Arrieta, Cristobal, Urrutia, Julio, Besa, Pablo, Montalba, Cristian, Lafont, Nelson, Andia, Marcelo E., and Uribe, Sergio
Health Informatics, Signal Processing, Healthy volunteers, Limits of agreement, Fat infiltration, Nuclear medicine, business.industry, business, Reproducibility, Medicine, Manual segmentation, T2 weighted, Paraspinal Muscle, and Thresholding
Fat infiltration of paraspinal muscles has been related with low back pain and quantified using T2w MR images and manual segmentation techniques. This methodology is time consuming and has low reproducibility. Moreover, the accuracy of T2w images to quantify fat has not been validated. This paper presents the development and validation of an OsiriX application to semi-automatically segment infiltrated fat on T2w images. This software was also utilized to validate the quantification of muscle fat infiltration with T2w images, considering Dixon fat images assessments as a gold standard. Two databases have been considered: 1)T2w images of 37 patients for comparison between manual and semi-automatic segmentations; 2)T2w and Dixon fat images of 10 healthy volunteers and 10 patients to validate the use of T2w images. The OsiriX application was based on automatic thresholding techniques. The fat infiltration fraction was measured in 4 muscle groups (erector spinae and multifidus) and in 5 cross-sectional areas (from L1-L2 to L5-S1) for both databases. From the 37 patients, we randomly selected 15 patients to assess reproducibility of the measurements. Manual and automatic fat quantifications using T2w images showed a correlation of 0.86, no statistical differences, bias of 0.05 % of the muscle area, and limits of agreement of [-12.52, 12.63]% of muscle area. Fat quantifications using Dixon and T2w images showed correlation of 0.94, no statistical differences, bias of 0.38 % of the muscle area, and limits of agreement of [-8.64, 9.41]% of the muscle area. T2w images segmented with this new OsiriX application are reliable and accurate for quantifying fat infiltration in paraspinal muscles.
Pinto, Jose Miguel, Arrieta, Cristobal, Andia, Marcelo E., Uribe, Sergio, Ramos-Grez, Jorge, Vargas, Alex, Irarrazaval, Pablo, and Tejos, Cristian
Biomedical Engineering, Biophysics, Image processing, Segmentation, Prosthesis design, Surgical planning, Critical factors, Image acquisition, Engineering drawing, Engineering, business.industry, business, Building process, Data mining, computer.software_genre, computer, and Triangulation (social science)
Additive manufacturing (AM) models are used in medical applications for surgical planning, prosthesis design and teaching. For these applications, the accuracy of the AM models is essential. Unfortunately, this accuracy is compromised due to errors introduced by each of the building steps: image acquisition, segmentation, triangulation, printing and infiltration. However, the contribution of each step to the final error remains unclear. We performed a sensitivity analysis comparing errors obtained from a reference with those obtained modifying parameters of each building step. Our analysis considered global indexes to evaluate the overall error, and local indexes to show how this error is distributed along the surface of the AM models. Our results show that the standard building process tends to overestimate the AM models, i.e. models are larger than the original structures. They also show that the triangulation resolution and the segmentation threshold are critical factors, and that the errors are concentrated at regions with high curvatures. Errors could be reduced choosing better triangulation and printing resolutions, but there is an important need for modifying some of the standard building processes, particularly the segmentation algorithms.