Exploiting Epistemic Uncertainty of Anatomy Segmentation for Anomaly Detection in Retinal OCT
Open accessThe DOI matches its Crossref record; 56 of the 56 references it cites were found in a registry or an index.
Abstract
Diagnosis and treatment guidance are aided by detecting relevant biomarkers in medical images. Although supervised deep learning can perform accurate segmentation of pathological areas, it is limited by requiring a priori definitions of these regions, large-scale annotations, and a representative patient cohort in the training set. In contrast, anomaly detection is not limited to specific definitions of pathologies and allows for training on healthy samples without annotation. Anomalous regions can then serve as candidates for biomarker discovery. Knowledge about normal anatomical structure brings implicit information for detecting anomalies. We propose to take advantage of this property using Bayesian deep learning, based on the assumption that epistemic uncertainties will correlate with anatomical deviations from a normal training set. A Bayesian U-Net is trained on a well-defined healthy environment using weak labels of healthy anatomy produced by existing methods. At test time, we capture epistemic uncertainty estimates of our model using Monte Carlo dropout. A novel post-processing technique is then applied to exploit these estimates and transfer their layered appearance to smooth blob-shaped segmentations of the anomalies. We experimentally validated this approach in retinal optical coherence tomography (OCT) images, using weak labels of retinal layers. Our method achieved a Dice index of 0.789 in an independent anomaly test set of age-related macular degeneration (AMD) cases. The resulting segmentations allowed very high accuracy for separating healthy and diseased cases with late wet AMD, dry geographic atrophy (GA), diabetic macular edema (DME) and retinal vein occlusion (RVO). Finally, we qualitatively observed that our approach can also detect other deviations in normal scans such as cut edge artifacts.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
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?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Hiasa al. [684] | et | 2019 | Bayesian U-Net | Segmentation | Cancer | 2 | MC dropout”
From A review of uncertainty quantification in deep learning: Techniques, applications and challenges (2021) · cites this paper as [684]
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
U-Net: Convolutional Networks for Biomedical Image Segmentation
Ronneberger, O., Fischer, P., & Brox, T. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-24574-4_28 · Cited by 99,404
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946 · cited 1×
“We used Kaiming initialization [43], Adam optimization [44], and a learning rate of 1 × 10−4 (multiplied by 0.”
MethodBatch 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×
“applied after each convolutional block, which consists of two 3×3 convolutions, each followed by batch-normalization [40] and a rectified linear unit (ReLU).”
MethodDelving 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×
“We used Kaiming initialization [43], Adam optimization [44], and a learning rate of 1 × 10−4 (multiplied by 0.”
MethodA survey on deep learning in medical image analysis
Litjens, G., Kooi, T., Bejnordi, B. E., Setio, A. A. A., Ciompi, F., Ghafoorian, M., van der Laak, J. A. W. M., van Ginneken, B., & Sánchez, C. I. (2017) · Medical Image Analysis · doi:10.1016/j.media.2017.07.005 · Cited by 15,332 · cited 1×
“Automated methods can partially address these issues by exploiting the potential of deep learning [3].”
BackgroundDermatologist-level classification of skin cancer with deep neural networks
Esteva, A., Kuprel, B., Novoa, R. A., Ko, J., Swetter, S. M., Blau, H. M., & Thrun, S. (2017) · Nature · doi:10.1038/nature21056 · Cited by 14,362 · cited 2×
“markers such as disease lesions [4], [7], [8], [34].”
BackgroundBatch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
Ioffe, S. (2024) · arXiv (Cornell University) · doi:10.57702/o9raffed · Cited by 14,214
Wong, W. L., Su, X., Li, X., Cheung, C. M. G., Klein, R., Cheng, C.-Y., & Wong, T. Y. (2014) · The Lancet Global Health · doi:10.1016/s2214-109x(13)70145-1 · Cited by 5,331 · cited 1×
“Age-related macular degeneration (AMD) is one of the leading causes of blindness in the world [26].”
BackgroundIdentifying Medical Diagnoses and Treatable Diseases by Image-Based Deep Learning
Kermany, D. S., Goldbaum, M., Cai, W., Valentim, C. C. S., Liang, H., Baxter, S. L., McKeown, A., Yang, G., Wu, X., Yan, F., et al. (2018) · Cell · doi:10.1016/j.cell.2018.02.010 · Cited by 4,842
What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?
Kendall, A., & Gal, Y. (2017) · arXiv · doi:10.48550/arxiv.1703.04977 · Cited by 3,088 · cited 5×
“Bayesian deep learning has emerged as an active field of research that aims to develop efficient tools for quantifying uncertainties [18]–[20].”
