Title :
Pattern recognition diagnostics for emphysema using phase only correlation technique
Author :
Purushotham, Thejaswini ; Rao, K.R.
Author_Institution :
Univ. of Texas, Arlington, TX, USA
Abstract :
Emphysema is a lung related disorder. It shows up on the CT (computer tomography) images as regions of low attenuation. Traditional methods used for detection of Emphysema have been derivation of several metrics of the CT image like the PI (pixel index) and BI (bullae index) which denote the percentage of pixels with low attenuation in the CT image. These techniques score low with low-dose CT images. Emphysema diagnosis using the POC (phase only correlation) algorithm is proposed in this paper. POC is a technique which is intensity independent gives a direct mapping between the extent of affected lung and the length of the POC peak.
Keywords :
computerised tomography; correlation methods; lung; medical disorders; medical image processing; pattern recognition; CT images; attenuation; bullae index; computer tomography; emphysema; lung disorder; pattern recognition diagnostics; phase only correlation technique; pixel index; Attenuation; Bismuth; Computed tomography; Image analysis; Lungs; Pattern analysis; Pattern recognition; Pixel; Radiation dosage; USA Councils; Correlation; POC; Pattern Recognition; emphysema;
Conference_Titel :
System Theory (SSST), 2010 42nd Southeastern Symposium on
Conference_Location :
Tyler, TX
Print_ISBN :
978-1-4244-5690-1
DOI :
10.1109/SSST.2010.5442800