Pith. sign in

REVIEW 5 major objections 5 minor 88 references

Feature Learning to Automatically Assess Radiographic Knee Osteoarthritis Severity

T0 review · 5 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper claims that convolutional neural networks can automate knee osteoarthritis grading from X-rays, with a jointly trained classifier and regressor reaching 64.6% multi-class accuracy and a mean-squared error of 0.480, on par with…

desk verdict A competent consolidation of prior knee-OA deep learning work with useful practical comparisons, but the central state-of-the-art claim is weakened by an omitted cited baseline and missing error bars. read the letter →

arxiv 1908.08840 v1 pith:JD7ANNZC submitted 2019-08-23 cs.CV cs.LG

Joseph Antony , Kevin McGuinness , Kieran Moran , Noel E O' Connor This is my paper · ORCID
classification cs.CVcs.LG
keywords kneeosteoarthritisKellgren-Lawrencegradingconvolutionalneuralnetworksordinalregressionmulti-objectivelearningX-rayimageclassificationfullyOAIdataset
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper argues that convolutional neural networks can take over the two jobs a radiologist does when grading knee osteoarthritis on X-rays: finding the knee joint and rating its severity on the Kellgren-Lawrence 0-4 scale. It claims that a lightweight CNN trained from scratch on public knee radiographs, optimizing a weighted mix of classification and regression losses, reaches 64.6% multi-class accuracy, and an ordinal-regression variant reaches 0.480 mean-squared error. These numbers beat both a handcrafted-feature classifier (WNDCHRM, 34.8% accuracy, MSE 2.112) and a fine-tuned large CNN (BVLC CaffeNet, 57.6%, MSE 0.836). The paper further claims the full pipeline, a fully convolutional network for localizing joints plus the trained CNN, performs on par with radiologic reliability readings, the accepted gold standard. If right, this means automatic KL grading on standard public datasets is no longer a proof of concept but a practical diagnostic support tool.

What carries the argument

The central mechanism is multi-objective convolutional learning: a single lightweight CNN of about 2.9 million parameters, built from blocks of cascaded 3x3 convolutions, batch normalization, and max pooling, ends in a shared fully connected layer that branches into a softmax head for the five KL classes and a linear head for a continuous severity value. The loss is a weighted sum of categorical cross-entropy and mean-squared error, with the ratio tuned to 0.5; cross-entropy preserves discrete grade information while MSE injects ordering and distance information between grades. The ordinal-regression variant replaces the regression branch with a dot product of the softmax probabilities against fixed weights [0,1,2,3,4], so the regression target is the expected grade under the class distribution, and this is the mechanism that produces the paper's lowest reported MSE.

What would settle it

Take a test set of OAI or MOST radiographs that also have OARSI sub-scores for joint space narrowing and osteophytes, and compare the ordinal-regression predictions against the measured sub-scores: if a model whose grade weights are proportional to measured joint space width achieves lower MSE than the fixed [0,1,2,3,4] model, the equidistant-spacing assumption behind the paper's continuous claim is false.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that feature learning with CNNs, rather than handcrafted image features, is the effective route to fine-grained knee OA quantification, and that the best quantification comes from training the network to predict KL grade as both a discrete class and a continuous value at once. The jointly trained CNN predicts a discrete grade through a softmax head and a continuous value through a linear regression head, with the loss a weighted sum of categorical cross-entropy and mean-squared error; it reaches 64.6% multi-class accuracy and MSE 0.507 on the regression head. A variant that treats the softmax head as a hidden layer and multiplies its probabilities by fixed weights [0,1,2,3,4], the ordinal-regression approach, reaches MSE 0.480 while keeping classification accuracy near 64.3%, which the paper reports as the best continuous-scale quantification. The claim is that this accuracy is on par with the reliability of expert radiologic readings, making the automatic system a credible replacement for manual KL scoring.

Load-bearing premise

The central claim assumes KL grades 0 through 4 are equally spaced points on a continuous severity scale; the paper itself notes they are not equidistant, so the reported continuous MSE depends on that spacing.

Editorial extensions

If this is right

  • An end-to-end pipeline using a fully convolutional network for knee localization plus the jointly trained CNN can produce KL grades and continuous severity scores directly from raw X-rays, removing manual ROI cropping.
  • The localizer alone detects knee joints at 100% accuracy for Jaccard index >= 0.5 over the combined OAI-MOST test set, so downstream grading need not depend on hand-drawn regions.
  • Training a CNN from scratch outperforms fine-tuning ImageNet-pretrained networks on this task (61.8% versus 57.6% accuracy), suggesting that domain-specific features learned on medical images outweigh transfer from natural images when enough data are available.
  • Joint training with two losses improves both class accuracy and regression MSE over training either head alone, making multi-objective convolutional learning a practical recipe for progressive-disease grading.
  • The system's accuracy is on par with radiologic reliability readings, implying automatic KL grading could serve as a second reader or initial screening tool in clinical workflows.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If KL grades are not equidistant, as the paper itself notes, the fixed weights [0,1,2,3,4] impose a spacing the data may not support; a natural extension is to learn those weights from joint-space-width or osteophyte measurements instead of fixing them.
  • The same weighted-loss recipe should transfer to other ordinal medical scales, such as Alzheimer's or cancer staging, where the disease is progressive but labels are discrete; the paper sketches this direction but does not test it.
  • Because the paper reports only MSE against integer KL labels, a stronger test of continuous quantification would compare predictions to OARSI sub-scores such as joint space narrowing and osteophytes; if those correlations are high, the continuous claim is independently supported.
  • The 64.6% accuracy may be partly dataset-specific: OAI and MOST share imaging protocols and reading standards, so a test on an external multi-center radiograph collection would reveal whether the learned features generalize.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 5 minor

Summary. This manuscript, written in thesis-chapter style, proposes and evaluates convolutional-neural-network methods for automatic knee osteoarthritis severity grading from X-ray images. The contributions include an FCN for knee joint localization, CNN classifiers and regressors trained from scratch, a jointly trained multi-objective network combining cross-entropy and MSE losses, and an ordinal regression variant with fixed grade weights. Experiments use the public OAI and MOST datasets; the final comparison (Table 43) reports that the jointly trained CNN reaches 64.6% multi-class accuracy and the ordinal regression reaches MSE 0.480, outperforming the WNDCHRM baseline and a fine-tuned BVLC CaffeNet.