BackgroundMethodInfluential
Cited by
A review of uncertainty quantification in deep learning: Techniques, applications and challenges
Abdar, M., Pourpanah, F., Hussain, S., Rezazadegan, D., Liu, L., Ghavamzadeh, M., Fieguth, P., Cao, X., Khosravi, A., Acharya, U. R., et al. (2021) · Information Fusion · doi:10.1016/j.inffus.2021.05.008 · Cited by 3,111
A survey of uncertainty in deep neural networks
Gawlikowski, J., Tassi, C. R. N., Ali, M., Lee, J., Humt, M., Feng, J., Kruspe, A., Triebel, R., Jung, P., Roscher, R., et al. (2023) · Artificial Intelligence Review · doi:10.1007/s10462-023-10562-9 · Cited by 1,406
“uncertainty is estimated in segmentation tasks using approximates of Bayesian inference [1], [2], [4], [152], [261], [262], [263], [264] or test-time data augmentation techniques [249].”
MethodA Unifying Review of Deep and Shallow Anomaly Detection
Ruff, L., Kauffmann, J. R., Vandermeulen, R. A., Montavon, G., Samek, W., Kloft, M., Dietterich, T. G., & Muller, K.-R. (2021) · Proceedings of the IEEE · doi:10.1109/jproc.2021.3052449 · Cited by 914
Uninformed Students: Student-Teacher Anomaly Detection With Discriminative Latent Embeddings
Bergmann, P., Fauser, M., Sattlegger, D., & Steger, C. (2020) · 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr42600.2020.00424 · Cited by 990
Deep learning and medical image processing for coronavirus (COVID-19) pandemic: A survey
Bhattacharya, S., Reddy Maddikunta, P. K., Pham, Q.-V., Gadekallu, T. R., Krishnan S, S. R., Chowdhary, C. L., Alazab, M., & Jalil Piran, M. (2021) · Sustainable Cities and Society · doi:10.1016/j.scs.2020.102589 · Cited by 483
“The authors in Seeb ¨ ock et al. (2019) use this idea to implement a Bayesian DL technique assuming that the epistemic uncertainties would relate to the anatomical eccentricities from the training dataset.”
Deep Learning for Medical Anomaly Detection – A Survey
Fernando, T., Gammulle, H., Denman, S., Sridharan, S., & Fookes, C. (2022) · ACM Computing Surveys · doi:10.1145/3464423 · Cited by 399
Seoni, S., Jahmunah, V., Salvi, M., Barua, P. D., Molinari, F., & Acharya, U. R. (2023) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2023.107441 · Cited by 239
A review of uncertainty estimation and its application in medical imaging
Zou, K., Chen, Z., Yuan, X., Shen, X., Wang, M., & Fu, H. (2023) · Meta-Radiology · doi:10.1016/j.metrad.2023.100003 · Cited by 158
Label-Free Segmentation of COVID-19 Lesions in Lung CT
Yao, Q., Xiao, L., Liu, P., & Zhou, S. K. (2021) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2021.3066161 · Cited by 176
A graph neural network method for distributed anomaly detection in IoT
Protogerou, A., Papadopoulos, S., Drosou, A., Tzovaras, D., & Refanidis, I. (2021) · Evolving Systems · doi:10.1007/s12530-020-09347-0 · Cited by 129
“Numerous literature research studies have been conducted within this application domain (Seeböck et al. 2019), (Chen et al. 2018), (Nair et al. 2020).”
BackgroundHuang, L., Ruan, S., Xing, Y., & Feng, M. (2024) · Medical Image Analysis · doi:10.1016/j.media.2024.103223 · Cited by 121
Deep Learning Approaches for Detecting COVID-19 From Chest X-Ray Images: A Survey
Alghamdi, H. S., Amoudi, G., Elhag, S., Saeedi, K., & Nasser, J. (2021) · IEEE Access · doi:10.1109/access.2021.3054484 · Cited by 126
“For example, DL has been successfully applied to microscopy images [11], brain tumor classification [12], MRI images [13], and retinal photographs [14].”
MethodUltra-short-term wind power forecasting based on deep Bayesian model with uncertainty
Liu, L., Liu, J., Ye, Y., Liu, H., Chen, K., Li, D., Xue, D. Q., & Sun, M. (2023) · Renewable Energy · doi:10.1016/j.renene.2023.01.038 · Cited by 128
Anomaly Detection in Medical Imaging - A Mini Review
Tschuchnig, M., & Gadermayr, M. (2022) · Springer eBooks · doi:10.1007/978-3-658-36295-9_5 · Cited by 166
“Several advancements have also been made in OCT segmentation of retina lesions [7], [8], [25], [26], [48], with one publication performing visual touring test using 2 experts, which were unable to recognize differences in the correctly reconstructed data [8].”
BackgroundCollin, A.-S., & De Vleeschouwer, C. (2021) · 2020 25th International Conference on Pattern Recognition (ICPR) · doi:10.1109/icpr48806.2021.9412842 · Cited by 106
“structures can be inferred from the reconstruction (red block), following a traditional approach based on the residual [2], or even from an estimation of the prediction uncertainty [3].”
