| 61. |
- Friman, Ola, 1975-, et al.
(författare)
-
Recognizing emphysema - A neural network approach
- 2002
-
Ingår i: Pattern Recognition, 2002. Proceedings. 16th International Conference on (Volume:1). - : IEEE Computer Society. ; , s. 512-515
-
Konferensbidrag (refereegranskat)abstract
- An accurate and fully automatic method for detecting and quantifying emphysema in CT-images is presented. The method is based on an image preprocessing step followed by a neural network classifier trained to separate true emphysema from artifacts. The proposed approach is shown to be superior to an established method when applied on real patient data.
|
|
| 62. |
|
|
| 63. |
- Gharehbaghi, Arash, et al.
(författare)
-
A novel method for discrimination between innocent and pathological heart murmurs
- 2015
-
Ingår i: Medical Engineering and Physics. - : Elsevier. - 1350-4533 .- 1873-4030. ; 37:7, s. 674-682
-
Tidskriftsartikel (refereegranskat)abstract
- This paper presents a novel method for discrimination between innocent and pathological murmurs using the growing time support vector machine (GTSVM). The proposed method is tailored for characterizing innocent murmurs (IM) by putting more emphasis on the early parts of the signal as IMs are often heard in early systolic phase. Individuals with mild to severe aortic stenosis (AS) and IM are the two groups subjected to analysis, taking the normal individuals with no murmur (NM) as the control group. The AS is selected due to the similarity of its murmur to IM, particularly in mild cases. To investigate the effect of the growing time windows, the performance of the GTSVM is compared to that of a conventional support vector machine (SVM), using repeated random sub-sampling method. The mean value of the classification rate/sensitivity is found to be 88%/86% for the GTSVM and 84%/83% for the SVM. The statistical evaluations show that the GTSVM significantly improves performance of the classification as compared to the SVM.
|
|
| 64. |
- Haufe, William, et al.
(författare)
-
Feasibility of an automated tissue segmentation technique in a longitudinal weight loss study
- 2016
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- To address the problems inherent in manual methods, a novel, semi-automated tissue segmentation image analysis technique has been developed. The purpose of this study was to demonstrate the feasibility and describe preliminary observations of applying this technique to quantify and monitor longitudinal changes in abdominal adipose tissue and thigh muscle volume in obese adults during weight loss. Abdominal adipose tissue and thigh muscle volume decreased during weight loss. As a proportion of body weight, adipose tissue volumes decreased during weight loss. By comparison, as a proportion of body weight, thigh muscle volume increased.
|
|
| 65. |
- Johansson, Björn, et al.
(författare)
-
Learning Corner Orientation Using Canonical Correlation
- 2001
-
Ingår i: Proceedings of the SSAB Symposium on Image Analysis. ; , s. 89-92
-
Konferensbidrag (refereegranskat)abstract
- This paper shows how canonical correlation can be used to learn a detector for corner orientation invariant to corner angle and intensity. Pairs of images with the same corner orientation but different angle and intensity are used as training samples. Three different image representations; intensity values, products between intensity values, and local orientation are examined. The last representation gives a well behaved result that is easy to decode into the corner orientation. To reduce dimensionality, parameters from a polynomial model fitted on the different representations is also considered. This reduction did not affect the performance of the system.
|
|
| 66. |
|
|
| 67. |
|
|
| 68. |
- Karlsson, Anette, et al.
(författare)
-
Automatic and quantitative assessment of regional muscle volume by multi-atlas segmentation using whole-body water–fat MRI
- 2015
-
Ingår i: Journal of Magnetic Resonance Imaging. - : John Wiley & Sons. - 1053-1807 .- 1522-2586. ; 41:6, s. 1558-1569
-
Tidskriftsartikel (refereegranskat)abstract
- PurposeTo develop and demonstrate a rapid whole-body magnetic resonance imaging (MRI) method for automatic quantification of total and regional skeletal muscle volume.Materials and MethodsThe method was based on a multi-atlas segmentation of intensity corrected water–fat separated image volumes. Automatic lean muscle tissue segmentations were achieved by nonrigid registration of atlas datasets with 10 different manually segmented muscle groups. Ten subjects scanned at 1.5 T and 3.0 T were used as atlases, initial validation and optimization. Further validation used 11 subjects scanned at 3.0 T. The automated and manual segmentations were compared using intraclass correlation, true positive volume fractions, and delta volumes.ResultsFor the 1.5 T datasets, the intraclass correlation, true positive volume fractions (mean ± standard deviation, SD), and delta volumes (mean ± SD) were 0.99, 0.91 ± 0.02, −0.10 ± 0.70L (whole body), 0.99, 0.93 ± 0.02, 0.01 ± 0.07L (left anterior thigh), and 0.98, 0.80 ± 0.07, −0.08 ± 0.15L (left abdomen). The corresponding values at 3.0 T were 0.97, 0.92 ± 0.03, −0.17 ± 1.37L (whole body), 0.99, 0.93 ± 0.03, 0.03 ± 0.08L (left anterior thigh), and 0.89, 0.90 ± 0.04, −0.03 ± 0.42L (left abdomen). The validation datasets showed similar results.ConclusionThe method accurately quantified the whole-body skeletal muscle volume and the volume of separate muscle groups independent of field strength and image resolution.
|
|
| 69. |
- Karlsson, Anette, et al.
(författare)
-
Automatic and Quantitative Assessment of Total and Regional Muscle Tissue Volume using Multi-Atlas Segmentation
- 2015
-
Ingår i: International Society for Magnetic Resonance in Medicince Annual Meeting.
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- The purpose is to develop and demonstrate a rapid whole-body MRI method for automatic quantification of total and regional lean skeletal muscle volume. Quantitative water and fat separated image volumes of the whole body are manually segmented and used as atlases. The atlases are non-rigidly registered onto to a new image volume and the muscle groups are classified using a voting scheme. A leave-one-out approach with subjects scanned in a 1.5 T and a 3.0 T scanner is used for validation. The method quantifies the whole-body skeletal muscle volumes and the volumes of separate muscle groups independently of image resolution.
|
|
| 70. |
- Karlsson, Anette, et al.
(författare)
-
Automatic and Quantitative Assessment of Total and Regional Muscle Tissue Volume using Multi-Atlas Segmentation
- 2014
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- Accurate and precise assessment of human muscle tissue is important for further understanding of different muscle diseases and syndromes. We present a rapid whole body MR method for automatic quantification of total and regional muscle volume. The method is based on multi-atlas segmentation of intensity corrected water-fat separated images. The method was validated with a leave-one-out approach, using manually segmented atlases from 10 subjects as ground truth. The result gave a coefficient of variation on total muscle volume equal to 1.25±1.35 % (mean ± standard deviation). The method enables cost-efficient large-scale studies, investigating conditions such as sarcopenia and muscular dystrophies.
|
|