Significance. If the results are reproducible and the comparisons are made on a common benchmark, the paper would provide a useful demonstration that multi-objective training and ordinal regression improve automatic KL grading over handcrafted-feature baselines. Strengths include the systematic progression from handcrafted features to transfer learning to from-scratch training, the use of two public datasets, the explicit error analysis, and the description of an end-to-end pipeline. The central claim of outperforming existing approaches is not currently established, however, because a published baseline (Tiulpin et al. [44]) cited in the paper itself reports higher accuracy, and because no variance or significance information is provided.

major comments (5)
  1. [Section 2.2.2 / Table 43] The paper's claim to 'outperform existing approaches' (abstract and Section 6) is not supported by the comparisons in Table 43. Section 2.2.2 cites Tiulpin et al. [44] with 66.7% average multi-class accuracy on the entire OAI dataset, which is higher than the 64.6% reported here for the jointly trained CNN. Table 43 only includes WNDCHRM and a fine-tuned BVLC CaffeNet, and no explicit comparison to [44] is provided. Please either run the proposed method and the Tiulpin et al. method on the same evaluation protocol, or revise the claims to state superiority only over the methods actually tested.
  2. [Section 5.2.6 / Table 43] The ordinal regression results are reported inconsistently. The text (Section 5.2.6, Results) states that after rounding the ordinal output, classification accuracy is 61.8% and MSE is 0.504, while the classification branch itself reaches 64.3%. Table 43 lists 'Ordinal Regression 64.3% 0.480', which does not match either quantity as described. Clarify which outputs and rounding procedures the table entries refer to, and correct the numbers in the abstract and conclusion if needed.
  3. [Section 5.2.4/5.2.5 (Tables 39-43)] The key accuracy and MSE differences (e.g., 61.8% vs 64.6%, and regression MSE 0.574 vs 0.507) are reported for single training runs, with no confidence intervals, repeated-seed statistics, or significance tests. Without such information, the improvement from joint training and ordinal regression cannot be distinguished from random variation. Please provide means and standard deviations over multiple runs or another appropriate statistical assessment.
  4. [Section 5.2.5 Discussion / 5.2.6] The continuous-scale quantification claim depends on treating KL grades 0-4 as equidistant integer targets (fixed weights [0,1,2,3,4] in the ordinal regression and integer labels in MSE). The paper itself notes that KL grades are not equidistant, citing references [22,81,25,24,82]. The reported MSE therefore measures deviation on this arbitrary integer scale, not on a validated continuous severity scale. Please either validate the scale (e.g., against quantitative joint-space-width measurements or OARSI grades) or clearly limit the continuous-scale claim to a modeling choice.
  5. [Section 5.3] The proposed end-to-end diagnostic system is described as combining the FCN localization and the jointly trained CNN, but no experiments evaluate the full pipeline. The results in Table 43 appear to be obtained from cropped knee images, and it is not stated whether these crops come from automatic localization or manual annotation (Section 5.2.4 describes both but Table 43 does not specify). The paper should report end-to-end performance with automatic localization, or explicitly state that the system was not evaluated as a whole.
minor comments (5)
  1. [Throughout (architectures)] There are inconsistent network descriptions; for example, Table 33 lists conv4-1 with 128 kernels but an output shape of 96, and several architecture tables use inconsistent naming conventions (e.g., 'Conv2 1' vs 'conv2-1').
  2. [Throughout] The manuscript uses 'WNDCHARM' and 'WNDCHRM' interchangeably (e.g., Section 5.1.1 vs Table 31); please standardize the spelling.
  3. [Section 2.2.1] There are typos such as 'purpotedly' and non-standard ligatures in 'off-the-shelf'; a careful proofread is needed.
  4. [Section 5.2.5] The opening of Section 5.2.5 refers to 'Section 5.4.5' instead of 'Section 5.2.5'.
  5. [Figure 51] Figure 51 is captioned 'The CNN configuration for ordinal regression' but appears to be a plot of test accuracy over epochs; the caption and the figure content should be matched.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the CNN and ordinal-regression results are empirical, evaluated on held-out public data; self-citations are contextual and not load-bearing.

full rationale

The paper's central claims are empirical comparisons of CNN training variants on fixed public OAI/MOST data splits. The jointly trained CNN and ordinal regression outputs are produced by standard forward passes, with no parameter fitted to the test labels; the loss weight 0.5 is selected by validation and the same splits are maintained for all compared methods. The 'continuous scale' output is explicitly acknowledged as a modeling limitation: no continuous ground-truth exists, discrete KL labels are used as regression targets, and the paper notes that KL categories are not equidistant (Section 5.2.5 Discussion). This is a stated assumption about the label space, not a circular derivation. The fixed weights [0,1,2,3,4] in ordinal regression are a deterministic design choice, and the lower MSE relative to rounded classification is partly a mathematical property of the mean; however, the paper's main comparisons against single-output regression and classification are empirical and not forced by construction. Self-citations [45,58,86,88] describe the authors' earlier pipelines or related work; none supplies an unverified uniqueness theorem or a fitted parameter that the present claims reduce to. The omission of Tiulpin's published 66.7% accuracy baseline from Table 43 is a comparison/completeness concern, not circularity. Hence there is no significant circularity.

Assumptions & free parameters 6 free parameters · 5 assumptions · 0 invented entities

The central claim depends on a handful of tuned hyperparameters, on the validity of KL grades as ground truth, and on the assumption that discrete grades can stand in for a continuous severity scale. The most consequential free parameter is the 0.5 weight balancing classification and regression losses; the paper selects it by searching 0.2 to 0.6 on validation data. No new physical entities are introduced.

free parameters (6)
  • loss_weight_lambda = 0.5
    Weight for the MSE term in the joint classification-regression loss; chosen by testing values 0.2 to 0.6 (Section 5.2.5, Initial Configuration).
  • l2_regularization_penalty = 0.01
    Applied to convolutional and fully connected layers in the best jointly trained and ordinal regression CNNs (Tables 38, 41); tuned by hand.
  • dropout_ratios = 0.25, 0.3, 0.5
    Dropout after convolutional and dense layers, varied across configurations (Sections 5.2.4, 5.2.5); chosen by validation performance.
  • input_image_size = 200x300
    Chosen to approximately preserve mean aspect ratio 1.6 of extracted knee joints (Section 5.2.4); affects all reported accuracies.
  • roi_crop_size = 640x560
    Fixed-size knee joint crop for center-based localization (Section 4.2.1); selected after visual testing.
  • ordinal_grade_weights = [0,1,2,3,4]
    Fixed weights applied to softmax probabilities for ordinal regression (Section 5.2.6); assumes equal spacing of KL grades, which the paper itself questions.
assumptions (5)
  • domain assumption KL grades are a valid ordinal ground truth for knee OA severity
    All training and evaluation uses KL grades from OAI/MOST as labels; the paper notes inter-rater variability and subjectivity (Sections 1, 5.2.5).
  • ad hoc to paper Discrete KL grades can be used as regression targets to represent a continuous severity scale
    Explicitly acknowledged: 'there is no ground truth i.e. KL grades on a continuous scale to train a network directly for regression output' (Section 5.2.4, Training CNNs for Regression).
  • domain assumption OAI and MOST images can be resized and combined without loss of diagnostic information
    Images are resized to fixed sizes (e.g., 256x256, 200x300) and datasets combined for training (Sections 4.2, 5.2.4).
  • domain assumption Manual annotations of knee joint centers and ROIs are accurate enough to serve as ground truth for FCN training
    A custom annotation tool was used to manually label 4,446 OAI and 2,920 MOST radiographs (Sections 4.2.1, 4.2.2).
  • domain assumption The FCN localization errors do not substantially affect downstream classification
    The paper reports a 3-4% accuracy decrease with automatic vs manual extraction (Section 4.2.1, Limitations).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Feature Learning to Automatically Assess Radiographic Knee Osteoarthritis Severity." pith.science (2026). https://pith.science/paper/JD7ANNZC