BackgroundMethodAI-based monitoring of retinal fluid in disease activity and under therapy
Schmidt-Erfurth, U. M., Reiter, G. S., Riedl, S., Seeböck, P., Vogl, W., Blodi, B. A., Domalpally, A., Fawzi, A. A., Jia, Y., Sarraf, D., et al. (2021) · Progress in Retinal and Eye Research · doi:10.1016/j.preteyeres.2021.100972 · Cited by 108
Self-citationSegmentMeIfYouCan: A Benchmark for Anomaly SegmentationPreprint
Chan, R., Lis, K., Uhlemeyer, S., Blum, H., Honari, S., Siegwart, R., Fua, P., Salzmann, M., & Rottmann, M. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2104.14812 · Cited by 224
“tomography images [45] or brain MRIs [46].”
The Role of Generative Adversarial Network in Medical Image Analysis: An In-depth Survey
Alamir, M., & Alghamdi, M. (2022) · ACM Computing Surveys · doi:10.1145/3527849 · Cited by 103
“(g ) Healthy retina OCT image[174].”
Tian, Y., Pang, G., Liu, F., Chen, Y., Shin, S. H., Verjans, J. W., Singh, R., & Carneiro, G. (2021) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-87240-3_13 · Cited by 77
Uncertainty-aware deep learning in healthcare: A scoping review
Loftus, T. J., Shickel, B., Ruppert, M. M., Balch, J. A., Ozrazgat-Baslanti, T., Tighe, P. J., Efron, P. A., Hogan, W. R., Rashidi, P., Upchurch, G. R., et al. (2022) · PLOS Digital Health · doi:10.1371/journal.pdig.0000085 · Cited by 73
Deep learning in retinal optical coherence tomography (OCT): A comprehensive survey
Viedma, I. A., Alonso‐Caneiro, D., Read, S. A., & Collins, M. J. (2022) · Neurocomputing · doi:10.1016/j.neucom.2022.08.021 · Cited by 82
Hanson, R. L. W., Airody, A., Sivaprasad, S., & Gale, R. P. (2022) · Eye · doi:10.1038/s41433-022-02360-4 · Cited by 77
Yan, Y., Wang, D., Zhou, G., & Chen, Q. (2021) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2021.3107586 · Cited by 60
Proxy-Bridged Image Reconstruction Network for Anomaly Detection in Medical Images
Zhou, K., Li, J., Luo, W., Li, Z., Yang, J., Fu, H., Cheng, J., Liu, J., & Gao, S. (2021) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2021.3118223 · Cited by 73
Hassan, T., Akram, M. U., Werghi, N., & Nazir, M. N. (2020) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2020.2982914 · Cited by 68
“[36] presented a framework based on Bayesian”
BackgroundMethodInfluentialHematoma Expansion Context Guided Intracranial Hemorrhage Segmentation and Uncertainty Estimation
Li, X., Luo, G., Wang, W., Wang, K., Gao, Y., & Li, S. (2021) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2021.3103850 · Cited by 63
“Uncertainty estimation is important for medical image segmentation tasks [32].”
BackgroundTian, Y., Liu, F., Pang, G., Chen, Y., Liu, Y., Verjans, J. W. H., Singh, R., & Carneiro, G. (2023) · Medical Image Analysis · doi:10.1016/j.media.2023.102930 · Cited by 78
“…methods are generally based on a one-class classifier (OCC) that learns a normal image distribution from the normal training images, and test image anomalies (or abnormal images) are detected based on the extent that they deviate from the learned distribution [8,11,22,36,42,46,48, 50,55–57,61,62].”
BackgroundMethodKumari, S., Prabha, C., Karim, A., Hassan, M. M., & Azam, S. (2024) · IET Information Security · doi:10.1049/2024/8821891 · Cited by 83
Approaches to quantify optical coherence tomography angiography metrics
Tan, B., Sim, R. Z. H., Chua, J., Wong, D. W. K., Yao, X., Garhöfer, G., Schmidl, D., Werkmeister, R. M., & Schmetterer, L. F. (2020) · Annals of Translational Medicine · doi:10.21037/atm-20-3246 · Cited by 51
“Various ML methods including support vector machine (88), random forest (89), neural network (90), and DL architectures were applied to retinal segmentation in healthy and diseased eyes (91-101).”
MethodCalderon-Ramirez, S., Yang, S., Moemeni, A., Colreavy‐Donnelly, S., Elizondo, D. A., Oala, L., Rodríguez‐Capitán, J., Jiménez-Navarro, M. F., López‐Rubio, E., & Molina‐Cabello, M. A. (2021) · IEEE Access · doi:10.1109/access.2021.3085418 · Cited by 51
“For instance, comparing uncertainty estimations of [25], [26] to the ones yielded in [40], is difficult as different uncertainty measures yield different uncertainty value ranges for different data sets.”
