• Aucun résultat trouvé

Organ Segmentation in CT Images With Weak Annotations: A Preliminary Study

N/A
N/A
Protected

Academic year: 2022

Partager "Organ Segmentation in CT Images With Weak Annotations: A Preliminary Study"

Copied!
1
0
0

Texte intégral

(1)

Organ Segmentation in CT Images With Weak Annotations: A Preliminary Study

Rosana El-Jurdi

(1,2)

, Caroline Petitjean

(1)

, Paul Honeine

(1)

, Fahed Abdallah

(2,3)

(1)LITIS, Universit´e de Rouen, France (2)Universit´e Libanaise, Liban (3)UTT, Troyes, France

Abstract

Medical image segmentation has unprecedented challenges, compared to natural image segmentation, in particular because of the scarcity of annotated datasets. Of particular interest is the ongoing 2019 SegTHOR competition, which consists in Segmenting THoracic Organs at Risk in CT images. While the fully supervised framework (i.e., pixel-level annotation) is considered in this competition, this paper seeks to push forward the competition to a new paradigm: weakly supervised segmentation, namely training with only bounding boxes that enclose the organs. After a pre-processing step, the proposed method operates a GrabCut algorithm that transforms the images into pixel-level annotated ones. And then a deep neural network is trained on the medical images, where several segmentation loss functions are examined. Experiments show the relevance of the proposed method, providing comparable results to the ongoing fully supervised segmentation competition.

Framework

SegTHOR Dataset

Constituted of 60 patients with lung cancer [1]

Contains groundtruth segmentations of 4 organs at risk Focus in this work is on heart segmentation

Challenges

Soft organs are hard to detect and share same gray scale intensities with neighboring tissues

Reasonable class imbalance is evident due to the vari- ation of organ size with respect to background

Low contrast in CT images

Outline of the method

Step 1: Data preprocessing

Step 2: From bounding boxes to label segments Step 3: Refining label estimates with deep learning

Proposed method

From bounding boxes to pixel annotations:

Grabcut method: an iterative intensity graph based algorithm that gener- ates label segments from bounding boxes, based on estimating Gaussian mixture models and optimizing graphs built from image pixels [2]

Training the segmentation model with U-net :

– Encoder: VGG16 backbone with its full convolutional layers removed, pre- trained with ImageNet weights.

– Decoder: deconvolutional layers that extract and concatenate feature maps to generate coarse grained labels.

– Skip grams combine feature maps from both encoder and decoder layers for accurate organ localization.

Addressing class imbalance using loss functions:

– Balanced cross entropy loss [3]: −

PN

i=1 βigi log(pi) – Dice loss [4]: 2 PN

i=1 pigi + ǫ PN

i=1 pi + PN

i=1 gi + ǫ – Tversky loss [5]:

PN

i=1 pigi + ǫ PN

i=1 pigi + α PN

i=1 pi(1 − gi) + β PN

i=1(1 − pi)gi + ǫ

Results and Analysis

Dataset Distribution: training (26 patients: 1477 slices), validation (4 patients:

155 slices), testing (20 patients: 693 slices)

Results: Dice similarity, with the mean, extrema and standard deviation values mean % std % max % min % Pre-trained only (with ImageNet) 10.09 6.42 32.19 0.15 Trained with cross entropy loss 71.87 12.88 89.78 0.0 Trained with balanced cross entropy loss 84.67 3.94 90.57 67.86

Trained with Dice loss 41.07 18.49 74.63 0.0

Trained with Tversky loss 28.94 10.97 53.89 1.28

Trained with Full Supervision 93.18 4.54 98.18 72.12

Analysis:

Cross entropy losses achieve the highest scores with a Dice accuracy of 84.67%.

Proper convergence of model with dice loss is dependent on smoothing variable ǫ Grabcut label estimates are poor representatives of labels

Conclusion

Proposed a weakly supervised framework for organ segmentation in CT images Examined multiple loss functions with balanced cross entropy resulting in best performance

Future work

Integrating to multi-class organ segmentation

Proposing a smoothing function that varies w.r.t training iterations

Proposing a training procedure that alternates between losses, taking into account the strengths of each loss function to handle class imbalance

References

[1] R. Trullo, C. Petitjean, S. Ruan, B. Dubray, D. Nie, and D. Shen, “Segmentation of organs at risk in thoracic ct images using a sharpmask architecture and conditional random fields,” in 14th IEEE International Symposium on Biomedical Imaging (ISBI), pp. 1003–1006, 2017.

[2] A. Khoreva, R. Benenson, J. H. Hosang, M. Hein, and B. Schiele, “Weakly supervised semantic labelling and instance segmentation,” CoRR, vol. abs/1603.07485, 2016.

[3] Y. S. Aurelio, G. M. de Almeida, C. L. de Castro, and A. P. Braga, “Learning from imbalanced data sets with weighted cross-entropy function,” Neural Processing Letters, Jan 2019.

[4] F. Milletari, N. Navab, and S. Ahmadi, “V-net: Fully convolutional neural networks for volumetric medical image segmentation,” CoRR, vol. abs/1606.04797, 2016.

[5] S. S. M. Salehi, D. Erdogmus, and A. Gholipour, “Tversky loss function for image segmentation using 3d fully convo- lutional deep networks,” CoRR, vol. abs/1706.05721, 2017.

Work supported by the National Council for Scientific Research of Lebanon, Agence Fran¸caise de la Francophonie, and ANR-18-CE23-0014

Références

Documents relatifs

For the problem of segmentation of healthy liver against lesions, the optimal size of texture was obtained for a size of 13 × 13 pixels.. The definition of a size of texture in

Finally, the proposed weakly supervised method provides an accuracy of up to 84.67% for the heart segmentation, using the balanced cross entropy with only 550 iterations, which is

This paper aims to explore these two challenges using Mask R-CNN trained with a data augmentation method designed to provide good performance on both rectilinear and fisheye images..

Since the mapping of I on B m becomes more accurate as the scale in- creases, it can be assumed that the knowledge extracted from B m to drive the segmentation of I becomes more

The Web-based abdominal organ CT image segmentation platform designed in this paper is equivalent to providing a service. There are mainly the following advantages. In

Methods An automatic method based on active contours without edges was used for left and the right ventricle cavity segmentation.. A large database of 1920 MR images obtained from

In this step, the atlas images and label images are down-sampled and resized to generate low-resolution atlas images I Ai l (i = 1,. , N ) which are then aligned with the

To segment tumor in PET images using multiple features, the spatial evidential clustering algorithm, e.g., SECM [8], is fully improved by integrating unsupervised distance