@misc{pith2026190808840,
  author       = {Pith},
  title        = {Pith review of: Feature Learning to Automatically Assess Radiographic Knee Osteoarthritis Severity},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JD7ANNZC}},
  note         = {Machine review of arXiv:1908.08840}
}
read the original abstract

This chapter presents the investigations and the results of feature learning using convolutional neural networks to automatically assess knee osteoarthritis (OA) severity and the associated clinical and diagnostic features of knee OA from X-ray images. Also, this chapter demonstrates that feature learning in a supervised manner is more effective than using conventional handcrafted features for automatic detection of knee joints and fine-grained knee OA image classification. In the general machine learning approach to automatically assess knee OA severity, the first step is to localize the region of interest that is to detect and extract the knee joint regions from the radiographs, and the next step is to classify the localized knee joints based on a radiographic classification scheme such as Kellgren and Lawrence grades. First, the existing approaches for detecting (or localizing) the knee joint regions based on handcrafted features are reviewed and outlined. Next, three new approaches are introduced: 1) to automatically detect the knee joint region using a fully convolutional network, 2) to automatically assess the radiographic knee OA using CNNs trained from scratch for classification and regression of knee joint images to predict KL grades in ordinal and continuous scales, and 3) to quantify the knee OA severity optimizing a weighted ratio of two loss functions: categorical cross entropy and mean-squared error using multi-objective convolutional learning and ordinal regression. Two public datasets: the OAI and the MOST are used to evaluate the approaches with promising results that outperform existing approaches. In summary, this work primarily contributes to the field of automated methods for localization (automatic detection) and quantification (image classification) of radiographic knee OA.

Figures

Figures reproduced from arXiv: 1908.08840 by the authors.

