Title :
Concordance Analysis of DEXA Data
Author :
Hashemi, Ray R. ; Tyler, Alexander A. ; Childers, Christopher ; Chausmer, Arthur
Author_Institution :
Dept. of Comput. Sci., Armstrong Atlantic State Univ., Savannah, GA
Abstract :
The bone density measurements done by DEXA scanning approach are used in diagnosis of osteoporosis. There are six major sites used for DEXA scanning of each hip (three sites on the left and three sites on the right) including: the femoral neck, the trochanteric site, and Ward´s Triangle with or without corresponding total hip measurements. In the past, two separate researches conducted by the authors to determine whether all six measurements are needed for diagnosis of osteoporosis. The results revealed that none of the six measurements are redundant. The null hypothesis that we investigate in this paper is that having non-redundant sites means the measurements of bone density for these sites are in concordance. The concordance analysis is implemented using a new neural network (CNN). The results reveal that the null hypothesis is true
Keywords :
X-ray absorption; bone; diseases; medical computing; neural nets; patient diagnosis; DEXA scanning; Ward Triangle; bone density measurements; concordance analysis; dual energy X-ray absorptiometry; femoral neck; neural network; osteoporosis diagnosis; trochanteric site; Cellular neural networks; Computer science; Data analysis; Density measurement; Hip; Humans; Informatics; Neck; Neural networks; Osteoporosis; Concordance Neural Network.; Concordance analysis; DEXA scanning; Osteoporosis;
Conference_Titel :
Information Technology, 2007. ITNG '07. Fourth International Conference on
Conference_Location :
Las Vegas, NV
Print_ISBN :
0-7695-2776-0
DOI :
10.1109/ITNG.2007.58