Loading [a11y]/accessibility-menu.js
Analysis of Breast Thermograms in Lateral Views using Texture Features | IEEE Conference Publication | IEEE Xplore
Scheduled Maintenance: On Tuesday, 25 February, IEEE Xplore will undergo scheduled maintenance from 1:00-5:00 PM ET (1800-2200 UTC). During this time, there may be intermittent impact on performance. We apologize for any inconvenience.

Analysis of Breast Thermograms in Lateral Views using Texture Features


Abstract:

Breast cancer is one of the major concerns for highest mortality rate among women in India. The gold standard imaging modality such as mammography fails to detect the bre...Show More

Abstract:

Breast cancer is one of the major concerns for highest mortality rate among women in India. The gold standard imaging modality such as mammography fails to detect the breast abnormalities in denser breasts and it can indicate a tumor only after it attains a certain size. Breast thermography which is non-intrusive and non-painful overcomes these limitations. It is proved to be a promising technique in detecting even the onset of tumors through temperature pattern on the breast. This paper illustrates feature-based analysis implemented on lateral view breast thermograms acquired in mass screening camps. The images were analyzed using texture transform based features. The objective of this study is to examine the significance of acquiring the thermograms in lateral views. The lateral views cover the whole breast for thermal analysis which greatly helps the radiologist to investigate the images further. Texture-based Grey Level Co-occurrence Matrix features and discrete wavelet transform based features are extracted to study the frontal and lateral view breast thermograms. The statistical analysis using one way ANOVA and independent t-test indicate significant values that clearly distinguish between normal and abnormal thermograms. These features are then fed into classifiers to validate that lateral view breast thermograms prove to be efficient in detecting the abnormalities better than in frontal views. The classification performance results show higher sensitivity rate of 82%, specificity rate of 83% and accuracy rate of 87% for lateral views when compared to the frontal views.
Date of Conference: 28-31 October 2018
Date Added to IEEE Xplore: 24 February 2019
ISBN Information:

ISSN Information:

Conference Location: Jeju, Korea (South)

References

References is not available for this document.