Pith. sign in

REVIEW 3 major objections 7 minor 49 references

Knee menisci segmentation and relaxometry of 3D ultrashort echo time (UTE) cones MR imaging using attention U-Net with transfer learning

T0 review · 3 major / 7 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 A 2D attention U-Net with transfer learning can outline knee menisci on 3D UTE cones MR images at the same level as two human radiologists, and the automatic outlines yield T1, T1ρ, and T2* values statistically indistinguishable from…

desk verdict A useful, credible feasibility study for automatic menisci segmentation and relaxometry in 3D UTE cones MRI, but the 'equivalence' claim is statistically overconfident and should be tightened before publication. read the letter →

arxiv 1908.01594 v1 pith:QTW4GHLI submitted 2019-08-05 eess.IV cs.CVphysics.med-ph

Michal Byra , Mei Wu , Xiaodong Zhang , Hyungseok Jang , Ya-Jun Ma , Eric Y Chang , Sameer Shah , Jiang Du This is my paper · ORCID
classification eess.IVcs.CVphysics.med-ph
keywords kneemeniscussegmentationultrashortechotimeMRIUTEconesattentionU-NettransferlearningT1rhorelaxometryT2*osteoarthritisbiomarkers
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

The paper is trying to establish that a fully automatic deep-learning pipeline can replace hand-drawn regions of interest for quantitative knee meniscus MRI on 3D ultrashort echo time (UTE) cones images. It trains two 2D attention U-Nets, each on the outlines of one of two radiologists, using transfer learning from a network pretrained on natural images. On held-out subjects the automatic segmentations reached Dice scores of 0.860 and 0.833, slightly above the 0.820 Dice score between the two radiologists themselves. The T1, T1ρ, and T2* averages computed from automatic and manual outlines showed no statistically significant differences and correlated at 0.90 to 0.97. If correct, this makes meniscal relaxometry feasible without time-consuming manual tracing, supporting osteoarthritis assessment.

What carries the argument

The central object is a 2D attention U-Net, a U-Net whose skip connections are gated by attention layers so that encoder features far from the initial meniscus localization are suppressed. The first two encoder blocks are initialized from the corresponding blocks of a deep network pretrained on a large natural-image dataset, and a trainable 1×1 convolution block in front converts the normalized grayscale subtracted UTE images into the three-channel input format that pretrained network expects. Training uses a Dice-score loss directly, which suits the menisci's small pixel fraction. The resulting segmentations are fed into nonlinear least-squares fitting of the UTE cones data to compute T1, T1ρ, and T2* maps.

What would settle it

Build a consensus ground truth for the same or new UTE cones volumes from a panel of several radiologists (or from higher-resolution histology where available), then compute Dice scores and relaxometry differences for the two CNNs against that consensus. If the CNNs' Dice against consensus falls well below the radiologists' average Dice against consensus, or if the automatic T1, T1ρ, or T2* values differ from the consensus-based values by more than the 1.95%, 2.56-3.03%, and 4.59-6.15% average relative errors reported here, the equivalence claim would fail.

Watch

Extended reading notes

Core claim

The central claim is that a 2D attention U-Net with transfer learning, trained on radiologist-outlined regions of interest in subtracted AdiabT1ρ-weighted UTE cones slices, segments the knee menisci at a level equivalent to inter-observer variability. The models achieved Dice scores of 0.860 and 0.833 against the radiologist whose outlines trained them, while the two radiologists' manual segmentations agreed with each other at 0.820. The average T1, T1ρ, and T2* relaxation times computed from the automatic segmentations were not statistically different from those computed from manual outlines, with Pearson correlations between 0.90 and 0.97 and average relative errors of roughly 2% for T1, 3% for T1ρ, and 5-6% for T2*. The two CNNs also agreed with each other more strongly than the radiologists did (Dice 0.882 versus 0.820), which the paper reads as evidence that the learned segmentations are driven by image content rather than by annotator idiosyncrasy.

