Search

Showing total 2 results
Start Over Topic article Journal bmc medical imaging
2 results

1. SEA-NET: medical image segmentation network based on spiral squeeze-and-excitation and attention modules.

Author

Xiong, Liangli, Yi, Chen, Xiong, Qiliang, and Jiang, Shaofeng

Subjects
IMAGE segmentation, DIAGNOSTIC imaging, IMAGE analysis
Abstract

Background: Medical image segmentation is an important processing step in most of medical image analysis. Thus, high accuracy and robustness are required for them. The current deep neural network based medical segmentation methods have good effect on image with balanced foreground and background, but it will loss the characteristics of small targets on image with imbalanced foreground and background after multiple convolutions. Methods: In order to retain the features of small targets in the deep network, we proposed a new medical image segmentation model based on the U-Net with squeeze-and-excitation and attention modules which form a spiral closed path,callled as Spiral Squeeze-and-Excitation and Attention NET (SEA-NET) in this paper. The segmentation model used squeeze-and-extraction modules to adjust the channel information to enhance the useful information and used attention modules to adjust the spatial information of the feature map to highlight the target area for small target segmentation when up-sampling. The deep semantic information is integrated into the shallow feature map by the attention model. Therefore, the deep semantic information cannot be scattered by continuous up-sampling. We used cross entropy loss + Tversky loss function for fast convergence and well processing the imbalanced data sets. Our proposed SEA-NET was tested on the brain MRI dataset LPBA40 and peripheral blood smear images. Conclusions: On brain MRI data, the average value of the Dice coefficient we obtained reached 98.1 % . On the peripheral blood smear dataset, our proposed model has a good segmentation effect on adhesion cells. Results: The experimental results proved that the proposed SEA-Net performed better than U-Net, U-Net++, etc. in medical image segmentation. [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

2. Ultrasound evaluation in combination with finger extension force measurements of the forearm musculus extensor digitorum communis in healthy subjects

Author

Sofia Brorsson, Ylva Aurell, Anna Nilsdotter, Marita Hilliges, and Christer Sollerman

Subjects
Male, Medicin och hälsovetenskap, Medical and Health Sciences, non invasive procedure, human experiment, finger, middle aged, Medicine, forearm, Ultrasonography, adult, Ultrasound, article, Healthy subjects, Anatomy, Middle Aged, Circumference, Forearm, aged, female, medicine.anatomical_structure, lcsh:R855-855.5, muscle mass, Radiology Nuclear Medicine and imaging, muscle force, Female, medicine.symptom, Muscle Contraction, Research Article, Muscle contraction, Adult, lcsh:Medical technology, rehabilitation care, Statistics, Nonparametric, histology, muscle contraction, extensor muscle, male, nonparametric test, Humans, controlled study, Radiology, Nuclear Medicine and imaging, human, normal human, skeletal muscle, Muscle, Skeletal, Aged, business.industry, echography, Skeletal muscle, arm muscle, postoperative care, thickness, physiology, muscle strength, business, Muscle architecture, Extensor Digitorum Communis
Abstract

Background The aim of this study was to evaluate the usefulness of an ultrasound-based method of examining extensor muscle architecture, especially the parameters important for force development. This paper presents the combination of two non-invasive methods for studying the extensor muscle architecture using ultrasound simultaneously with finger extension force measurements. Methods M. extensor digitorum communis (EDC) was examined in 40 healthy subjects, 20 women and 20 men, aged 35–73 years. Ultrasound measurements were made in a relaxed position of the hand as well as in full contraction. Muscle cross-sectional area (CSA), pennation angle and contraction patterns were measured with ultrasound, and muscle volume and fascicle length were also estimated. Finger extension force was measured using a newly developed finger force measurement device. Results The following muscle parameters were determined: CSA, circumference, thickness, pennation angles and changes in shape of the muscle CSA. The mean EDC volume in men was 28.3 cm3 and in women 16.6 cm3. The mean CSA was 2.54 cm2 for men and 1.84 cm2 for women. The mean pennation angle for men was 6.5° and for women 5.5°. The mean muscle thickness for men was 1.2 cm and for women 0.76 cm. The mean fascicle length for men was 7.3 cm and for women 5.0 cm. Significant differences were found between men and women regarding EDC volume (p < 0.001), CSA (p < 0.001), pennation angle (p < 0.05), muscle thickness (p < 0.001), fascicle length (p < 0.001) and finger force (p < 0.001). Changes in the shape of muscle architecture during contraction were more pronounced in men than women (p < 0.01). The mean finger extension force for men was 96.7 N and for women 39.6 N. Muscle parameters related to the extension force differed between men and women. For men the muscle volume and muscle CSA were related to extension force, while for women muscle thickness was related to the extension force. Conclusion Ultrasound is a useful tool for studying muscle architectures in EDC. Muscle parameters of importance for force development were identified. Knowledge concerning the correlation between muscle dynamics and force is of importance for the development of new hand training programmes and rehabilitation after surgery.

Published
2008
Full Text
View/download PDF