Figure 1
Figure 1. A healthy knee and a knee joint affected with OA. [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. The Kellgren and Lawrence grading system to assess the severity of knee [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Samples of bilateral PA fixed flexion knee OA radiographs. [PITH_FULL_IMAGE:figures/full_fig_p014_3.png] view at source ↗
Figures from the paper (49 more)
Figure 4
Figure 4. Figure 4: The OAI baseline data set distribution based on KL grades. [PITH_FULL_IMAGE:figures/full_fig_p015_4.png]
Figure 5
Figure 5. Figure 5: The MOST data set distribution based on KL grades. [PITH_FULL_IMAGE:figures/full_fig_p015_5.png]
Figure 6
Figure 6. Figure 6: A knee OA X-ray image with the region of interest: the knee joints. [PITH_FULL_IMAGE:figures/full_fig_p016_6.png]
Figure 7
Figure 7. Figure 7: Pre-selected knee joint centres (20×20 pixels) extracted from knee joint images for template matching [PITH_FULL_IMAGE:figures/full_fig_p017_7.png]
Figure 8
Figure 8. Figure 8: Detecting the knee joint centres and extracting the knee joints. [PITH_FULL_IMAGE:figures/full_fig_p019_8.png]
Figure 9
Figure 9. Figure 9: Automatic localisation of knee joints with reference to the centre of the [PITH_FULL_IMAGE:figures/full_fig_p021_9.png]
Figure 10
Figure 10. Figure 10: (a) An input X-ray image and (b) The binary mask annotations for knee [PITH_FULL_IMAGE:figures/full_fig_p022_10.png]
Figure 11
Figure 11. Figure 11: An instance of input, ground truth and output (predictions) of FCN. [PITH_FULL_IMAGE:figures/full_fig_p023_11.png]
Figure 12
Figure 12. Figure 12: An input image, ground truth, and outcome of the final FCN. [PITH_FULL_IMAGE:figures/full_fig_p024_12.png]
Figure 13
Figure 13. Figure 13: Prediction of the FCN with max pooling and up-sampling layers. [PITH_FULL_IMAGE:figures/full_fig_p025_13.png]
Figure 14
Figure 14. Figure 14: Predictions of the FCN with 3 Convolution-Pooling stages. [PITH_FULL_IMAGE:figures/full_fig_p026_14.png]
Figure 15
Figure 15. Figure 15: Training and validation losses of the FCN. [PITH_FULL_IMAGE:figures/full_fig_p028_15.png]
Figure 16
Figure 16. Figure 16: Error analysis: X-ray images, ground truth, FCN output - weak [PITH_FULL_IMAGE:figures/full_fig_p029_16.png]
Figure 17
Figure 17. Figure 17: Error analysis: X-ray images, ground truth, FCN output - detections [PITH_FULL_IMAGE:figures/full_fig_p030_17.png]
Figure 18
Figure 18. Figure 18: A knee X-ray image with the detected centres and the extracted left and [PITH_FULL_IMAGE:figures/full_fig_p030_18.png]
Figure 19
Figure 19. Figure 19: Anomalies in the automatic extraction of the ROI. [PITH_FULL_IMAGE:figures/full_fig_p031_19.png]
Figure 20
Figure 20. Figure 20: The actual ROI for the knee joints in Figure 19. [PITH_FULL_IMAGE:figures/full_fig_p031_20.png]
Figure 21
Figure 21. Figure 21: Variations in the aspect ratio of the extracted knee joints. [PITH_FULL_IMAGE:figures/full_fig_p032_21.png]
Figure 22
Figure 22. Figure 22: The actual ROI for the extracted knee joints in Figure 21. [PITH_FULL_IMAGE:figures/full_fig_p032_22.png]
Figure 23
Figure 23. Figure 23: Automatic localisation of the Region of Interest. [PITH_FULL_IMAGE:figures/full_fig_p033_23.png]
Figure 24
Figure 24. Figure 24: (a) An input X-ray image and (b) The binary mask annotations for the [PITH_FULL_IMAGE:figures/full_fig_p034_24.png]
Figure 25
Figure 25. Figure 25: Training and validation losses of the FCN. [PITH_FULL_IMAGE:figures/full_fig_p034_25.png]
Figure 26
Figure 26. Figure 26: An input X-ray image, ground truth and output prediction of the FCN. [PITH_FULL_IMAGE:figures/full_fig_p035_26.png]
Figure 27
Figure 27. Figure 27: Qualitative Evaluation: An input X-ray image, ground truth, and FCN [PITH_FULL_IMAGE:figures/full_fig_p036_27.png]
Figure 28
Figure 28. Figure 28: Qualitative Evaluation: An input X-ray image, ground truth, and FCN [PITH_FULL_IMAGE:figures/full_fig_p036_28.png]
Figure 29
Figure 29. Figure 29: Qualitative Evaluation: An input X-ray image, ground truth, and FCN [PITH_FULL_IMAGE:figures/full_fig_p037_29.png]
Figure 30
Figure 30. Figure 30: Error Analysis: An input X-ray image, ground truth, and FCN detec [PITH_FULL_IMAGE:figures/full_fig_p037_30.png]
Figure 31
Figure 31. Figure 31: Error Analysis: An input X-ray image, ground truth, and FCN detec [PITH_FULL_IMAGE:figures/full_fig_p038_31.png]
Figure 32
Figure 32. Figure 32: Error Analysis: An input X-ray image, ground truth, and FCN detec [PITH_FULL_IMAGE:figures/full_fig_p038_32.png]
Figure 33
Figure 33. Figure 33: Error Analysis: An input X-ray image, ground truth, and FCN detec [PITH_FULL_IMAGE:figures/full_fig_p039_33.png]
Figure 34
Figure 34. Figure 34: Error Analysis: An input X-ray image, ground truth, and FCN detec [PITH_FULL_IMAGE:figures/full_fig_p039_34.png]
Figure 35
Figure 35. Figure 35: Learning curves:training and validation losses (left), and validation [PITH_FULL_IMAGE:figures/full_fig_p046_35.png]
Figure 36
Figure 36. Figure 36: Learning curves: training and validation losses, and accuracies of the [PITH_FULL_IMAGE:figures/full_fig_p053_36.png]
Figure 37
Figure 37. Figure 37: Learning curves: training and validation losses, and accuracies of the [PITH_FULL_IMAGE:figures/full_fig_p055_37.png]
Figure 38
Figure 38. Figure 38: Learning curves: training and validation losses, and accuracies of the [PITH_FULL_IMAGE:figures/full_fig_p057_38.png]
Figure 39
Figure 39. Figure 39: Learning curves: training and validation losses, and accuracies of the [PITH_FULL_IMAGE:figures/full_fig_p058_39.png]
Figure 40
Figure 40. Figure 40: Learning curves: training and validation losses for the best performing [PITH_FULL_IMAGE:figures/full_fig_p062_40.png]
Figure 41
Figure 41. Figure 41: Initial configuration to jointly train a CNN for classification and [PITH_FULL_IMAGE:figures/full_fig_p064_41.png]
Figure 42
Figure 42. Figure 42: Learning curves for (a) classification and (b) regression in jointly trained [PITH_FULL_IMAGE:figures/full_fig_p065_42.png]
Figure 43
Figure 43. Figure 43: Learning curves for (a) classification and (b) regression in jointly trained [PITH_FULL_IMAGE:figures/full_fig_p068_43.png]
Figure 44
Figure 44. Figure 44: Confusion matrix for the multi-class classification using the jointly [PITH_FULL_IMAGE:figures/full_fig_p069_44.png]
Figure 45
Figure 45. Figure 45: ROC for the multi-class classification using the jointly trained CNN. [PITH_FULL_IMAGE:figures/full_fig_p070_45.png]
Figure 46
Figure 46. Figure 46: Mis-classifications: grade 1 joints predicted as grade 0, 2, and 3. [PITH_FULL_IMAGE:figures/full_fig_p070_46.png]
Figure 47
Figure 47. Figure 47: Misclassification: other grade knee joints predicted as grade 1. [PITH_FULL_IMAGE:figures/full_fig_p071_47.png]
Figure 48
Figure 48. Figure 48: An instance of more severe misclassification: grade 0 and grade 3. [PITH_FULL_IMAGE:figures/full_fig_p071_48.png]
Figure 49
Figure 49. Figure 49: The CNN configuration for ordinal regression. [PITH_FULL_IMAGE:figures/full_fig_p072_49.png]
Figure 50
Figure 50. Figure 50: Learning curves for (a) classification and (b) ordinal regression . [PITH_FULL_IMAGE:figures/full_fig_p074_50.png]
Figure 51
Figure 51. Figure 51: The CNN configuration for ordinal regression. [PITH_FULL_IMAGE:figures/full_fig_p075_51.png]
Figure 52
Figure 52. Figure 52: The proposed pipeline for quantifying knee OA severity. [PITH_FULL_IMAGE:figures/full_fig_p076_52.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

88 extracted references · 76 canonical work pages

  1. [44]

    Scientific reports 8(1) (2018) 1727

    Tiulpin, A., Thevenot, J., Rahtu, E., Lehenkari, P., Saarakkala, S.: Automatic knee osteoarthritis diagnosis from plain radiographs: a deep learning-based approach. Scientific reports 8(1) (2018) 1727

    work page 2018

  2. [1]

    IEEE Transactions on Biomedical Engineering 56(2) (2009)

    Shamir, L., Ling, S.M., Scott Jr, W.W., Bos, A., Orlov, N., Macura, T.J., Eckley, D.M., Ferrucci, L., Goldberg, I.G.: Knee X-ray image analysis method for automated detection of osteoarthritis. IEEE Transactions on Biomedical Engineering 56(2) (2009)

    2009

  3. [2]

    In: International Conference on Medical Image Computing and Computer-Assisted Intervention, Springer (2015) 127–134

    Thomson, J., ONeill, T., Felson, D., Cootes, T.: Automated shape and texture analysis for detection of osteoarthritis from radiographs of the knee. In: International Conference on Medical Image Computing and Computer-Assisted Intervention, Springer (2015) 127–134

    2015

  4. [3]

    Osteoarthritis and Cartilage 17(10) (2009) 1307–1312 Feature Learning to Assess Knee OA Severity 81

    Shamir, L., Ling, S.M., Scott, W., Hochberg, M., Ferrucci, L., Goldberg, I.G.: Early detection of radiographic knee osteoarthritis using computer-aided analysis. Osteoarthritis and Cartilage 17(10) (2009) 1307–1312 Feature Learning to Assess Knee OA Severity 81