BackgroundMethodQuantitative Comparison of Monte-Carlo Dropout Uncertainty Measures for Multi-class Segmentation
Camarasa, R., Bos, D., Hendrikse, J., Nederkoorn, P. J., Kooi, M. E., van der Lugt, A., & de Bruijne, M. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-60365-6_4 · Cited by 63
Tian, Y., Pang, G., Liu, Y., Wang, C., Chen, Y., Liu, F., Singh, R., Verjans, J. W. H., Wang, M., & Carneiro, G. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-45676-3_2 · Cited by 61
REB: Reducing biases in representation for industrial anomaly detection
Lyu, S., Mo, D., & Wong, W. K. (2024) · Knowledge-Based Systems · doi:10.1016/j.knosys.2024.111563 · Cited by 47
“This technique has been widely applied in many fields, such as industrial product surface inspection [53], video analysis [1], and medical diagnosis [47, 48].”
MethodBurlina, P., Paul, W., Liu, T. Y. A., & Bressler, N. M. (2022) · JAMA Ophthalmology · doi:10.1001/jamaophthalmol.2021.5557 · Cited by 37
Patro, K. K., Allam, J. P., Hammad, M. A., Tadeusiewicz, R., & Pławiak, P. (2023) · Journal of Applied Biomedicine · doi:10.1016/j.bbe.2023.01.005 · Cited by 39
“DL has been proven its effectiveness in brain tumour classification [14], MRI image analysis [15], skin cancer, lung infection [16] and retinal images [17].”
BackgroundLee, M. S., Yang, S. W., & Han, S. W. (2023) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv56688.2023.00065 · Cited by 42
“Uncertainty quantification is important for precise decision-making in various domains [1], such as autonomous driving [11, 6] or medical image analysis [23, 36, 35, 29, 34].”
BackgroundMedical Imaging with Deep Learning for COVID- 19 Diagnosis: A Comprehensive ReviewPreprint
Bharati, S., Podder, P., Mondal, M. R. H., & Prasath, V. B. S. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2107.09602 · Cited by 58
“The researchers of [88] apply this concept to develop a DL-based Bayesian system, adopting that epistemic suspicions are related to the training dataset with its anatomical eccentricities.”
BackgroundThe MVTec AD 2 Dataset: Advanced Scenarios for Unsupervised Anomaly Detection
Heckler-Kram, L., Neudeck, J.-H., Scheler, U., König, R., & Steger, C. T. (2026) · International Journal of Computer Vision · doi:10.1007/s11263-026-02743-0 · Cited by 96
Self-supervised anomaly detection, staging and segmentation for retinal images
Li, Y., Lao, Q., Kang, Q., Jiang, Z., Du, S., Zhang, S., & Li, K. (2023) · Medical Image Analysis · doi:10.1016/j.media.2023.102805 · Cited by 36
Gupta, R., Tripathi, V., & Gupta, A. K. (2021) · International Conference on Computational Performance Evaluation (ComPE) · doi:10.1109/compe53109.2021.9752188 · Cited by 33
RetractedZhang, H., Yang, B., Li, S., Zhang, X., Li, X., Liu, T., Higashita, R., & Liu, J. (2025) · Computerized Medical Imaging and Graphics · doi:10.1016/j.compmedimag.2025.102539 · Cited by 44
Deep Learning based Anomaly Detection in Images: Insights, Challenges and Recommendations
Alloqmani, A., Yoosef, B., Irshad, A., & Alsolami, F. J. (2021) · International Journal of Advanced Computer Science and Applications · doi:10.14569/ijacsa.2021.0120428 · Cited by 34
Computer aided diagnosis of diabetic macular edema in retinal fundus and OCT images: A review
Chakrapani, P. K., Kumar, P., Geetha, M. K., & Bhandary, S. V. (2023) · Journal of Applied Biomedicine · doi:10.1016/j.bbe.2022.12.005 · Cited by 30
Mou, L., Liang, L., Gao, Z., & Wang, X. (2022) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2022.103619 · Cited by 29
Anomaly Detection using Score-based Perturbation Resilience
Shin, W., Lee, J., Lee, T.-H., Lee, S., & Yun, J. P. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.02136 · Cited by 32
“This approach is used not only in the manufacturing industry but also in medical diagnostics [53, 30, 10], video surveillance [36, 21, 13], and autonomous driving [18, 44, 15].”
