Matches in SemOpenAlex for { <https://semopenalex.org/work/W3135252693> ?p ?o ?g. }
- W3135252693 abstract "Deep neural networks are known to be data-driven and label noise can have a marked impact on model performance. Recent studies have shown great robustness to classic image recognition even under a high noisy rate. In medical applications, learning from datasets with label noise is more challenging since medical imaging datasets tend to have asymmetric (class-dependent) noise and suffer from high observer variability. In this paper, we systematically discuss and define the two common types of label noise in medical images - disagreement label noise from inconsistency expert opinions and single-target label noise from wrong diagnosis record. We then propose an uncertainty estimation-based framework to handle these two label noise amid the medical image classification task. We design a dual-uncertainty estimation approach to measure the disagreement label noise and single-target label noise via Direct Uncertainty Prediction and Monte-Carlo-Dropout. A boosting-based curriculum training procedure is later introduced for robust learning. We demonstrate the effectiveness of our method by conducting extensive experiments on three different diseases: skin lesions, prostate cancer, and retinal diseases. We also release a large re-engineered database that consists of annotations from more than ten ophthalmologists with an unbiased golden standard dataset for evaluation and benchmarking." @default.
- W3135252693 created "2021-03-15" @default.
- W3135252693 creator A5003662421 @default.
- W3135252693 creator A5005014252 @default.
- W3135252693 creator A5006822602 @default.
- W3135252693 creator A5017469047 @default.
- W3135252693 creator A5021790939 @default.
- W3135252693 creator A5053275663 @default.
- W3135252693 creator A5065250332 @default.
- W3135252693 creator A5078933997 @default.
- W3135252693 creator A5091299040 @default.
- W3135252693 date "2021-02-28" @default.
- W3135252693 modified "2023-10-01" @default.
- W3135252693 title "Improving Medical Image Classification with Label Noise Using Dual-uncertainty Estimation" @default.
- W3135252693 cites W1866072925 @default.
- W3135252693 cites W1979773093 @default.
- W3135252693 cites W2041184937 @default.
- W3135252693 cites W2103868202 @default.
- W3135252693 cites W2133012565 @default.
- W3135252693 cites W2145073242 @default.
- W3135252693 cites W2165835468 @default.
- W3135252693 cites W2168396085 @default.
- W3135252693 cites W2194775991 @default.
- W3135252693 cites W2296073425 @default.
- W3135252693 cites W2543927648 @default.
- W3135252693 cites W2600383743 @default.
- W3135252693 cites W2604156156 @default.
- W3135252693 cites W2739943470 @default.
- W3135252693 cites W2752971446 @default.
- W3135252693 cites W2765407302 @default.
- W3135252693 cites W2795282075 @default.
- W3135252693 cites W2806389172 @default.
- W3135252693 cites W2887842788 @default.
- W3135252693 cites W2891018693 @default.
- W3135252693 cites W2912934043 @default.
- W3135252693 cites W2945634539 @default.
- W3135252693 cites W2949226441 @default.
- W3135252693 cites W2950300355 @default.
- W3135252693 cites W2963081269 @default.
- W3135252693 cites W2963351448 @default.
- W3135252693 cites W2963466845 @default.
- W3135252693 cites W2963735582 @default.
- W3135252693 cites W2963759070 @default.
- W3135252693 cites W2964059111 @default.
- W3135252693 cites W2971609010 @default.
- W3135252693 cites W2978426779 @default.
- W3135252693 cites W2979302305 @default.
- W3135252693 cites W2979431884 @default.
- W3135252693 cites W2980265083 @default.
- W3135252693 cites W2981873476 @default.
- W3135252693 cites W2986198551 @default.
- W3135252693 cites W2988966271 @default.
- W3135252693 cites W2996988779 @default.
- W3135252693 cites W2999199619 @default.
- W3135252693 cites W3005731330 @default.
- W3135252693 cites W3028394061 @default.
- W3135252693 cites W3035325670 @default.
- W3135252693 cites W3036586801 @default.
- W3135252693 cites W3042609801 @default.
- W3135252693 cites W3101156210 @default.
- W3135252693 cites W3122459568 @default.
- W3135252693 cites W3138166738 @default.
- W3135252693 doi "https://doi.org/10.48550/arxiv.2103.00528" @default.
- W3135252693 hasPublicationYear "2021" @default.
- W3135252693 type Work @default.
- W3135252693 sameAs 3135252693 @default.
- W3135252693 citedByCount "6" @default.
- W3135252693 countsByYear W31352526932021 @default.
- W3135252693 crossrefType "posted-content" @default.
- W3135252693 hasAuthorship W3135252693A5003662421 @default.
- W3135252693 hasAuthorship W3135252693A5005014252 @default.
- W3135252693 hasAuthorship W3135252693A5006822602 @default.
- W3135252693 hasAuthorship W3135252693A5017469047 @default.
- W3135252693 hasAuthorship W3135252693A5021790939 @default.
- W3135252693 hasAuthorship W3135252693A5053275663 @default.
- W3135252693 hasAuthorship W3135252693A5065250332 @default.
- W3135252693 hasAuthorship W3135252693A5078933997 @default.
- W3135252693 hasAuthorship W3135252693A5091299040 @default.
- W3135252693 hasBestOaLocation W31352526931 @default.
- W3135252693 hasConcept C104317684 @default.
- W3135252693 hasConcept C115961682 @default.
- W3135252693 hasConcept C119857082 @default.
- W3135252693 hasConcept C124101348 @default.
- W3135252693 hasConcept C144133560 @default.
- W3135252693 hasConcept C153180895 @default.
- W3135252693 hasConcept C154945302 @default.
- W3135252693 hasConcept C162853370 @default.
- W3135252693 hasConcept C163294075 @default.
- W3135252693 hasConcept C185592680 @default.
- W3135252693 hasConcept C2776145597 @default.
- W3135252693 hasConcept C29265498 @default.
- W3135252693 hasConcept C35772409 @default.
- W3135252693 hasConcept C41008148 @default.
- W3135252693 hasConcept C46686674 @default.
- W3135252693 hasConcept C55493867 @default.
- W3135252693 hasConcept C63479239 @default.
- W3135252693 hasConcept C86251818 @default.
- W3135252693 hasConcept C99498987 @default.
- W3135252693 hasConceptScore W3135252693C104317684 @default.
- W3135252693 hasConceptScore W3135252693C115961682 @default.