinal Multi‐Lesion Segmentation
Zheng, H., & Liu, F. (2024) · International Journal of Imaging Systems and Technology · doi:10.1002/ima.23163 · Cited by 6
“The value of 𝛿 indicates the degree of blur-ring of Gaussian filtering, the larger the value the greater the degree of blurring of Gaussian filtering [34].”
Popescu, D., & Ichim, L. (2018) · Symmetry · doi:10.3390/sym10030073 · Cited by 6
“The authors in [45] used CNNs (Convolutional NNs) to detect hemorrhages.”
MethodKhomri, B., Christodoulidis, A., Djerou, L., Babahenini, M. C., & Farida Cheriet, A. (2018) · Journal of Biomedical Optics · doi:10.1117/1.jbo.23.5.056004 · Cited by 6
Deep Learning Techniques for Diabetic Retinopathy Detection
Qummar, S., Khan, F. G., Shah, S., Khan, A., Din, A., & Gao, J. (2020) · Current Medical Imaging Formerly Current Medical Imaging Reviews · doi:10.2174/1573405616666200213114026 · Cited by 6
Classification of retinal images in stages of diabetic retinopathy using deep learning
Lyona, C., Menezes, J., Shinde, T., Gavhane, M., Rohatgi, R., & Chavan, S. (2020) · International Conference on Advances in Computing, Communication Control and Networking (ICAC3N) · doi:10.1109/icacccn51052.2020.9362913 · Cited by 6
“The efficiency of convolutional neural network (CNN) model was proved in detecting hemorrhages in DR images by Van Grinsven et al. [15].”
MethodComputer-aided diagnostic model for retinal vascular diseases using graph-based attention mechanism
Sivapriya, G., Devi, R. M., & Keerthika, P. (2023) · The Journal of Supercomputing · doi:10.1007/s11227-023-05581-w · Cited by 6
“Particularly, research offering better segmentation results have always included deep learning (DL)-based approaches [12].”
A meta-analysis on diabetic retinopathy and deep learning applications
Erciyas, A., & Barışçı, N. (2023) · Multimedia Tools and Applications · doi:10.1007/s11042-023-17784-7 · Cited by 9
On Outsourcing Artificial Neural Network Learning of Privacy-Sensitive Medical Data to the Cloud
Melissourgos, D., Gao, H., Ma, C., Chen, S., & Wu, S. S. (2021) · IEEE International Conference on Tools with Artificial Intelligence (ICTAI) · doi:10.1109/ictai52525.2021.00062 · Cited by 5
IoT‐Based Solutions for Smart Healthcare
Jain, P., Panesar, S. F., Talwar, B. F., & Sah, M. K. (2021) · Wiley eBooks · doi:10.1002/9781119792345.ch2 · Cited by 8
Damrawi, G. E., Zahran, M. A., Amin, E., & Abdelsalam, M. M. (2021) · Journal of Ambient Intelligence and Humanized Computing · doi:10.1007/s12652-021-03565-3 · Cited by 6
“Another medical image analysis via an improved Convolutional Neural Network (CNN) in (van Grinsven et al. 2016).”
MethodPerformance Analysis of Deep Learning based Segmentation of Retinal Lesions in Fundus Images
Baskaran, N. K., & Mahesh, T. R. (2023) · International Conference on Electronics and Renewable Systems (ICEARS) · doi:10.1109/icears56392.2023.10085616 · Cited by 6
“[16] explored a method to detect hemorrhages in images of the retinal fundus.”
MethodIntelligent identification and classification of diabetic retinopathy using fuzzy inference system
Medhi, J. P., Sandeep, R., Datta, P., & Nizami, T. K. (2023) · Computer Methods in Biomechanics and Biomedical Engineering Imaging & Visualization · doi:10.1080/21681163.2023.2235014 · Cited by 5
Multi-scale multi-attention network for diabetic retinopathy grading
Xia, H., Long, J., Song, S., & Tan, Y. (2023) · Physics in Medicine and Biology · doi:10.1088/1361-6560/ad111d · Cited by 6
AI and retinal image analysis at Baidu
Yang, Y., Yang, D., Xu, Y., Wang, L., Huang, Y., Li, X., Liu, X. E., & La, L. V. (2019) · Elsevier eBooks · doi:10.1016/b978-0-08-102816-2.00020-4 · Cited by 5
A Deep Learning Approach for Retinal Image Feature Extraction
Hoque, M. E., Kipli, K., Zulcaffle, T. M. A., Al-Hababi, A. Y. S., Mat, D. A. A., Sapawi, R., & Joseph, A. (2021) · Pertanika journal of science & technology · doi:10.47836/pjst.29.4.17 · Cited by 7
Automatic classification of diabetic retinopathy based on convolutional neural networks
Fang, M., Zhang, X., Zhang, W., Xue, J., & Wu, L. (2018) · International Conference on Image, Video Processing and Artificial Intelligence (IVPAI) · doi:10.1117/12.2503883 · Cited by 5
“In recent years, inspired by the success of CNNs in computer vision [7-9], many scholars tried to use CNNs to learn high-level features of DR lesions [10-11].”
BackgroundSmall Scale Feature Propagation Using Deep Residual Learning for Diabetic Retinopathy Classification
Metan, A. C., Lambert, A. J., & Pickering, M. R. (2019) · International Conference on Image, Vision and Computing (ICIVC) · doi:10.1109/icivc47709.2019.8981096 · Cited by 5
“Unlike how previously published methods trained their algorithms on smaller subsets of images initially [7], or have used class weighting for each batch individually [2], we have found that this method somewhat inhibits the class imbalance issue while also enabling us to utilize the whole dataset.”
MethodSCRD‐Net: A Deep Convolutional Neural Network Model for Glaucoma Detection in Retina Tomography
Wang, H., Hu, J., & Zhang, J. (2021) · Complexity · doi:10.1155/2021/9858343 · Cited by 6
“Numerous classification schemes have been proposed for diabetic retinopathy [23, 24], hemorrhage detection [25], optic disc abnormality detection [26], age-related macular degeneration (AMD) [27, 28], diabetic macular edema (DME) [29], multiclassification [30, 31], and other conditions [32–34].”
BackgroundDIAGNOSING THORAX DISEASES USING DEEP LEARNING MODELS
Shadeed, G. A., Tawfeeq, M. A., & Mahmoud, S. M. (2020) · Journal of Engineering and Sustainable Development · doi:10.31272/jeasd.conf.1.12 · Cited by 5
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.