Unsupervised feature extraction of anterior chamber OCT images for ordering and classification

Amil, Pablo; González, Laura; Arrondo, Elena; Salinas, Cecilia; Güell, José L.; Masoller, Cristina; Parlitz, Ulrich

doi:10.1038/s41598-018-38136-8

Cited by 15 publications

(19 citation statements)

References 26 publications

(24 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…OCT anterior chamber images are routinely used for the early diagnosis of glaucoma. We show that, when images with artifacts (outliers) are removed from the training dataset, the performance of the unsupervised ordering algorithm [46] improves significantly. We also compare the performance of these methods with the performance of other popular methods used in the literature.…”

Section: Introductionmentioning

confidence: 94%

“…This database consists of 1213 OCT images of the anterior chamber of the eye of healthy and non-healthy patients of the Instituto de Microcirugia Ocular in Barcelona. The database was analyzed in Amil et al [46] where an unsupervised algorithm for ordering the images was proposed. The images had been classified in four categories (closed, narrow, open, and wide open) by two expert ophthalmologists.…”

Section: Anterior Chamber Oct Imagesmentioning

confidence: 99%

“…The distance matrix D ij was calculated as described in detail in Amil et al [46]: by comparing pixel-by-pixel, after preprocessing the images to adjust the alignment and to enhance the contrast. For the algorithms that don't use the distance matrix (OCSVM and distance to center of mass), the same pre-processing was used.…”

Section: Anterior Chamber Oct Imagesmentioning

confidence: 99%

“…For the OCT database, there is no a priori definition of outliers (i.e., no ground truth), all the images were drawn from the same database. However, as a proxy for determining the performance of the outlier finding methods, we used the performance of the unsupervised methods proposed in Amil et al [46] when ignoring the images identified as outliers.…”

Section: Anterior Chamber Oct Imagesmentioning

confidence: 99%

See 3 more Smart Citations

Outlier Mining Methods Based on Graph Structure Analysis

2019

Self Cite

View full text Add to dashboard Cite

Outlier detection in high-dimensional datasets is a fundamental and challenging problem across disciplines that has also practical implications, as removing outliers from the training set improves the performance of machine learning algorithms. While many outlier mining algorithms have been proposed in the literature, they tend to be valid or efficient for specific types of datasets (time series, images, videos, etc.). Here we propose two methods that can be applied to generic datasets, as long as there is a meaningful measure of distance between pairs of elements of the dataset. Both methods start by defining a graph, where the nodes are the elements of the dataset, and the links have associated weights that are the distances between the nodes. Then, the first method assigns an outlier score based on the percolation (i.e., the fragmentation) of the graph. The second method uses the popular IsoMap non-linear dimensionality reduction algorithm, and assigns an outlier score by comparing the geodesic distances with the distances in the reduced space. We test these algorithms on real and synthetic datasets and show that they either outperform, or perform on par with other popular outlier detection methods. A main advantage of the percolation method is that is parameter free and therefore, it does not require any training; on the other hand, the IsoMap method has two integer number parameters, and when they are appropriately selected, the method performs similar to or better than all the other methods tested.

show abstract

Section: Introductionmentioning

confidence: 94%

Section: Anterior Chamber Oct Imagesmentioning

confidence: 99%

Section: Anterior Chamber Oct Imagesmentioning

confidence: 99%

Section: Anterior Chamber Oct Imagesmentioning

confidence: 99%

See 2 more Smart Citations

Outlier Mining Methods Based on Graph Structure Analysis

2019

Self Cite

View full text Add to dashboard Cite

show abstract

“…Importantly, the retina is a noninvasive window for studying brain deceases such as Alzheimer or dementia, because changes in retinal microvasculature may reflect similar changes in cerebral microvasculature. 8 Here we first present novel algorithms to analyze a database of AS-OCT images: we demonstrate machine learning algorithms to order the images of an AS-OCT image database according to the degree of angle-closure, 9 and also, to identify the images in the database that contain artifacts. 10 In the second part we present novel algorithms for the analysis of color retinal fundus images.…”

Section: Introductionmentioning

confidence: 99%