Load-bearing premise

The load-bearing premise is that the two radiologists' manually drawn outlines, made on subjectively selected subtracted images, are the correct meniscus boundary; because those outlines are both the training labels and the evaluation standard, any systematic error in them is learned and reproduced by the network.

Editorial extensions

If this is right

  • Automatic segmentations can replace manual tracing for meniscal T1, T1ρ, and T2* relaxometry on similar 3T UTE cones scans without statistically changing the average values.
  • The segmentation accuracy is at or above radiologist inter-observer level: a model trained on one radiologist's outlines agrees with the other radiologist about as well as the two radiologists agree with each other.
  • The CNNs produced no false positives on slices without menisci, supporting use of the model in fully automated slice-filtering and quantification pipelines.
  • The method could enable efficient extraction of osteoarthritis-relevant biomarkers, including relaxation times and ROI areas, from whole-knee UTE acquisitions.
  • Because the two CNNs agreed with each other more strongly than the two radiologists did, automatic segmentations may reduce inter-observer variability in meniscal measurements.

Reading between the lines

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

  • A further step, not tested here, would be to evaluate the same models against a consensus ground truth from many annotators; since the two CNNs agree more than the radiologists do, the models may be averaging out annotator noise, but any systematic boundary bias shared by the radiologists would still be baked into both the labels and the evaluation.
  • The error pattern—smallest for T1 and largest for T2*—suggests T2* relaxometry is the most sensitive to boundary disagreements; an OA study using T2* as its primary biomarker may need tighter segmentation agreement than this pipeline currently guarantees on atypical slices.
  • Because the method is trained on 2D slices, through-plane meniscus curvature is ignored; a likely extension, consistent with earlier work the paper cites, is to add a 3D refinement network, and a testable prediction is that this would push held-out Dice further above the radiologist baseline.
  • The same automatic masks could also be used for morphometric measures such as meniscal volume or extrusion on UTE data, an extension the paper mentions as future whole-joint work but does not validate here.
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

3 major / 7 minor

Summary. This paper presents a transfer-learning-based 2D attention U-Net for segmenting knee menisci in 3D UTE cones MR images and for computing T1, T1ρ, and T2* from the resulting masks. Sixty-one subjects were imaged; two radiologists independently outlined menisci on subtracted AdiabT1ρ-weighted images; the data were split into training/validation/test sets (36/10/15), with 191 test slices from 15 menisci. The authors report Dice scores of 0.860 and 0.833 for the two CNNs versus radiologist inter-observer Dice of 0.820, high correlations (0.90-0.97) between manual and automatic relaxometry values, and no statistically significant differences in average T1, T1ρ, and T2* between manual and automatic segmentations. They conclude that the automatic segmentations are equivalent to the radiologists' inter-observer variability and can replace manual ROIs for meniscal relaxometry.

Significance. If the claims hold, this would be a useful practical result: it is, to the authors' knowledge, the first demonstration of fully automatic meniscal segmentation and relaxometry on 3D UTE cones data. The study has real strengths: a held-out test set, validation-based model selection, two independent radiologist reference standards, separate reporting for medial and lateral menisci, AUC and false-positive checks, and Bland-Altman plots. The segmentation performance (Dice around 0.83-0.86) is in line with prior knee menisci segmentation work. However, the load-bearing statistical claims of equivalence in both segmentation and relaxometry are not supported by the analyses as reported.