    2009

  5. [4]

    Source code for biology and medicine 3(1) (2008) 13

    Shamir, L., Orlov, N., Eckley, D.M., Macura, T., Johnston, J., Goldberg, I.G.: Wndchrm–an open source utility for biological image analysis. Source code for biology and medicine 3(1) (2008) 13

    2008

  6. [5]

    Pattern recognition letters 29(11) (2008) 1684–1693

    Orlov, N., Shamir, L., Macura, T., Johnston, J., Eckley, D.M., Goldberg, I.G.: WND-CHARM: Multi-purpose image classification using compound image trans- forms. Pattern recognition letters 29(11) (2008) 1684–1693

    2008

  7. [6]

    Osteoarthritis and Cartilage 16(11) (2008) 1300–1306

    Oka, H., Muraki, S., Akune, T., Mabuchi, A., Suzuki, T., Yoshida, H., Yamamoto, S., Nakamura, K., Yoshimura, N., Kawaguchi, H.: Fully automatic quantification of knee osteoarthritis severity on plain radiographs. Osteoarthritis and Cartilage 16(11) (2008) 1300–1306

    2008

  8. [7]

    European journal of radiology 82(1) (2013) 112–117

    Park, H.J., Kim, S.S., Lee, S.Y., Park, N.H., Park, J.Y., Choi, Y.J., Jeon, H.J.: A practical MRI grading system for osteoarthritis of the knee: association with Kellgren–Lawrence radiographic scores. European journal of radiology 82(1) (2013) 112–117

    2013

Show all 88 references
  1. [8]

    PhD thesis, Stanford University (2010)

    Lee, H.: Unsupervised feature learning via sparse hierarchical representations. PhD thesis, Stanford University (2010)

    2010

  2. [9]

    PhD Thesis, Stanford University (2013)

    Le, Q.V.: Scalable feature learning. PhD Thesis, Stanford University (2013)

    2013

  3. [10]

    PhD Thesis, University of Washington (2013)

    Yang, S.: Feature Engineering in Fine-Grained Image Classification. PhD Thesis, University of Washington (2013)

    2013

  4. [11]

    PhD Thesis, Imperial College London (2013)

    Donoghue, C.R.: Analysis of MRI for Knee Osteoarthritis using Machine Learning. PhD Thesis, Imperial College London (2013)

    2013

  5. [12]

    Osteoarthritis and Cartilage 42(1) (2017) 60–88

    Litjens, G., Kooi, T., Bejnordi, B.E., Setio, A.A.A., Ciompi, F., Ghafoorian, M., van der Laak, J.A., van Ginneken, B., S´ anchez, C.I.: A survey on deep learning in medical image analysis. Osteoarthritis and Cartilage 42(1) (2017) 60–88

    2017

  6. [13]

    Annual Review of Biomedical Engineering (0) (2017)

    Shen, D., Wu, G., Suk, H.I.: Deep learning in medical image analysis. Annual Review of Biomedical Engineering (0) (2017)

    2017

  7. [14]

    In: Advances in neural information processing systems

    Krizhevsky, A., Sutskever, I., Hinton, G.E.: Imagenet classification with deep convolutional neural networks. In: Advances in neural information processing systems. (2012) 1097–1105

    2012

  8. [15]

    Tajbakhsh, N., Shin, J.Y., Gurudu, S.R., Hurst, R.T., Kendall, C.B., Gotway, M.B., Liang, J.: Convolutional neural networks for medical image analysis: full training or fine tuning? IEEE transactions on medical imaging 35(5) (2016) 1299– 1312

    2016

  9. [16]

    In: International conference on medical image computing and computer- assisted intervention, Springer (2013) 246–253

    Prasoon, A., Petersen, K., Igel, C., Lauze, F., Dam, E., Nielsen, M.: Deep feature learning for knee cartilage segmentation using a triplanar convolutional neural network. In: International conference on medical image computing and computer- assisted intervention, Springer (20...

    2013

  10. [17]

    Medical image analysis 35 (2017) 18–31

    Havaei, M., Davy, A., Warde-Farley, D., Biard, A., Courville, A., Bengio, Y., Pal, C., Jodoin, P.M., Larochelle, H.: Brain tumor segmentation with deep neural networks. Medical image analysis 35 (2017) 18–31

    2017

  11. [18]

    NeuroImage 108 (2015) 214–224

    Zhang, W., Li, R., Deng, H., Wang, L., Lin, W., Ji, S., Shen, D.: Deep convolutional neural networks for multi-modality isointense infant brain image segmentation. NeuroImage 108 (2015) 214–224

    2015

  12. [19]

    In: SPIE Medical Imaging, International Society for Optics and Photonics (2015) 94131G–94131G

    Roth, H.R., Farag, A., Lu, L., Turkbey, E.B., Summers, R.M.: Deep convolutional networks for pancreas segmentation in ct imaging. In: SPIE Medical Imaging, International Society for Optics and Photonics (2015) 94131G–94131G

    2015

  13. [20]

    In: Advances in neural information processing systems

    Ciresan, D., Giusti, A., Gambardella, L.M., Schmidhuber, J.: Deep neural networks segment neuronal membranes in electron microscopy images. In: Advances in neural information processing systems. (2012) 2843–2851 82 Joseph Antony, Kevin McGuinness, Kieran Moran, Noel E O’Connor

    2012

  14. [21]

    Osteoarthritis and Cartilage 16(2) (2008) 234–243

    Marijnissen, A.C., Vincken, K.L., Vos, P.A., Saris, D., Viergever, M., Bijlsma, J., Bartels, L., Lafeber, F.: Knee images digital analysis (kida): a novel method to quantify individual radiographic features of knee osteoarthritis in detail. Osteoarthritis and Cartilage 16(2) (...