MethodUnsupervised 3D Brain Anomaly Detection
Viana, J. S., de la Rosa, E., Vyvere, T. V., Robben, D., Sima, D. M., & CENTER-TBI Participants and Investigators (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-72084-1_13 · Cited by 32
Hassan, T., Akram, M. U., & Werghi, N. (2020) · IEEE International Conference on Bioinformatics and Bioengineering (BIBE) · doi:10.1109/bibe50027.2020.00099 · Cited by 30
Udayaraju, P., Jeyanthi, P., & Sekhar, B. V. D. S. (2023) · Soft Computing · doi:10.1007/s00500-023-08928-w · Cited by 24
“Seebock et al. (2020) described a Bayesian DL approach depending on the assumption that epistemological uncertainty correlates with anatomical deviation from the common training set.”
RetractedPeng, Y., Lin, A., Wang, M., Lin, T., Liu, L., Wu, J., Zou, K., Shi, T., Feng, L., Liang, Z., et al. (2024) · Cell Reports Medicine · doi:10.1016/j.xcrm.2024.101876 · Cited by 50
“14 Liu et al. used uncertainty strategy to enhance the accuracy of boundary localization of 7 retinal biomarkers but did not investigate image classification.”
A Bayesian deep learning method for freeway incident detection with uncertainty quantification
Liu, G., Jin, H., Li, J., Hu, X., & Li, J. (2022) · Accident Analysis & Prevention · doi:10.1016/j.aap.2022.106796 · Cited by 29
Improving Uncertainty Estimations for Mammogram Classification using Semi-Supervised Learning
Calderon-Ramirez, S., Murillo-Hernandez, D., Rojas-Salazar, K., Calvo-Valverde, L.-A., Yang, S., Moemeni, A., Elizondo, D. A., López‐Rubio, E., & Molina‐Cabello, M. A. (2021) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn52387.2021.9533719 · Cited by 26
An improved agglomerative hierarchical clustering anomaly detection method for scientific data
Shi, P., Zhao, Z., Zhong, H., Shen, H., & Ding, L. (2020) · Concurrency and Computation Practice and Experience · doi:10.1002/cpe.6077 · Cited by 22
“7 However, the inconvenience of obtaining labeled samples acts as a block of common usage of supervised and semi-supervised methods.”
BackgroundModality preserving U-Net for segmentation of multimodal medical images
Wu, B., Zhang, F., Xu, L., Shen, S., Shao, P., Sun, M., Liu, P., Yao, P., & Xu, R. X. (2023) · Quantitative Imaging in Medicine and Surgery · doi:10.21037/qims-22-1367 · Cited by 25
“In the process of biomedical image analysis, image segmentation help to focus on pathology (4), track disease progression (5), and characterize anatomical structures and defects (6).”
BackgroundOnitiu, D. (2022) · Information & Communications Technology Law · doi:10.1080/13600834.2022.2116354 · Cited by 36
Leveraging the Bhattacharyya coefficient for uncertainty quantification in deep neural networks
Molle, P. V., Verbelen, T., Vankeirsbilck, B., Vylder, J. D., Diricx, B., Kimpe, T. R. L., Simoens, P., & Dhoedt, B. (2021) · Neural Computing and Applications · doi:10.1007/s00521-021-05789-y · Cited by 25
Araújo, T., Aresta, G., Schmidt-Erfurth, U. M., & Bogunović, H. (2023) · Scientific Reports · doi:10.1038/s41598-023-43018-9 · Cited by 30
“A notable approach is uncertainty estimation via Monte Carlo (MC) dropout 18 , which has been applied for referable DR detection in eye fundus images 44 as well as in OCT classification and segmentation tasks 7,44–46 .”
MethodSelf-citationVLFATRollout: Fully transformer-based classifier for retinal OCT volumes
Oghbaie, M., Araújo, T., Schmidt-Erfurth, U. M., & Bogunović, H. (2024) · Computerized Medical Imaging and Graphics · doi:10.1016/j.compmedimag.2024.102452 · Cited by 18
Self-citationQuantifying deep neural network uncertainty for atrial fibrillation detection with limited labels
Chen, B., Javadi, G., Hamilton, A., Sibley, S. R., Laird, P., Abolmaesumi, P., Maslove, D. M., & Mousavi, P. (2022) · Scientific Reports · doi:10.1038/s41598-022-24574-y · Cited by 23
“…and uncertain estimation has been applied to generate more accurate, higher quality results for medical image translation 35 segmentation 32,40 , enable better targeting of ultrasound-guided prostate biopsies 41 , as well as in non-healthcare domains such as improving the data…”
MethodDiabetic retinopathy detection through generative AI techniques: A review
Bansal, V. K., Jain, A., & Walia, N. K. (2024) · Results in Optics · doi:10.1016/j.rio.2024.100700 · Cited by 25
Lesion-aware convolutional neural network for chest radiograph classification
Li, F., Shi, J., Yan, L., Wang, Y.-G., Zhang, X.-D., Jiang, M., Wu, Z., & Zhou, K.-Q. (2020) · Clinical Radiology · doi:10.1016/j.crad.2020.08.027 · Cited by 19
Applications of deep learning for detecting ophthalmic diseases with ultrawide-field fundus images
Tang, Q., Wang, H., Wu, D., & Zhang, M. (2023) · International Journal of Ophthalmology · doi:10.18240/ijo.2024.01.24 · Cited by 16
Wang, Y., Zhu, Y., Lou, G., Zhang, P., Chen, J., & Li, J. (2021) · Journal of Biomedical Informatics · doi:10.1016/j.jbi.2021.103930 · Cited by 18
Seeböck, P., Orlando, J. I., Michl, M., Mai, J., Schmidt-Erfurth, U. M., & Bogunović, H. (2024) · Medical Image Analysis · doi:10.1016/j.media.2024.103104 · Cited by 17
Self-citationThe Evidence Contraction Issue in Deep Evidential Regression: Discussion and Solution
Wu, Y., Shi, B., Dong, B., Zheng, Q., & Wei, H. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i19.30172 · Cited by 29
“Despite the attractiveness of Deep Learning, their deployment in high-risk domains such as Weather Prediction (Bi et al. 2023), Vehicle Control (Choi et al. 2019) and Medical Di-agnostics (Seebock et al. 2020) is still limited, which is attributed to the fact that Deep Learning models are subject to uncertainty.”