major comments (3)
  1. [Methods, 'Model development and performance evaluation'; Results, Table 2] The claim of 'no associated differences' in T1, T1ρ, and T2* rests on t-tests applied to 191 2D images derived from only 15 test menisci. Adjacent 2D slices from the same meniscus are spatially correlated, so the effective sample size is at most 15, not 191. With n=15, the tests are underpowered, and a non-significant p-value does not establish equivalence. The manuscript should report per-meniscus or per-subject averages and use an equivalence test (e.g., TOST with a pre-specified margin) or a linear mixed-effects model that accounts for within-meniscus correlation. Without this, the abstract's statement that there were 'no associated differences' is not warranted.
  2. [Results, Table 1; Discussion] The segmentation-equivalence claim ('equivalent to the inter-observer variability of two radiologists') is based on descriptive comparison of average Dice scores (0.860 and 0.833 versus 0.820). No formal non-inferiority or equivalence test is reported for the CNN-versus-radiologist agreement relative to the radiologist inter-observer agreement. The only significance test mentioned for Dice (CNN1 vs CNN2 0.882 versus Rad1 vs Rad2 0.820, p<0.001) is not described in the Methods and appears to treat per-slice Dice values as independent. A formal test, or a clearly worded claim of 'comparable' rather than 'equivalent', is needed.
  3. [Methods, 'UTE imaging and data collection'; Discussion] The reference standard is based on ROIs drawn on subtracted AdiabT1ρ-weighted images selected 'based on subjective assessment', and the Discussion acknowledges the selection was 'strictly subjective'. Because the CNNs are trained to reproduce these specific ROIs, systematic errors in the manual outlines or in the choice of subtraction image are propagated into the automatic segmentations and the relaxometry comparison. The authors should assess the sensitivity of their conclusions to the choice of subtraction image, or at minimum discuss how this limitation bears on the claim that the method can 'replace' manual ROIs.
minor comments (7)
  1. [Abstract and Conclusions] The phrase 'no associated differences' should be rephrased to 'no statistically significant differences were detected' unless equivalence testing with a prespecified margin is provided.
  2. [Methods, 'Model development and performance evaluation'] The test set is described as '191 2D images from 15 menisci' while the data split describes 15 subjects; clarify whether '15 menisci' refers to 15 subjects or 15 individual menisci, and state how many medial/lateral menisci are included.
  3. [Table 1] There is a formatting error in the 95% CI for the CNN2 row: '0.885 0.904' is missing a dash and should read '0.885-0.904'.
  4. [Figure 1 caption] The caption says 'NIR value' but the text uses 'NAFP' (number of adiabatic full passages); please correct this inconsistency.
  5. [Discussion] The text refers to 'cartridge compartments' where 'cartilage compartments' is intended; also, in the Methods there is a duplicated 'T1' in 'per adiabatic T1T1ρ preparation.'
  6. [Methods, network description] The '1D convolutional block' added in front of the VGG19 blocks is described as consisting of 1x1 convolutional filters; calling it a 1D convolutional block is confusing and should be clarified.
  7. [General] No code or data availability statement is provided, which limits reproducibility of the transfer-learning implementation.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: held-out evaluation and independent relaxometry fitting keep the segmentation and relaxometry claims self-contained.

full rationale

The paper's derivation chain is a supervised segmentation study: radiologist ROIs are used to train two attention U-Nets, the trained models are applied to a held-out test set (36/10/15 subject split; 191 2D images from 15 menisci), and relaxometry values are then computed from both manual and automatic ROIs using the same UTE-Cones fitting procedure. The only fitted quantities are the CNN weights, and they are fit to training ROIs; the test-set Dice scores and relaxometry correlations are therefore not forced by construction. The relaxometry results are not an input to the CNN: T1, T1rho, and T2* are obtained by Levenberg-Marquardt fitting of UTE-Cones data within the ROIs, so agreement between manual and automatic segmentations is a post-hoc validation, not a tautology. Citations to prior work on UTE-Cones sequences and fitting equations (e.g., refs. 8, 9, 33) supply imaging methods, not the segmentation result, and no uniqueness theorem or ansatz is imported from the authors' own prior work to foreclose alternatives. The paper itself acknowledges that the image selection for ROI outlining was 'strictly subjective,' which is a ground-truth validity limitation, but it does not make the derivation circular. The statistical concern that 191 2D images from 15 menisci are treated as independent in t-tests, and that no formal equivalence test is reported, affects the strength of the 'no associated differences' claim, but this is a correctness/statistical-power issue rather than a circularity issue. Overall, the central claims are grounded in held-out data and independent fitting, so no circular step is present.

