Building a reference multimedia database for interstitial lung diseases

Depeursinge, Adrien; Vargas, Alejandro; Platon, Alexandra; Geissbühler, Antoine; Poletti, Pierre–Alexandre; Müller, Henning

doi:10.1016/j.compmedimag.2011.07.003

Cited by 214 publications

(131 citation statements)

References 22 publications

Supporting

Mentioning

117

Contrasting

Unclassified

Order By: Relevance

“…The second dataset is a larger, publicly available dataset on interstitial lung diseases (see [12] for a description). In this texture classification problem with scans from 73 patients, we do patch-wise classification of five types of lung tissue (healthy tissue: 22%, emphysema: 3%, ground glass: 16%, fibrosis: 15%, micronodules: 44%).…”

Section: Datasets and Problemsmentioning

confidence: 99%

Learning Features for Tissue Classification with the Classification Restricted Boltzmann Machine

Tulder

Bruijne

2014

Medical Computer Vision: Algorithms for Big Data

View full text Add to dashboard Cite

Abstract. Performance of automated tissue classification in medical imaging depends on the choice of descriptive features. In this paper, we show how restricted Boltzmann machines (RBMs) can be used to learn features that are especially suited for texture-based tissue classification. We introduce the convolutional classification RBM, a combination of the existing convolutional RBM and classification RBM, and use it for discriminative feature learning. We evaluate the classification accuracy of convolutional and non-convolutional classification RBMs on two lung CT problems. We find that RBM-learned features outperform conventional RBM-based feature learning, which is unsupervised and uses only a generative learning objective, as well as often-used filter banks. We show that a mixture of generative and discriminative learning can produce filters that give a higher classification accuracy.

show abstract

Section: Datasets and Problemsmentioning

confidence: 99%

Learning Features for Tissue Classification with the Classification Restricted Boltzmann Machine

Tulder

Bruijne

2014

Medical Computer Vision: Algorithms for Big Data

View full text Add to dashboard Cite

show abstract

“…The second dataset is the publicly available ILD dataset [3] which contains 905 selected image slices from 120 patients including in total 13 ILD labels (such as ground glass, fibrosis, micronodules, consolidation, etc). This dataset is sparsely annotated, and thus could only be used for the patch-based model.…”

Section: Dataset and Implementation Detailsmentioning

confidence: 99%

“…This is a common problem in medical imaging, which explains the fact of missing ground truth or partially labeled databases. Indeed, to the best of our knowledge, the only available ILD dataset [3] is sparsely annotated; i.e. it includes only some parts of selected image slices out of the whole volumetric dataset.…”

Section: Introductionmentioning

confidence: 99%

Deep patch-based priors under a fully convolutional encoder-decoder architecture for interstitial lung disease segmentation

Vakalopoulou

Chassagnon

Paragios

et al. 2018

2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018)

View full text Add to dashboard Cite

Interstitial lung diseases (ILD) encompass a large spectrum of diseases sharing similarities in their physiopathology and computed tomography (CT) appearance. In this paper, we propose the adaption of a deep convolutional encoder-decoder (CED) that has shown high accuracy for image segmentation. Such architectures require annotation of the total region with pathological findings. This is difficult to acquire, due to uncertainty in the definition and extent of disease patterns and the need of significant human effort, especially for large datasets. Therefore, often current methods use patch-based implementations of convolutional neural networks, which however tend to produce spatially inhomogeneous segmentations due to their local contextual view. We exploit the advantages of both architectures by using the output of a patch-based classifier as a prior to a CED. Our method could advance the state-of-the-art in lung tissue segmentation using only a small number of newly annotated images.

show abstract

“…In this work, we have used publicly available dataset of ILD cases [1]. This database has Region of Interest (ROI) marked by three experienced radiologists.…”

Section: Datasetmentioning

confidence: 99%

“…Computed Tomography is nothing but detailed image created by computer by taking information from multiple X-rays of the lungs and surrounding structures. Since last three decades Computed Tomography scan of lungs are becoming very famous because of its discriminative information for different ILDs [1], [2]. There are two stages in computer aided diagnosis system.…”

Section: Introductionmentioning

confidence: 99%

Interstitial Lung Disease Classification Using Feed Forward Neural Networks

Dudhane

Shingadkar²,

Sanghavi³

et al. 2017

Proceedings of the International Conference on Communication and Signal Processing 2016 (ICCASP 2016)

View full text Add to dashboard Cite

This paper presents a classification scheme for Interstitial Lung Disease (ILD) pattern using Patch based approach and ANN classifier. Various methods are proposed for classification of ILD patterns but still accuracy can be improved when ILD cases are complex. In this paper, we have extracted features from size patch using histogram of Local Binary Patterns (LBP) and second order statistics such as Grey level cooccurrence matrix (GLCM), Grey level run length matrix (GLRLM). A two layer Feed-Forward Neural Network trained with Scaled Conjugate Gradient Back-propagation algorithm is used for classification. Accuracy of the work carried out in this paper isResults are verified and compared with different classifiers such as k-NN and SVM. This study has been carried out on publicly available database of ILD cases. ILD patches have been collected from -D Region of interest (ROI) marked by expert radiologist. Five frequently seen ILD patterns: Normal, Emphysema, Fibrosis, Ground Glass and Micronodule are considered in this study. Experimental results with proposed scheme outperforms in classification of ILD patterns.

show abstract

Building a reference multimedia database for interstitial lung diseases

Cited by 214 publications

References 22 publications

Learning Features for Tissue Classification with the Classification Restricted Boltzmann Machine

Learning Features for Tissue Classification with the Classification Restricted Boltzmann Machine

Deep patch-based priors under a fully convolutional encoder-decoder architecture for interstitial lung disease segmentation

Interstitial Lung Disease Classification Using Feed Forward Neural Networks

Contact Info

Product

Resources

About