BackgroundTruong, A. M., & Luong, H. Q. (2024) · Current Research in Food Science · doi:10.1016/j.crfs.2024.100758 · Cited by 16
“…unsupervised learning for anomaly detection, which can be classified into two different approaches: one utilizing per-pixel reconstruction errors (Seeb ¨ ock et al. (2020); Collin and De Vleeschouwer (2021); Bergmann et al. (2020)), and the other evaluating the density obtained from the…”
MethodQuantifying Uncertainty of Deep Neural Networks in Skin Lesion Classification
Molle, P. V., Verbelen, T., Boom, C. D., Vankeirsbilck, B., Vylder, J. D., Diricx, B., Kimpe, T. R. L., Simoens, P., & Dhoedt, B. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-32689-0_6 · Cited by 15
Natural gas leakage detection from offshore platform by OGI camera and unsupervised deep learning
Li, J., Qian, X., Shi, J., Xie, Z., Chang, Y., & Chen, G. (2024) · Journal of Loss Prevention in the Process Industries · doi:10.1016/j.jlp.2024.105449 · Cited by 12
A Quantitative Comparison of Epistemic Uncertainty Maps Applied to Multi-Class Segmentation
Camarasa, R., Bos, D., Hendrikse, J., Nederkoorn, P. J., Kooi, M. E., van der Lugt, A., & de Bruijne, M. (2021) · The Journal of Machine Learning for Biomedical Imaging · doi:10.59275/j.melba.2021-ec49 · Cited by 20
“The use of description measure to assess uncertainty is widely spread in the literature (Seeb¨ock et al., 2019; Kendall et al., 2017).”
MethodOutbreak COVID-19 in Medical Image Processing Using Deep Learning: A State-of-the-Art Review
Kaur, J., & Kaur, P. (2021) · Archives of Computational Methods in Engineering · doi:10.1007/s11831-021-09667-7 · Cited by 13
New Deep Learning Models for Medical Imaging: Deep Belief Network, GAN, Autoencoder
Gupta, R., Shukla, S. K., & Tripathi, V. (2023) · International Conference on Smart Electronics and Communication (ICOSEC) · doi:10.1109/icosec58147.2023.10276257 · Cited by 11
Uncertainty CNNs: A path to enhanced medical image classification performance
Papageorgiou, V. Ε., Πετμεζάς, Γ., Dogoulis, P., Cordy, M., & Maglaveras, N. (2025) · Mathematical Biosciences & Engineering · doi:10.3934/mbe.2025020 · Cited by 13
Li, Y., Yu, B., Si, M., Yang, M., Cui, W., Zhou, Y., Fu, S., Wang, H., Liu, X., & Zhang, H. (2025) · International Ophthalmology · doi:10.1007/s10792-025-03439-z · Cited by 15
Quantile Autoencoder With Abnormality Accumulation for Anomaly Detection of Multivariate Sensor Data
Ryu, S., Yim, J., Seo, J., Yu, Y., & Seo, H. (2022) · IEEE Access · doi:10.1109/access.2022.3187426 · Cited by 13
One Step Closer to Unbiased Aleatoric Uncertainty Estimation
Wang, Z., Martin, Z., Das, S., Weng, T.-W., Megretski, A., Daniel, L., & Nguyen, L. M. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i15.29627 · Cited by 20
Learning and predicting the unknown class using evidential deep learning
Nagahama, A. (2023) · Scientific Reports · doi:10.1038/s41598-023-40649-w · Cited by 14
Prototype-based fuzzy rough sets for outlier detection
Cai, M., Qi, D., Huang, C., & Zhan, J. (2025) · Fuzzy Sets and Systems · doi:10.1016/j.fss.2025.109460 · Cited by 12
Inoda, S., Takahashi, H., Yamagata, H., Hisadome, Y., Kondo, Y., Tampo, H., Sakamoto, S., Katada, Y., Kurihara, T., Kawashima, H., et al. (2022) · Scientific Reports · doi:10.1038/s41598-022-25894-9 · Cited by 9
Yun, C., Eom, B., Park, S.-J., Kim, C., Kim, D., Jabeen, F., Kim, W. H., Kim, H. J., & Kim, J. (2023) · Sensors · doi:10.3390/s23052864 · Cited by 10
Comparing code-free and bespoke deep learning approaches in ophthalmology
Wong, C. Y. T., O’Byrne, C., Taribagil, P., Liu, T., Antaki, F., & Keane, P. A. (2024) · Graefe s Archive for Clinical and Experimental Ophthalmology · doi:10.1007/s00417-024-06432-x · Cited by 10
“This is because investments in the construction of traditional DL models for mass data analyses tend to be financially dissuasive in the face of little proof of success [44, 45].”