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

The paper contributes an empirical ML pipeline. Its quantitative claims depend on learned network weights, chosen hyperparameters, and the assumptions that radiologist ROIs and 2D slice-level analysis are valid. No new physical entities are introduced.

free parameters (6)
  • CNN weights and biases = Trained on 36 subjects, 458 2D images augmented to 2748
    All learned network parameters are fitted to the training data; the central segmentation claim depends on them.
  • Learning rate and momentum = 0.001 and 0.9
    Adam optimizer hyperparameters chosen by hand; standard values.
  • Batch size = 32
    Chosen by hand during training.
  • Learning rate decay and early stopping = Drop factor 0.5 every 5 epochs, patience 15 epochs
    Training schedule chosen by hand and monitored on validation set.
  • Edge-aware contrast manipulation parameters = Not specified
    Reported as 'selected experimentally'; exact values not given, so reproduction requires tuning.
  • Preprocessing crop and resizing = Crop to 192x192, resize to 224x224
    Input geometry selected to match VGG19 input size.
assumptions (5)
  • standard math Dice score-based loss is appropriate for heavily imbalanced segmentation
    The paper relies on this standard choice, citing prior segmentation literature.
  • domain assumption Features from the first two VGG19 blocks trained on ImageNet transfer usefully to grayscale MR images
    The paper adds a 1x1 convolutional block to adapt grayscale inputs, but does not ablate the transfer learning contribution.
  • domain assumption Manual ROIs by two radiologists on subtracted AdiabT1rho-weighted images are valid ground truth
    All training labels and evaluation metrics depend on this assumption; the image choice was subjective.
  • domain assumption 2D slices containing the meniscus are sufficient for relaxometry estimation
    The method averages per-slice relaxation values from 2D ROIs and does not validate against a full 3D volume measurement.
  • domain assumption UTE-Cones relaxometry fitting equations from prior literature are accurate
    T1, T1rho, and T2* values are computed with established fitting procedures cited from earlier work.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Knee menisci segmentation and relaxometry of 3D ultrashort echo time (UTE) cones MR imaging using attention U-Net with transfer learning." pith.science (2026). https://pith.science/paper/QTW4GHLI

@misc{pith2026190801594,
  author       = {Pith},
  title        = {Pith review of: Knee menisci segmentation and relaxometry of 3D ultrashort echo time (UTE) cones MR imaging using attention U-Net with transfer learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/QTW4GHLI}},
  note         = {Machine review of arXiv:1908.01594}
}
abstract

The purpose of this work is to develop a deep learning-based method for knee menisci segmentation in 3D ultrashort echo time (UTE) cones magnetic resonance (MR) imaging, and to automatically determine MR relaxation times, namely the T1, T1$\rho$, and T2* parameters, which can be used to assess knee osteoarthritis (OA). Whole knee joint imaging was performed using 3D UTE cones sequences to collect data from 61 human subjects. Regions of interest (ROIs) were outlined by two experienced radiologists based on subtracted T1$\rho$-weighted MR images. Transfer learning was applied to develop 2D attention U-Net convolutional neural networks for the menisci segmentation based on each radiologist's ROIs separately. Dice scores were calculated to assess segmentation performance. Next, the T1, T1$\rho$, T2* relaxations, and ROI areas were determined for the manual and automatic segmentations, then compared.The models developed using ROIs provided by two radiologists achieved high Dice scores of 0.860 and 0.833, while the radiologists' manual segmentations achieved a Dice score of 0.820. Linear correlation coefficients for the T1, T1$\rho$, and T2* relaxations calculated using the automatic and manual segmentations ranged between 0.90 and 0.97, and there were no associated differences between the estimated average meniscal relaxation parameters. The deep learning models achieved segmentation performance equivalent to the inter-observer variability of two radiologists. The proposed deep learning-based approach can be used to efficiently generate automatic segmentations and determine meniscal relaxations times. The method has the potential to help radiologists with the assessment of meniscal diseases, such as OA.

