Segmentation and classification of breast lesions using dynamic and textural features in dynamic contrast enhanced-magnetic resonance imaging

Roberta Fusco, Mario Sansone, Carlo Sansone, Antonella Petrillo

Research output: Chapter in Book/Report/Conference proceedingConference contribution

11 Citations (Scopus)

Abstract

The aim of this study is to propose an approach, based on Multi Layer Perceptron classification of dynamic and textural features, for breast lesions segmentation and classification using Dynamic Contrast Enhanced-Magnetic Resonance Imaging data. We compared the performance obtainable with dynamic, textural and spatio-temporal features. In particular, 98 dynamic features, 60 textural features and 72 spatio-temporal features were considered. The dataset included 20 breast lesions, 10 benign and 10 malignant. The performance of lesion segmentation have been evaluated with respect to manual segmentation provided by an expert radiologist. Results of lesion classification were compared to histological findings. Our results indicate that Multi Layer Perceptron can achieve better results in terms of sensitivity, specificity and accuracy when dynamic features are considered both for lesion segmentation and classification (accuracy of 91 % and 70 %, respectively).

Original languageEnglish
Title of host publicationProceedings - IEEE Symposium on Computer-Based Medical Systems
DOIs
Publication statusPublished - 2012
Event25th IEEE International Symposium on Computer-Based Medical Systems, CBMS 2012 - Rome, Italy
Duration: Jun 20 2012Jun 22 2012

Other

Other25th IEEE International Symposium on Computer-Based Medical Systems, CBMS 2012
CountryItaly
CityRome
Period6/20/126/22/12

Fingerprint

Magnetic resonance
Breast
Magnetic Resonance Imaging
Imaging techniques
Neural Networks (Computer)
Multilayer neural networks
Sensitivity and Specificity

ASJC Scopus subject areas

  • Computer Science Applications
  • Radiology Nuclear Medicine and imaging

Cite this

Segmentation and classification of breast lesions using dynamic and textural features in dynamic contrast enhanced-magnetic resonance imaging. / Fusco, Roberta; Sansone, Mario; Sansone, Carlo; Petrillo, Antonella.

Proceedings - IEEE Symposium on Computer-Based Medical Systems. 2012. 6266312.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Fusco, R, Sansone, M, Sansone, C & Petrillo, A 2012, Segmentation and classification of breast lesions using dynamic and textural features in dynamic contrast enhanced-magnetic resonance imaging. in Proceedings - IEEE Symposium on Computer-Based Medical Systems., 6266312, 25th IEEE International Symposium on Computer-Based Medical Systems, CBMS 2012, Rome, Italy, 6/20/12. https://doi.org/10.1109/CBMS.2012.6266312
Fusco, Roberta ; Sansone, Mario ; Sansone, Carlo ; Petrillo, Antonella. / Segmentation and classification of breast lesions using dynamic and textural features in dynamic contrast enhanced-magnetic resonance imaging. Proceedings - IEEE Symposium on Computer-Based Medical Systems. 2012.
@inproceedings{db326d04aeb944d8ae65bcc6cd7cf17b,
title = "Segmentation and classification of breast lesions using dynamic and textural features in dynamic contrast enhanced-magnetic resonance imaging",
abstract = "The aim of this study is to propose an approach, based on Multi Layer Perceptron classification of dynamic and textural features, for breast lesions segmentation and classification using Dynamic Contrast Enhanced-Magnetic Resonance Imaging data. We compared the performance obtainable with dynamic, textural and spatio-temporal features. In particular, 98 dynamic features, 60 textural features and 72 spatio-temporal features were considered. The dataset included 20 breast lesions, 10 benign and 10 malignant. The performance of lesion segmentation have been evaluated with respect to manual segmentation provided by an expert radiologist. Results of lesion classification were compared to histological findings. Our results indicate that Multi Layer Perceptron can achieve better results in terms of sensitivity, specificity and accuracy when dynamic features are considered both for lesion segmentation and classification (accuracy of 91 {\%} and 70 {\%}, respectively).",
author = "Roberta Fusco and Mario Sansone and Carlo Sansone and Antonella Petrillo",
year = "2012",
doi = "10.1109/CBMS.2012.6266312",
language = "English",
isbn = "9781467320511",
booktitle = "Proceedings - IEEE Symposium on Computer-Based Medical Systems",

}

TY - GEN

T1 - Segmentation and classification of breast lesions using dynamic and textural features in dynamic contrast enhanced-magnetic resonance imaging

AU - Fusco, Roberta

AU - Sansone, Mario

AU - Sansone, Carlo

AU - Petrillo, Antonella

PY - 2012

Y1 - 2012

N2 - The aim of this study is to propose an approach, based on Multi Layer Perceptron classification of dynamic and textural features, for breast lesions segmentation and classification using Dynamic Contrast Enhanced-Magnetic Resonance Imaging data. We compared the performance obtainable with dynamic, textural and spatio-temporal features. In particular, 98 dynamic features, 60 textural features and 72 spatio-temporal features were considered. The dataset included 20 breast lesions, 10 benign and 10 malignant. The performance of lesion segmentation have been evaluated with respect to manual segmentation provided by an expert radiologist. Results of lesion classification were compared to histological findings. Our results indicate that Multi Layer Perceptron can achieve better results in terms of sensitivity, specificity and accuracy when dynamic features are considered both for lesion segmentation and classification (accuracy of 91 % and 70 %, respectively).

AB - The aim of this study is to propose an approach, based on Multi Layer Perceptron classification of dynamic and textural features, for breast lesions segmentation and classification using Dynamic Contrast Enhanced-Magnetic Resonance Imaging data. We compared the performance obtainable with dynamic, textural and spatio-temporal features. In particular, 98 dynamic features, 60 textural features and 72 spatio-temporal features were considered. The dataset included 20 breast lesions, 10 benign and 10 malignant. The performance of lesion segmentation have been evaluated with respect to manual segmentation provided by an expert radiologist. Results of lesion classification were compared to histological findings. Our results indicate that Multi Layer Perceptron can achieve better results in terms of sensitivity, specificity and accuracy when dynamic features are considered both for lesion segmentation and classification (accuracy of 91 % and 70 %, respectively).

UR - http://www.scopus.com/inward/record.url?scp=84867319584&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84867319584&partnerID=8YFLogxK

U2 - 10.1109/CBMS.2012.6266312

DO - 10.1109/CBMS.2012.6266312

M3 - Conference contribution

AN - SCOPUS:84867319584

SN - 9781467320511

BT - Proceedings - IEEE Symposium on Computer-Based Medical Systems

ER -