BackgroundDeep Convolutional Neural Networks for Detecting COVID-19 Using Medical Images: A Survey
Khattab, R., Abdelmaksoud, I. R., & Abdelrazek, S. M. (2023) · New Generation Computing · doi:10.1007/s00354-023-00213-6 · Cited by 11
Martinez, J. A. C., da Costa, G. A. O. P., Messias, C. G., Soler, L. D. S., Almeida, C. A., & Feitosa, R. Q. (2024) · ISPRS Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.isprsjprs.2024.03.004 · Cited by 9
Inoda, S., Takahashi, H., Arai, Y., Tampo, H., Matsui, Y., Kawashima, H., & Yanagi, Y. (2023) · Graefe s Archive for Clinical and Experimental Ophthalmology · doi:10.1007/s00417-023-06054-9 · Cited by 9
“[10, 11] Moreover, AI can not only detect exudative AMD, [12] but also predict AMD progression.”
BackgroundSchwarzenbacher, L., Seeböck, P., Schartmüller, D., Leydolt, C., Menapace, R., & Schmidt-Erfurth, U. M. (2022) · Acta Ophthalmologica · doi:10.1111/aos.15141 · Cited by 9
“Up to now, PCopacification (PCO) is the most common complication of cataract surgery, occurring in 18.4 to 50% of patients after two to five years and is treated with Nd:YAG laser capsulotomy if not prevented by a PPCCC or PPLC (Apple et al. 1992; Stager et al. 2006).”
BackgroundInfluentialSelf-citationVisual acuity prediction on real-life patient data using a machine learning based multistage system
Schlosser, T., Beuth, F., Meyer, T., Kumar, A. S., Stolze, G., Furashova, O., Engelmann, K., & Kowerko, D. (2024) · Scientific Reports · doi:10.1038/s41598-024-54482-2 · Cited by 11
“Moreover, Schmidt-Erfurth et al. recently reported the potential of AI-based approaches for targeted optimization of diagnosis and therapy for eye diseases 7 .”
BackgroundRobustness via Uncertainty-aware Cycle ConsistencyPreprint
Upadhyay, U., Chen, Y., & Akata, Z. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2110.12467 · Cited by 24
“Critical applications such as medical image-to-image translation [27, 28, 29, 30, 31, 32] is an excellent testbed for our model as confidence in the network’s predictions is desirable [33, 34] especially under the influence of missing imaging modalities.”
MethodBlattmann, M., Lindenmeyer, A., Franke, S., Neumuth, T., & Schneider, D. (2025) · PLOS Digital Health · doi:10.1371/journal.pdig.0000801 · Cited by 16
Wang, H., Guo, X., Song, K., Sun, M., Shao, Y., Xue, S., Zhang, T., & Zhang, T. (2024) · Neural Networks · doi:10.1016/j.neunet.2024.106817 · Cited by 10
Deep Learning Approaches for Detecting COVID-19 from Chest X-ray Images: A Survey (Preprint)Preprint
Alghamdi, H., Amoudi, G., Elhag, S., Saeedi, K., & Nasser, J. (2020) · doi:10.2196/preprints.26506 · Cited by 8
Jebril, H., Esengönül, M., & Bogunović, H. (2024) · Bioengineering · doi:10.3390/bioengineering11070682 · Cited by 11
“This approach was inspired by the work of [37], which focused on weak-label anomaly detection on structural OCT B-scans.”
MethodAnomaly Detection in Retinal OCT Images With Deep Learning-Based Knowledge Distillation
Aresta, G., Araújo, T., Schmidt-Erfurth, U. M., & Bogunović, H. (2025) · Translational Vision Science & Technology · doi:10.1167/tvst.14.3.26 · Cited by 9
“For instance, Seebock et al. 38 proposed to use an uncertainty-aware retinal layer segmentation model trained in normal data to identify potential anomalous OCT regions.”