    2008

  15. [22]

    Osteoarthritis and Cartilage 5(4) (1997) 241–250

    Felson, D.T., McAlindon, T.E., Anderson, J.J., Weissman, B.W., Aliabadi, P., Evans, S., Levy, D., LaValley, M.P.: Defining radiographic osteoarthritis for the whole knee. Osteoarthritis and Cartilage 5(4) (1997) 241–250

    1997

  16. [23]

    Bone 51(2) (2012) 278–288

    Braun, H.J., Gold, G.E.: Diagnosis of osteoarthritis: Imaging. Bone 51(2) (2012) 278–288

    2012

  17. [24]

    Osteoarthritis and Cartilage 16(8) (2008) 873–882

    Emrani, P.S., Katz, J.N., Kessler, C.L., Reichmann, W.M., Wright, E.A., McAlin- don, T.E., Losina, E.: Joint space narrowing and kellgren–lawrence progression in knee osteoarthritis: an analytic literature synthesis. Osteoarthritis and Cartilage 16(8) (2008) 873–882

    2008

  18. [25]

    Hart, D., Spector, T.: Kellgren & lawrence grade 1 osteophytes in the kneedoubtful or definite? Osteoarthritis and cartilage 11(2) (2003) 149–150

    2003

  19. [26]

    IOSR Journal of Electronics and Communication Engineering (IOSR-JECE) 9(5) (2014) 23–28

    Shaikh, H., Panbude, J., Joshi, A.: Image segmentation techniques and its appli- cations for knee joints: A survey. IOSR Journal of Electronics and Communication Engineering (IOSR-JECE) 9(5) (2014) 23–28

    2014

  20. [27]

    In: Biomedical and Pharmaceutical Engineering, 2006

    Sun, Y., Teo, E., Zhang, Q.: Discussions of knee joint segmentation. In: Biomedical and Pharmaceutical Engineering, 2006. ICBPE 2006. International Conference on, IEEE (2006)

    2006

  21. [28]

    International Journal of Computer Applications 145(1) (2016)

    Gornale, S.S., Patravali, P.U., Manza, R.R.: Detection of osteoarthritis using knee x-ray image analyses: A machine vision based approach. International Journal of Computer Applications 145(1) (2016)

    2016

  22. [29]

    In: Scandinavian Conference on Image Analysis, Springer (2017) 290–301

    Tiulpin, A., Thevenot, J., Rahtu, E., Saarakkala, S.: A novel method for automatic localization of joint area on knee plain radiographs. In: Scandinavian Conference on Image Analysis, Springer (2017) 290–301

    2017

  23. [30]

    Magnetic resonance imaging 17(7) (1999) 1033–1042

    Stammberger, T., Eckstein, F., Michaelis, M., Englmeier, K.H., Reiser, M.: Interobserver reproducibility of quantitative cartilage measurements: comparison of b-spline snakes and manual segmentation. Magnetic resonance imaging 17(7) (1999) 1033–1042

    1999

  24. [31]

    Osteoarthritis and cartilage 7(1) (1999) 95–109

    Cohen, Z.A., Mccarthy, D.M., Kwak, S.D., Legrand, P., Fogarasi, F., Ciaccio, E.J., Ateshian, G.A.: Knee cartilage topography, thickness, and contact areas from mri: in-vitro calibration and in-vivo measurements. Osteoarthritis and cartilage 7(1) (1999) 95–109

    1999

  25. [32]

    Osteoarthritis and cartilage 22(10) (2014) 1724–1731

    Hirvasniemi, J., Thevenot, J., Immonen, V., Liikavainio, T., Pulkkinen, P., J¨ ams¨ a, T., Arokoski, J., Saarakkala, S.: Quantification of differences in bone texture from plain radiographs in knees with and without osteoarthritis. Osteoarthritis and cartilage 22(10) (2014) 1724–1731

    2014

  26. [33]

    Medical physics 37(5) (2010) 2030–2042

    Woloszynski, T., Podsiadlo, P., Stachowiak, G., Kurzynski, M.: A signature dissimilarity measure for trabecular bone texture in knee radiographs. Medical physics 37(5) (2010) 2030–2042

    2010

  27. [34]

    Annals of the BMVA 2013(7) (2013) 1–22

    Zhao, F., Xie, X.: An overview of interactive medical image segmentation. Annals of the BMVA 2013(7) (2013) 1–22

    2013

  28. [35]

    PhD thesis, ETH Zurich (2005)

    Pirnog, C.D.: Articular cartilage segmentation and tracking in sequential MR images of the knee. PhD thesis, ETH Zurich (2005)

    2005

  29. [36]

    Medical physics 27(3) (2000) 580–591

    Duryea, J., Li, J., Peterfy, C., Gordon, C., Genant, H.: Trainable rule-based algorithm for the measurement of joint space width in digital radiographic images of the knee. Medical physics 27(3) (2000) 580–591

    2000

  30. [37]

    Medical physics 35(5) (2008) 1870–1883 Feature Learning to Assess Knee OA Severity 83

    Podsiadlo, P., Wolski, M., Stachowiak, G.: Automated selection of trabecular bone regions in knee radiographs. Medical physics 35(5) (2008) 1870–1883 Feature Learning to Assess Knee OA Severity 83

    2008

  31. [38]

    IPTEK The Journal for Technology and Science 22(3) (2011)

    Anifah, L., Purnama, I.K.E., Hariadi, M., Purnomo, M.H.: Automatic segmenta- tion of impaired joint space area for osteoarthritis knee on x-ray image using gabor filter based morphology process. IPTEK The Journal for Technology and Science 22(3) (2011)

    2011

  32. [39]

    In: Innova- tive Computing, Information and Control, 2006

    Lee, H.C., Lee, J.S., Lin, M.C.J., Wu, C.H., Sun, Y.N.: Automatic assessment of knee osteoarthritis parameters from two-dimensional x-ray image. In: Innova- tive Computing, Information and Control, 2006. ICICIC’06. First International Conference on. Volume 2., IEEE (2006) 673–676

    2006

  33. [40]

    Asian Journal of Scientific Research 6(4) (2013) 805

    Subramoniam, M., Rajini, V.: Local binary pattern approach to the classification of osteoarthritis in knee x-ray images. Asian Journal of Scientific Research 6(4) (2013) 805

    2013

  34. [41]

    Biomedical Research (2015)

    Subramoniam, B., et al.: A non-invasive computer aided diagnosis of osteoarthritis from digital x-ray images. Biomedical Research (2015)

    2015

  35. [42]

    International Journal of Engineering and Techniques 1(3) (2015)

    Deokar, D.D., Patil, C.G.: Effective feature extraction based automatic knee osteoarthritis detection and classification using neural network. International Journal of Engineering and Techniques 1(3) (2015)

    2015

  36. [43]

    PloS one 11(2) (2016) e0148724

    Yoo, T.K., Kim, D.W., Choi, S.B., Park, J.S.: Simple scoring system and artificial neural network for knee osteoarthritis risk prediction: A cross-sectional study. PloS one 11(2) (2016) e0148724

    2016

  37. [45]

    In: Pattern Recognition (ICPR), 2016 23rd International Conference on, IEEE (2016) 1195– 1200

    Antony, J., McGuinness, K., O’Connor, N.E., Moran, K.: Quantifying radiographic knee osteoarthritis severity using deep convolutional neural networks. In: Pattern Recognition (ICPR), 2016 23rd International Conference on, IEEE (2016) 1195– 1200

