Finding the optimal parameter setting for an ensemble-based lesion detector

In this paper, we show how the optimal setting of an ensemble-based lesion detector can improve the performance of the system. That is, instead of the individually optimal parameter settings of the member algorithms, we determine such a setting that maximizes the accuracy of the ensemble. Our case study is an ensemble composed for microaneurysm detection on retinal images that is currently ranked as first in an online competition. The determination of the optimal parameter setting is a very time consuming problem, especially in fusion-based systems. Thus, we suggest several speedups in a stochastic optimum search framework regarding the evaluation of the corresponding special energy function. In other words, we also suggest some modifications and tuning of stochastic search approaches for ensemble-based image detectors evaluated on test databases.

[1]  H. Kälviäinen,et al.  DIARETDB 1 diabetic retinopathy database and evaluation protocol , 2007 .

[2]  Qin Li,et al.  Detection of microaneurysms using multi-scale correlation coefficients , 2010, Pattern Recognit..

[3]  Ludmila I. Kuncheva,et al.  Combining Pattern Classifiers: Methods and Algorithms , 2004 .

[4]  H. B. Mitchell.  Image Fusion: Theories, Techniques and Applications , 2010 .

[5]  Lionel Valet,et al.  Choquet Integral Parameter Optimization for a Fusion System Devoted to Image Interpretation , 2012, IPMU.

[6]  Bálint Antal,et al.  An Ensemble-Based System for Microaneurysm Detection and Diabetic Retinopathy Grading , 2012, IEEE Transactions on Biomedical Engineering.

[7]  András Hajdu,et al.  Microaneurysm detection in retinal images using a rotating cross-section based model , 2011, 2011 IEEE International Symposium on Biomedical Imaging: From Nano to Macro.

[8]  S. Abdelazeem,et al.  Micro-aneurysm detection using vessels removal and circular Hough transform , 2002, Proceedings of the Nineteenth National Radio Science Conference.

[9]  P F Sharp,et al.  An image-processing strategy for the segmentation and quantification of microaneurysms in fluorescein angiograms of the ocular fundus. , 1996, Computers and biomedical research, an international journal.

[10]  Joni-Kristian Kämäräinen,et al.  The DIARETDB1 Diabetic Retinopathy Database and Evaluation Protocol , 2007, BMVC.

[11]  C. D. Gelatt,et al.  Optimization by Simulated Annealing , 1983, Science.

[12]  Pascale Massin,et al.  Automatic detection of microaneurysms in color fundus images , 2007, Medical Image Anal..

[13]  Subhash C. Bagui,et al.  Combining Pattern Classifiers: Methods and Algorithms , 2005, Technometrics.

[14]  W. Marsden.  I and J , 2012 .

[15]  Aaas News,et al.  Book Reviews , 1893, Buffalo Medical and Surgical Journal.

[16]  Georg Ch. Pflug,et al.  Simulated Annealing for noisy cost functions , 1996, J. Glob. Optim..

[17]  Karel J. Zuiderveld,et al.  Contrast Limited Adaptive Histogram Equalization , 1994, Graphics Gems.

[18]  Heikki Kälviäinen,et al.  DIARETDB 0 : Evaluation Database and Methodology for Diabetic Retinopathy Algorithms , 2007 .

[19]  Patrick Pérez,et al.  Object removal by exemplar-based inpainting , 2003, 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2003. Proceedings..

[20]  Thomas Walter,et al.  Automatic Detection of Microaneurysms in Color Fundus Images of the Human Retina by Means of the Bounding Box Closing , 2002, ISMDA.

[21]  Qin Li,et al.  Retinopathy Online Challenge: Automatic Detection of Microaneurysms in Digital Color Fundus Photographs , 2010, IEEE Transactions on Medical Imaging.

[22]  B. S.  Simulated Annealing with Noisy or Imprecise Energy Measurements , 2004 .

[23]  Aliaa A. A. Youssif,et al.  Comparative Study of Contrast Enhancement and Illumination Equalization Methods for Retinal Vasculat , 2006 .