Figures

Figures reproduced from arXiv: 1908.01594 by the authors.

Figure 1
Figure 1. The MR images obtained using the UTE 3D cones for the NIR [PITH_FULL_IMAGE:figures/full_fig_p025_1.png] view at source ↗
Figure 2
Figure 2. The proposed 2D attention U-Net CNN for the menisci segmentation. Gray colors indicate the convolutional blocks initiated with the weights extracted from the VGG19 network. For each block the number of filters is indicated below the block type. AL – attention layer, Conv – 2D convolutional block, Max pool – max pooling operator, Up – up sampling with a 2D transposed convolutional block (kernel size of 2x2, stride of… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

49 extracted references · 48 canonical work pages

  1. [1]

    Osteoarthritis: a disease of the joint as an organ

    Loeser RF, Goldring SR, Scanzello CR, Goldring MB. Osteoarthritis: a disease of the joint as an organ. Arthritis Rheum. 2012;64(6):1697-1707

    work page 2012

  2. [2]

    the Effects of Meniscus Injury on the Development of Knee Osteoarthritis: data From the Osteoarthritis Initiative

    Badlani JT, Borrero C, Golla S, Harner CD, Irrgang JJ. the Effects of Meniscus Injury on the Development of Knee Osteoarthritis: data From the Osteoarthritis Initiative. Am J Sports Med. 2013;41(6):1238-1244

    work page 2013

  3. [3]

    Meniscal tear and extrusion are strongly associated with progression of symptomatic knee osteoarthritis as assessed by quantitative magnetic resonance imaging

    Berthiaume M-J, Raynauld J-P, Martel-Pelletier J, et al. Meniscal tear and extrusion are strongly associated with progression of symptomatic knee osteoarthritis as assessed by quantitative magnetic resonance imaging. Ann Rheum Dis. 2005;64(4):556-563

    work page 2005

  4. [4]

    The association of meniscal pathologic changes with cartilage loss in symptomatic knee osteoarthritis

    Hunter DJ, Zhang YQ, Niu JB, et al. The association of meniscal pathologic changes with cartilage loss in symptomatic knee osteoarthritis. Arthritis Rheum. 2006;54(3):795-801

    work page 2006

  5. [5]

    Cartilage and meniscal T2 relaxation time as non-invasive biomarker for knee osteoarthritis and cartilage repair procedures

    Baum T, Joseph GB, Karampinos DC, Jungmann PM, Link TM, Bauer JS. Cartilage and meniscal T2 relaxation time as non-invasive biomarker for knee osteoarthritis and cartilage repair procedures. Osteoarthr Cartil. 2013;21(10):1474-1484

    work page 2013

  6. [6]

    Joseph GB, Baum T, Alizai H, et al. Baseline mean and heterogeneity of MR cartilage T2 are associated with morphologic degeneration of cartilage, meniscus, and bone marrow over 3 years - data from the Osteoarthritis Initiative. Osteoarthr Cartil. 2012;20(7):727-735

    work page 2012

  7. [7]

    UTE imaging in the musculoskeletal system

    Chang EY, Du J, Chung CB. UTE imaging in the musculoskeletal system. J Magn Reson imaging. 2015;41(4):870-883

    work page 2015

  8. [8]

    3D adiabatic T1ρ prepared ultrashort echo time cones sequence for whole knee imaging

    Ma Y-J, Carl M, Searleman A, Lu X, Chang EY, Du J. 3D adiabatic T1ρ prepared ultrashort echo time cones sequence for whole knee imaging. Magn Reson Med. 2018;80(4):1429-1439

    work page 2018

Show all 49 references
  1. [9]

    Whole knee joint T1 values measured in vivo at 3T by combined 3D ultrashort echo time cones actual flip angle and variable flip angle methods