    2016

  38. [46]

    Machine Vision for three- demensional Sciences (1990)

    Sobel, I.: An isotropic 3 × 3 image gradient operator. Machine Vision for three- demensional Sciences (1990)

    1990

  39. [47]

    In: Computer Vision and Pattern Recognition, 2005

    Dalal, N., Triggs, B.: Histograms of oriented gradients for human detection. In: Computer Vision and Pattern Recognition, 2005. CVPR 2005. IEEE Computer Society Conference on. Volume 1., IEEE (2005) 886–893

    2005

  40. [48]

    In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition

    Long, J., Shelhamer, E., Darrell, T.: Fully convolutional networks for semantic segmentation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. (2015) 3431–3440

    2015

  41. [49]

    In: 3D Vision (3DV), 2016 Fourth International Conference on, IEEE (2016) 565–571

    Milletari, F., Navab, N., Ahmadi, S.A.: V-net: Fully convolutional neural networks for volumetric medical image segmentation. In: 3D Vision (3DV), 2016 Fourth International Conference on, IEEE (2016) 565–571

    2016

  42. [50]

    arXiv preprint arXiv:1701.03056 (2017)

    Kayalibay, B., Jensen, G., van der Smagt, P.: Cnn-based segmentation of medical imaging data. arXiv preprint arXiv:1701.03056 (2017)

    2017 arXiv

  43. [51]

    arXiv preprint arXiv:1702.05970 (2017)

    Christ, P.F., Ettlinger, F., Gr¨ un, F., Elshaera, M.E.A., Lipkova, J., Schlecht, S., Ahmaddy, F., Tatavarty, S., Bickel, M., Bilic, P., et al.: Automatic liver and tumor segmentation of ct and mri volumes using cascaded fully convolutional neural networks. arXiv preprint arXi...

    2017 arXiv

  44. [52]

    (2016) Available at http://cs231n.github.io/ convolutional-networks/

    Li, F.F., Karpathy, A., Johnson, J.: CS231n: Convolutional Neural Net- works for Visual Recognition. (2016) Available at http://cs231n.github.io/ convolutional-networks/

    2016

  45. [53]

    arXiv preprint arXiv:1409.1556 (2014)

    Simonyan, K., Zisserman, A.: Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556 (2014)

    2014 arXiv

  46. [54]

    OpenCV library documentation (2001) 84 Joseph Antony, Kevin McGuinness, Kieran Moran, Noel E O’Connor

    Kilian, J.: Simple image analysis by moments. OpenCV library documentation (2001) 84 Joseph Antony, Kevin McGuinness, Kieran Moran, Noel E O’Connor

    2001

  47. [55]

    In: Color Imaging: Processing, Hardcopy, and Applications

    Bressan, M., Dance, C.R., Poirier, H., Arregui, D.: Local contrast enhancement. In: Color Imaging: Processing, Hardcopy, and Applications. (2007) 64930Y

    2007

  48. [56]

    Computer vision, graphics, and image processing 39(3) (1987) 355–368

    Pizer, S.M., Amburn, E.P., Austin, J.D., Cromartie, R., Geselowitz, A., Greer, T., ter Haar Romeny, B., Zimmerman, J.B., Zuiderveld, K.: Adaptive histogram equalization and its variations. Computer vision, graphics, and image processing 39(3) (1987) 355–368

    1987

  49. [57]

    In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition

    Erhan, D., Szegedy, C., Toshev, A., Anguelov, D.: Scalable object detection using deep neural networks. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. (2014) 2147–2154

    2014

  50. [58]

    In: International Conference on Machine Learning and Data Mining in Pattern Recognition, Springer (2017) 376–390

    Antony, J., McGuinness, K., Moran, K., OConnor, N.E.: Automatic detection of knee joints and quantification of knee osteoarthritis severity using convolutional neural networks. In: International Conference on Machine Learning and Data Mining in Pattern Recognition, Springer (20...

    2017

  51. [59]

    IEEE Transactions on pattern analysis and machine intelligence 24(7) (2002) 971–987

    Ojala, T., Pietikainen, M., Maenpaa, T.: Multiresolution gray-scale and rotation invariant texture classification with local binary patterns. IEEE Transactions on pattern analysis and machine intelligence 24(7) (2002) 971–987

    2002

  52. [60]

    Astrophysics Source Code Library (2013)

    Shamir, L., Orlov, N., Eckley, D.M., Macura, T., Johnston, J., Goldberg, I.: Wnd- charm: Multi-purpose image classifier. Astrophysics Source Code Library (2013)

    2013

  53. [61]

    IEEE Transactions on Systems, Man, and Cybernetics 8(6) (1978) 460–473

    Tamura, H., Mori, S., Yamawaki, T.: Textural features corresponding to visual perception. IEEE Transactions on Systems, Man, and Cybernetics 8(6) (1978) 460–473

    1978

  54. [62]

    IEEE Transactions on systems, man, and cybernetics 3(6) (1973) 610–621

    Haralick, R.M., Shanmugam, K., et al.: Textural features for image classification. IEEE Transactions on systems, man, and cybernetics 3(6) (1973) 610–621

    1973

  55. [63]

    IEEE Transactions on Image processing 11(10) (2002) 1160–1167

    Grigorescu, S.E., Petkov, N., Kruizinga, P.: Comparison of texture features based on gabor filters. IEEE Transactions on Image processing 11(10) (2002) 1160–1167

    2002

  56. [64]

    JOSA 70(8) (1980) 920–930

    Teague, M.R.: Image analysis via the general theory of moments. JOSA 70(8) (1980) 920–930

    1980

  57. [65]

    In: Proceedings of British Machine Vision Conference

    Chatfield, K., Simonyan, K., Vedaldi, A., Zisserman, A.: Return of the devil in the details: Delving deep into convolutional nets. In: Proceedings of British Machine Vision Conference. (2014)

    2014

  58. [66]

    In: Proceedings of the ACM International Conference on Multimedia

    Jia, Y., Shelhamer, E., Donahue, J., Karayev, S., Long, J., Girshick, R., Guadar- rama, S., Darrell, T.: Caffe: Convolutional architecture for fast feature embedding. In: Proceedings of the ACM International Conference on Multimedia. (2014) 675– 678

    2014

  59. [67]

    In: Proceedings of the British Machine Vision Conference, BMVA Press (2014)

    Karayev, S., Trentacoste, M., Han, H., Agarwala, A., Darrell, T., Hertzmann, A., Winnemoeller, H.: Recognizing image style. In: Proceedings of the British Machine Vision Conference, BMVA Press (2014)

    2014

  60. [68]

