Name
Title | ||
|---|---|---|
Learning from only positive and unlabeled data to detect lesions in vascular CT images |
Abstract | ||
|---|---|---|
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. |
Year | DOI | Venue |
|---|---|---|
2011 | 10.1007/978-3-642-23626-6_2 | Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention |
Keywords | Field | DocType |
training data,coronary heart disease,training set,small amount,additional iteration,state-of-the-art supervised method,present result,vascular ct image,ct image,unlabeled data,synthetic data | Training set,Pattern recognition,Computer science,Support vector machine,Synthetic data,Artificial intelligence,Classifier (linguistics),True positive rate,Machine learning | Conference |
Volume | Issue | ISSN |
14 | Pt 3 | 0302-9743 |
Citations | PageRank | References |
5 | 0.42 | 9 |
Authors | ||
5 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Maria A. Zuluaga | 1 | 279 | 25.84 |
| Don Hush | 2 | 320 | 51.44 |
| Edgar J. F. Delgado Leyton | 3 | 20 | 1.85 |
| Marcela Hernández Hoyos | 4 | 228 | 15.91 |
| Maciej Orkisz | 5 | 315 | 24.14 |