Self-citationUnsupervised anomaly detection in MR images using multicontrast information
Kim, B., Kwon, K., Oh, C., & Park, H. (2021) · Medical Physics · doi:10.1002/mp.15269 · Cited by 9
Efficient Guided Grad-CAM Tuned Patch Neural Network for Accurate Anomaly Detection in Full Images
Rajkumar, R., Shanthi, D., & Manivannan, K. (2024) · Information Technology And Control · doi:10.5755/j01.itc.53.2.34525 · Cited by 9
“It has been functional for many tasks, with technologies becoming increasingly commonplace in the medical sector [32], [29], [5], [37], [33].”
BackgroundZhang, W., Wu, H., & Qin, J. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-73337-6_2 · Cited by 10
Feng, D., Chen, X., Wang, X., Lv, J., Bai, L., Zhang, S., & Zhou, Z. (2022) · Proceedings - IEEE Symposium on Computer-Based Medical Systems · doi:10.1109/cbms55023.2022.00061 · Cited by 8
“To establish the desirable model, we propose to integrate uncertainty estimation in training phase instead of only being used in testing phase for evaluation as in published studies [1-6].”
MethodDeep Reinforcement Learning for Detection of Inner Ear Abnormal Anatomy in Computed Tomography
Diez, P. L., Sørensen, K. A., Sundgaard, J. V., Диаб, Х. М., Margeta, J., Patou, F., & Paulsen, R. R. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-16437-8_67 · Cited by 7
Detecting Absence of Bone Wall in Jugular Bulb by Image Transformation Surrogate Tasks
Li, X., Zhou, Y., Yin, H., Wang, Z., Li, Z., & Zhang, H. (2021) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2021.3139917 · Cited by 6
Continuous image anomaly detection based on contrastive lifelong learning
Fan, W., Shangguan, W., & Bouguila, N. (2023) · Applied Intelligence · doi:10.1007/s10489-022-04401-7 · Cited by 10
Influence of uncertainty estimation techniques on false-positive reduction in liver lesion detection
Bhat, I., Pluim, J. P. W., Viergever, M. A., & Kuijf, H. J. (2022) · The Journal of Machine Learning for Biomedical Imaging · doi:10.59275/j.melba.2022-5937 · Cited by 8
“Seeb¨ock et al. (2020) used the uncertainty map to detect anomalous regions in retinal OCT images.”
BackgroundDeep Reinforcement Learning for Detection of Abnormal Anatomies
Diez, P. L., Juhl, K. A., Sundgaard, J. V., Diab, H., Margeta, J., Patou, F., & Paulsen, R. R. (2022) · Proceedings of the Northern Lights Deep Learning Workshop · doi:10.7557/18.6280 · Cited by 6
“While for detection of abnormal anatomies in image data, most studies employ more advanced deep learning methods due to the increased complexity of the data [1, 11].”
BackgroundHassan, T., Akram, M. U., & Werghi, N. (2020) · arXiv (Cornell University) · Cited by 6
Metadata
- Authors
- Seebock, P., Orlando, J. I., Schlegl, T., Waldstein, S. M., Bogunovic, H., Klimscha, S., Langs, G., & Schmidt-Erfurth, U.
- Affiliations
- Medical University of Vienna (AT)
- ORCID
- Venue
- IEEE Transactions on Medical Imaging
- Published
- 2020-01
- Volume
- 39
- Issue
- 1
- Pages
- 87-98
- ISSN
- 0278-0062 (ISSN-L)
- 1558-254X
- OpenAlex
- W2947805333
- Cited by
- 164
- References
- 56
- Citations by year
- 2026: 13 · 2025: 17 · 2024: 25 · 2023: 31 · 2022: 30 · 2021: 32 · 2020: 14
- FWCI
- 12.86
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Licence
- Locations
- IEEE Transactions on Medical Imaging (publishedVersion)
- PubMed (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- Christian Doppler Forschungsgesellschaft
- Nvidia
- National Foundation for Research, Technology, and Development
- Austrian Science Fund (FWF I2714-B31)
- Vienna Science and Technology Fund (FA7464A0249, VRG12-009)
- Topics
- Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Retinal Imaging and Analysis
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Medical Image Segmentation Techniques
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Keywords
- anomaly detection, retinal optical coherence tomography, epistemic uncertainty, Monte Carlo dropout, retinal layer segmentation, weak labels, age-related macular degeneration, geographic atrophy, diabetic macular edema, retinal vein occlusion, biomarker discovery
- MeSH terms
- Supervised Machine Learning, Algorithms, Humans, Image Processing, Computer-Assisted, Retina, Artifacts, Tomography, Optical Coherence
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.