    International Journal of Computer Vision 115(3) (2015) 211–252

    Russakovsky, O., Deng, J., Su, H., Krause, J., Satheesh, S., Ma, S., Huang, Z., Karpathy, A., Khosla, A., Bernstein, M., et al.: ImageNet large scale visual recognition challenge. International Journal of Computer Vision 115(3) (2015) 211–252

    2015

  61. [69]

    (2014) 3320–3328

    Yosinski, J., Clune, J., Bengio, Y., Lipson, H.: How transferable are features in deep neural networks? In: Advances in Neural Information Processing Systems. (2014) 3320–3328

    2014

  62. [70]

    In: Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics

    Glorot, X., Bengio, Y.: Understanding the difficulty of training deep feedforward neural networks. In: Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics. (2010) 249–256

    2010

  63. [71]

    Journal of machine learning research 9 (2008) 1871–1874 Feature Learning to Assess Knee OA Severity 85

    Fan, R.E., Chang, K.W., Hsieh, C.J., Wang, X.R., Lin, C.J.: LIBLINEAR: A library for large linear classification. Journal of machine learning research 9 (2008) 1871–1874 Feature Learning to Assess Knee OA Severity 85

    2008

  64. [72]

    In: Biomedical Imaging (ISBI), 2016 IEEE 13th International Symposium on, IEEE (2016) 1430–1434

    Miao, S., Wang, Z.J., Zheng, Y., Liao, R.: Real-time 2d/3d registration via cnn regression. In: Biomedical Imaging (ISBI), 2016 IEEE 13th International Symposium on, IEEE (2016) 1430–1434

    2016

  65. [73]

    Medical image analysis 36 (2017) 41–51

    Spampinato, C., Palazzo, S., Giordano, D., Aldinucci, M., Leonardi, R.: Deep learning for automated skeletal bone age assessment in x-ray images. Medical image analysis 36 (2017) 41–51

    2017

  66. [74]

    In: Proceedings Volume 9785, Medical Imaging 2016: Computer-Aided Diagnosis, SPIE Medical Imaging (2016)

    Roth, H.R., Wang, Y., Yao, J., Lu, L., Burns, J.E., Summers, R.M.: Deep convolutional networks for automated detection of posterior-element fractures on spine ct. In: Proceedings Volume 9785, Medical Imaging 2016: Computer-Aided Diagnosis, SPIE Medical Imaging (2016)

    2016

  67. [75]

    In: CVPR09

    Deng, J., Dong, W., Socher, R., Li, L.J., Li, K., Fei-Fei, L.: ImageNet: A Large- Scale Hierarchical Image Database. In: CVPR09. (2009)

    2009

  68. [76]

    In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition

    Liu, S., Yang, J., Huang, C., Yang, M.H.: Multi-objective convolutional learning for face labeling. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. (2015) 3451–3459

    2015

  69. [77]

    In: Advances in neural information processing systems

    Tompson, J.J., Jain, A., LeCun, Y., Bregler, C.: Joint training of a convolutional network and a graphical model for human pose estimation. In: Advances in neural information processing systems. (2014) 1799–1807

    2014

  70. [78]

    In: German Conference on Pattern Recognition, Springer (2014) 107–118

    Ranftl, R., Pock, T.: A deep variational model for image segmentation. In: German Conference on Pattern Recognition, Springer (2014) 107–118

    2014

  71. [79]

    IEEE Transactions on Image Processing 14(9) (2005) 1360–1371

    Ning, F., Delhomme, D., LeCun, Y., Piano, F., Bottou, L., Barbano, P.E.: Toward automatic phenotyping of developing embryos from videos. IEEE Transactions on Image Processing 14(9) (2005) 1360–1371

    2005

  72. [80]

    In: European Conference on Computer Vision, Springer (2016) 19–35

    Rudd, E.M., G¨ unther, M., Boult, T.E.: Moon: A mixed objective optimization network for the recognition of facial attributes. In: European Conference on Computer Vision, Springer (2016) 19–35

    2016

  73. [81]

    Annals of the rheumatic diseases 67(7) (2008) 1034–1036

    Schiphof, D., Boers, M., Bierma-Zeinstra, S.M.: Differences in descriptions of kellgren and lawrence grades of knee osteoarthritis. Annals of the rheumatic diseases 67(7) (2008) 1034–1036

    2008

  74. [82]

    Journal of Musculoskeletal Research 13(04) (2010) 197–201

    Shamir, L., Felson, D.T., Ferrucci, L., Goldberg, I.G.: Assessment of osteoarthritis initiative–kellgren and lawrence scoring projects quality using computer analysis. Journal of Musculoskeletal Research 13(04) (2010) 197–201

    2010

  75. [83]

    IEEE Transactions on Knowledge and Data Engineering 28(1) (2016) 127–146

    Guti´ errez, P.A., Perez-Ortiz, M., Sanchez-Monedero, J., Fernandez-Navarro, F., Hervas-Martinez, C.: Ordinal regression methods: survey and experimental study. IEEE Transactions on Knowledge and Data Engineering 28(1) (2016) 127–146

    2016

  76. [84]

    Journal of Machine Learning Research 18(55) (2017) 1–35

    Pedregosa, F., Bach, F., Gramfort, A.: On the consistency of ordinal regression methods. Journal of Machine Learning Research 18(55) (2017) 1–35

    2017

  77. [85]

    arXiv preprint arXiv:1612.00775 (2016)

    Beckham, C., Pal, C.: A simple squared-error reformulation for ordinal classifica- tion. arXiv preprint arXiv:1612.00775 (2016)

    2016 arXiv

  78. [86]

    In: International Conference on Medical Imaging with Deep Learning

    G´ orriz, M., Antony, J., McGuinness, K., Gir´ o-i Nieto, X., OConnor, N.E.: Assessing knee oa severity with cnn attention-based end-to-end architectures. In: International Conference on Medical Imaging with Deep Learning. (2019) 197–214

    2019

  79. [87]

    Arthritis Care & Research 51(4) (2004) 558–561

    Williams, D.A., Farrell, M.J., Cunningham, J., Gracely, R.H., Ambrose, K., Cupps, T., Mohan, N., Clauw, D.J.: Knee pain and radiographic osteoarthritis interact in the prediction of levels of self-reported disability. Arthritis Care & Research 51(4) (2004) 558–561

    2004

  80. [88]

    Scientific reports 9(1) (2019) 5761

    Abedin, J., Antony, J., McGuinness, K., Moran, K., OConnor, N.E., Rebholz- Schuhmann, D., Newell, J.: Predicting knee osteoarthritis severity: comparative modeling based on patients data and plain x-ray images. Scientific reports 9(1) (2019) 5761

    2019

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.