Premium
Automated volume analysis of head and neck lesions on CT scans using 3D level set segmentation
Author(s) -
Street Ethan,
Hadjiiski Lubomir,
Sahiner Berkman,
Gujar Sachin,
Ibrahim Mohannad,
Mukherji Suresh K.,
Chan HeangPing
Publication year - 2007
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.2794174
Subject(s) - segmentation , gold standard (test) , artificial intelligence , computer science , metric (unit) , head and neck , medical imaging , data set , computer vision , minimum bounding box , quality assurance , nuclear medicine , pattern recognition (psychology) , medicine , radiology , operations management , external quality assessment , surgery , pathology , economics , image (mathematics)
The authors have developed a semiautomatic system for segmentation of a diverse set of lesions in head and neck CT scans. The system takes as input an approximate bounding box, and uses a multistage level set to perform the final segmentation. A data set consisting of 69 lesions marked on 33 scans from 23 patients was used to evaluate the performance of the system. The contours from automatic segmentation were compared to both 2D and 3D gold standard contours manually drawn by three experienced radiologists. Three performance metric measures were used for the comparison. In addition, a radiologist provided quality ratings on a 1 to 10 scale for all of the automatic segmentations. For this pilot study, the authors observed that the differences between the automatic and gold standard contours were larger than the interobserver differences. However, the system performed comparably to the radiologists, achieving an average area intersection ratio of 85.4 % compared to an average of 91.2 % between two radiologists. The average absolute area error was 21.1 % compared to 10.8 % , and the average 2D distance was 1.38 mm compared to 0.84 mm between the radiologists. In addition, the quality rating data showed that, despite the very lax assumptions made on the lesion characteristics in designing the system, the automatic contours approximated many of the lesions very well.