    Ma Y-J, Zhao W, Wan L, et al. Whole knee joint T1 values measured in vivo at 3T by combined 3D ultrashort echo time cones actual flip angle and variable flip angle methods. Magn Reson Med. 2019;81(3):1634-1644

    2019

  2. [10]

    Correlation of semiquantitative vs quantitative MRI meniscus measures in osteoarthritic knees: results from the Osteoarthritis Initiative

    Blöcker K, Guermazi A, Wirth W, et al. Correlation of semiquantitative vs quantitative MRI meniscus measures in osteoarthritic knees: results from the Osteoarthritis Initiative. Skeletal Radiol. 2014;43(2):227-232

    2014

  3. [11]

    Segmentation of joint and musculoskeletal tissue in the study of arthritis

    Pedoia V, Majumdar S, Link TM. Segmentation of joint and musculoskeletal tissue in the study of arthritis. Magn Reson Mater Physics, Biol Med. 2016;29(2):207-221

    2016

  4. [12]

    A survey on deep learning in medical image analysis

    Litjens G, Kooi T, Bejnordi BE, et al. A survey on deep learning in medical image analysis. Med Image Anal. 2017;42:60-88

    2017

  5. [13]

    U-net: Convolutional networks for biomedical image segmentation

    Ronneberger O, Fischer P, Brox T. U-net: Convolutional networks for biomedical image segmentation. In: International Conference on Medical Image Computing and Computer-Assisted Intervention. 2015:234-241

    2015

  6. [14]

    Automated segmentation of knee bone and cartilage combining statistical shape knowledge and convolutional neural networks: Data from the Osteoarthritis Initiative

    Ambellan F, Tack A, Ehlke M, Zachow S. Automated segmentation of knee bone and cartilage combining statistical shape knowledge and convolutional neural networks: Data from the Osteoarthritis Initiative. Med Image Anal. 2018;52:109- 118

    2018

  7. [15]

    Deep convolutional neural network and 3D deformable approach for tissue segmentation in musculoskeletal magnetic resonance imaging

    Liu F, Zhou Z, Jang H, Samsonov A, Zhao G, Kijowski R. Deep convolutional neural network and 3D deformable approach for tissue segmentation in musculoskeletal magnetic resonance imaging. Magn Reson Med. 2018;79(4):2379- 2391

    2018

  8. [16]

    Deep feature learning for knee cartilage segmentation using a triplanar convolutional neural network

    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. 2013:246-253

    2013

  9. [17]

    Deep convolutional neural network for segmentation of knee joint anatomy

    Zhou Z, Zhao G, Kijowski R, Liu F. Deep convolutional neural network for segmentation of knee joint anatomy. Magn Reson Med. 2018;80(6), 2759-2770

    2018

  10. [18]

    Use of 2D U-Net Convolutional Neural Networks for Automated Cartilage and Meniscus Segmentation of Knee MR Imaging Data to Determine Relaxometry and Morphometry

    Norman B, Pedoia V, Majumdar S. Use of 2D U-Net Convolutional Neural Networks for Automated Cartilage and Meniscus Segmentation of Knee MR Imaging Data to Determine Relaxometry and Morphometry. Radiology. 2018;288(1):177-185

    2018

  11. [19]

    Knee menisci segmentation using convolutional neural networks: data from the Osteoarthritis Initiative

    Tack A, Mukhopadhyay A, Zachow S. Knee menisci segmentation using convolutional neural networks: data from the Osteoarthritis Initiative. Osteoarthr Cartil. 2018;26(5):680-688

    2018

  12. [20]

    Automated segmentation of the menisci from MR images

    Fripp J, Bourgeat P, Engstrom C, Ourselin S, Crozier S, Salvado O. Automated segmentation of the menisci from MR images. In: IEEE International Symposium on Biomedical Imaging: From Nano to Macro. 2009:510-513

    2009

  13. [21]

