A Classification Framework for Anomaly Detection
The Journal of Machine Learning Research
Automatic Detection of Calcified Coronary Plaques in Computed Tomography Data Sets
MICCAI '08 Proceedings of the 11th international conference on Medical Image Computing and Computer-Assisted Intervention - Part I
Nonparametric Intensity Priors for Level Set Segmentation of Low Contrast Structures
MICCAI '09 Proceedings of the 12th International Conference on Medical Image Computing and Computer-Assisted Intervention: Part I
Density-based similarity measures for content based search
Asilomar'09 Proceedings of the 43rd Asilomar conference on Signals, systems and computers
LIBSVM: A library for support vector machines
ACM Transactions on Intelligent Systems and Technology (TIST)
Feature selection for SVM-based vascular anomaly detection
MCV'10 Proceedings of the 2010 international MICCAI conference on Medical computer vision: recognition techniques and applications in medical imaging
A fast lesion registration to assist coronary heart disease diagnosis in CTA images
ICCVG'12 Proceedings of the 2012 international conference on Computer Vision and Graphics
| Hi-index | 0.00 |
Detecting vascular lesions is an important task in the diagnosis and follow-up of the coronary heart disease. While most existing solutions tackle calcified and non-calcified plaques separately, we present a new algorithm capable of detecting both types of lesions in CT images. It builds up on a semi-supervised classification framework, in which the training set is made of both unlabeled data and a small amount of data labeled as normal. Our method takes advantage of the arrival of newly acquired data to re-train the classifier and improve its performance. We present results on synthetic data and on datasets from 15 patients. With a small amount of labeled training data our method achieved a 89.8% true positive rate, which is comparable to state-of-the-art supervised methods, and the performance can improve after additional iterations.