Feature selection using ROC curves on classification problems

Serrano, Adela; Soria, Emilio; Martín, Juan Enrique Pérez; Magdalena‐Benedito, Rafael; Gómez, Juan Pablo

doi:10.1109/ijcnn.2010.5596692

Cited by 20 publications

(15 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To obtain the most relevant features for the respective classification, we have adopted iterative sequential maximisation of task performance (also called the 'wrapper feature selection' [41]) in which initially the data is divided into k folds (in our case k = 5). Then the first feature is selected which has Gammanormalised derivative maximum mean classification performance across the folds.…”

Section: Feature Extraction and Selectionmentioning

confidence: 99%

A Novel Adaptive Deformable Model for Automated Optic Disc and Cup Segmentation to Aid Glaucoma Diagnosis

et al. 2017

View full text Add to dashboard Cite

show abstract

Section: Feature Extraction and Selectionmentioning

confidence: 99%

A Novel Adaptive Deformable Model for Automated Optic Disc and Cup Segmentation to Aid Glaucoma Diagnosis

et al. 2017

View full text Add to dashboard Cite

show abstract

“…From the available set of features, the feature with the highest Area Under the Curve (AUC) [24] is selected. The next feature is chosen in such a way that when it is used along with the first selected feature, it will give the highest AUC compared to other non-selected features.…”

Section: Feature Selectionmentioning

confidence: 99%

“…In the filter approach, the features are ranked with respect to their effectiveness in classification and higher ranked features are thresholded out. In order to determine most relevant features, an independent evaluation criterion for binary classification is used [18] and AUC is selected as its evaluation measure [24]. The features with higher AUC are ranked higher and the features with AUC greater than 0.9 are selected.…”

Section: Feature Selectionmentioning

confidence: 99%

Retinal Area Detector From Scanning Laser Ophthalmoscope (SLO) Images for Diagnosing Retinal Diseases

Haleem

Han

Hemert

et al. 2015

IEEE J. Biomed. Health Inform.

View full text Add to dashboard Cite

• This is the accepted manuscript version of the paper. c 2014 IEEE. Abstract-Scanning Laser Ophthalmoscopes (SLOs) can be used for early detection of retinal diseases. With the advent of latest screening technology, the advantage of using SLO is its wide Field of View (FOV), which can image a large part of the retina for better diagnosis of the retinal diseases. On the other hand, during the imaging process, artefacts such as eyelashes and eyelids are also imaged along with the retinal area. This brings a big challenge on how to exclude these artefacts. In this paper, we propose a novel approach to automatically extract out true retinal area from an SLO image based on image processing and machine learning approaches. To reduce the complexity of image processing tasks and provide a convenient primitive image pattern, we have grouped pixels into different regions based on the regional size and compactness, called superpixels. The framework then calculates image based features reflecting textural and structural information and classifies between retinal area and artefacts. The experimental evaluation results have shown good performance with an overall accuracy of 92%.

show abstract

“…The features are calculated for red and green channels (blue channel is zero in SLO images) and classification power was calculated using Area Under the Curve (AUC) [11]. AUC is taken using 5-fold cross validation on the training set.…”

Section: Feature Generation and Selectionmentioning

confidence: 99%