    Automated segmentation and analysis of normal and osteoarthritic knee menisci from magnetic resonance images - data from the Osteoarthritis Initiative

    Paproki A, Engstrom C, Chandra SS, Neubert A, Fripp J, Crozier S. Automated segmentation and analysis of normal and osteoarthritic knee menisci from magnetic resonance images - data from the Osteoarthritis Initiative. Osteoarthr Cartil. 2014;22(9):1259-1270

    2014

  14. [22]

    The unified extreme learning machines and discriminative random fields for automatic knee cartilage and meniscus segmentation from multi-contrast MR images

    Zhang K, Lu W, Marziliano P. The unified extreme learning machines and discriminative random fields for automatic knee cartilage and meniscus segmentation from multi-contrast MR images. Mach Vis Appl. 2013;24(7):1459- 1472

    2013

  15. [23]

    T1ρ relaxation time of the meniscus and its relationship with T1ρ of adjacent cartilage in knees with acute ACL injuries at 3 T

    Bolbos RI, Link TM, Ma CB, Majumdar S, Li X. T1ρ relaxation time of the meniscus and its relationship with T1ρ of adjacent cartilage in knees with acute ACL injuries at 3 T. Osteoarthr Cartil. 2009;17(1):12-18

    2009

  16. [24]

    Meniscal measurements of T1ρ and T2 at MR imaging in healthy subjects and patients with osteoarthritis

    Rauscher I, Stahl R, Cheng J, et al. Meniscal measurements of T1ρ and T2 at MR imaging in healthy subjects and patients with osteoarthritis. Radiology. 2008;249(2):591-600

    2008

  17. [25]

    Semi-automated segmentation to assess the lateral meniscus in normal and osteoarthritic knees

    Swanson MS, Prescott JW, Best TM, et al. Semi-automated segmentation to assess the lateral meniscus in normal and osteoarthritic knees. Osteoarthr Cartil. 2010;18(3):344-353

    2010

  18. [26]

    Ultrashort TE T1ρ (UTE T1ρ) imaging of the Achilles tendon and meniscus

    Du J, Carl M, Diaz E, et al. Ultrashort TE T1ρ (UTE T1ρ) imaging of the Achilles tendon and meniscus. Magn Reson Med. 2010;64(3):834-842

    2010

  19. [27]

    Very deep convolutional networks for large-scale image recognition

    Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition. arXiv e‐prints. 2014

    2014

  20. [28]

    Going deeper with convolutions

    Szegedy C, Liu W, Jia Y, et al. Going deeper with convolutions. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2015:1-9

    2015

  21. [29]

    Imagenet: A large-scale hierarchical image database

    Deng J, Dong W, Socher R, Li L-J, Li K, Fei-Fei L. Imagenet: A large-scale hierarchical image database. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2009:248-255

    2009

  22. [30]

    Object detectors emerge in deep scene cnns

    Zhou B, Khosla A, Lapedriza A, Oliva A, Torralba A. Object detectors emerge in deep scene cnns. arXiv e‐prints. 2014

    2014

  23. [31]

    A Deep Feature Fusion Methodology for Breast Cancer Diagnosis Demonstrated on Three Imaging Modality Datasets

    Antropova N, Huynh BQ, Giger ML. A Deep Feature Fusion Methodology for Breast Cancer Diagnosis Demonstrated on Three Imaging Modality Datasets. Med Phys. 2017;44(10):5126-5171

    2017

  24. [32]

    Training constrained deconvolutional networks for road scene semantic segmentation

    Ros G, Stent S, Alcantarilla PF, Watanabe T. Training constrained deconvolutional networks for road scene semantic segmentation. arXiv e‐prints. 2016

    2016

  25. [33]

    An algorithm for least-squares estimation of nonlinear parameters

    Marquardt DW. An algorithm for least-squares estimation of nonlinear parameters. J Soc Ind Appl Math. 1963;11(2):431-441

    1963

  26. [34]

    Elastix: a toolbox for intensity-based medical image registration

    Klein S, Staring M, Murphy K, Viergever MA, Pluim JPW. Elastix: a toolbox for intensity-based medical image registration. IEEE Trans Med Imaging. 2010;29(1):196-205

    2010

  27. [35]

    Fast parallel image registration on CPU and GPU for diagnostic classification of Alzheimer’s disease

    Shamonin DP, Bron EE, Lelieveldt BPF, Smits M, Klein S, Staring M. Fast parallel image registration on CPU and GPU for diagnostic classification of Alzheimer’s disease. Front Neuroinform. 2014;7:50

    2014

  28. [36]

    Learn to pay attention

    Jetley S, Lord NA, Lee N, Torr PHS. Learn to pay attention. arXiv e‐prints. 2018

    2018

  29. [37]

    Attention U-Net: Learning Where to Look for the Pancreas

    Oktay O, Schlemper J, Folgoc L Le, et al. Attention U-Net: Learning Where to Look for the Pancreas. arXiv e‐prints. 2018

    2018

  30. [38]

    Convolutional neural networks for medical image analysis: Full training or fine tuning? IEEE Trans Med Imaging

    Tajbakhsh N, Shin JY, Gurudu SR, et al. Convolutional neural networks for medical image analysis: Full training or fine tuning? IEEE Trans Med Imaging. 2016;35(5):1299-1312

    2016

  31. [39]

    Good practice in CNN feature transfer

    Zheng L, Zhao Y, Wang S, Wang J, Tian Q. Good practice in CNN feature transfer. arXiv e‐prints. 2016

    2016

  32. [40]

    Breast mass classification in sonography with transfer learning using a deep convolutional neural network and color conversion

    Byra M, Galperin M, Ojeda-Fournier H, et al. Breast mass classification in sonography with transfer learning using a deep convolutional neural network and color conversion. Med Phys. 2019;46(2):746-755

    2019

  33. [41]

    V-net: Fully convolutional neural networks for volumetric medical image segmentation

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

    2016

  34. [42]

    Generalised Dice overlap as a deep learning loss function for highly unbalanced segmentations

    Sudre CH, Li W, Vercauteren T, Ourselin S, Cardoso MJ. Generalised Dice overlap as a deep learning loss function for highly unbalanced segmentations. In: Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support. Springer; 2017:240-248

    2017

  35. [43]

    Local Laplacian Filters: Edge-aware Image Processing with a Laplacian Pyramid

    Paris S, Hasinoff SW, Kautz J. Local Laplacian Filters: Edge-aware Image Processing with a Laplacian Pyramid. Commun ACM. 2015;58(3):81-91

    2015

  36. [44]

    Adam: A method for stochastic optimization

    Kingma D, Ba J. Adam: A method for stochastic optimization. arXiv e‐prints. 2014

    2014

  37. [45]

    Understanding the difficulty of training deep feedforward neural networks

    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

  38. [46]

    TensorFlow: A System for Large-Scale Machine Learning

    Abadi M, Barham P, Chen J, et al. TensorFlow: A System for Large-Scale Machine Learning. In: USENIX Symposium on Operating Systems Design and Implementation. 2016:265-283

    2016

  39. [47]

    An introduction to ROC analysis

    Fawcett T. An introduction to ROC analysis. Pattern Recognit Lett. 2006;27(8):861-874

    2006

  40. [48]

    Segmentation of knee images: a grand challenge

    Heimann T, Morrison BJ, Styner MA, Niethammer M, Warfield S. Segmentation of knee images: a grand challenge. http://www.ski10.org/ski10.pdf

  41. [49]

    Why rankings of biomedical image analysis competitions should be interpreted with care

    Maier-Hein L, Eisenmann M, Reinke A, et al. Why rankings of biomedical image analysis competitions should be interpreted with care. Nat Commun. 2018;9(1):5217. Table 1. The average Dice scores (plus median and 95% CI) calculated using the manual and automatic segmentations. Ra...

    2018

Pith tools

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