Abstract
Globally, obesity is on the rise. According to the world health organization over 300 million people in the world are obese. India is the third most obese country in the world. It is noteworthy that 30 million people are obese in India alone. If a person's body weight is 20% higher than normal, it means he is obese. People who are obese are at a much higher risk of developing many severe ailments than normal or healthy weight. Visceral adipose tissue (VAT) quantity and subcutaneous adipose tissue (SAT) (volume) is generally higher for those with obesity. Hence, developing a computer vision-based fully automated visceral and subcutaneous adipose tissue segmentation and quantification system is an urgent research problem. This can be very useful for physicians to predict a wide variety of highly absurd diseases earlier. In this research, a fully automated hybrid deep learning framework has been developed to optimize VAT and SAT segmentation from non-contrast abdominal MRI images. The texture layer is included in traditional convolutional neural network architecture for CNN hybridization to improve classification performance and reduce computational cost. This will reduce the unwanted iteration which takes place during feature learning, which further reduces the computational complexity and unwanted memory requirements of the traditional CNN model. Back-propagation algorithms are generally used to train CNN, which unnecessarily consumes a lot of time to train the model. This research has developed an improved back-propagation algorithm to reduce unwanted training time. The performance of the proposed method is evaluated using mean absolute error (MAE), mean square error (MSE), and root mean square error (RMSE). Experimental results make it clear that the performance of the proposed VAT segmentation system is better than state-of-the-art methods.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Ackermans LLGC et al (2021) Deep learning automated segmentation for muscle and adipose tissue from abdominal computed tomography in polytrauma patients. Sensors (Basel) 6:16–21
Arthur RM, Wang S, Trobaugh JW (2011) Changes in body-surface electrocardiograms from geometric remodeling with obesity. IEEE Trans Biomed Eng 58(6):1565–1573
Cao C (2019) An improved faster R-CNN for small object detection. IEEE Access 7:106838–106846
Chang X et al (2016) Compound rank-projections for bilinear analysis. IEEE Trans Neural Netw Learn Syst 27:1502–1513
Corbeil P et al (2001) Increased risk for falling associated with obesity: mathematical modeling of postural control. IEEE Trans Neural Syst Rehabil Eng 9:126–136
Dabiri S et al (2020) Deep learning method for localization and segmentation of abdominal CT. Comput Med Imaging Graph 85:101776
Ding X et al (2014) Automated epicardial fat volume quantification from non-contrast CT. International Society for Optics and Photonics, p 90340
Ding X et al (2015) Automated pericardium delineation and epicardial fat volume quantification from noncontrast CT. Med Phys 42:5015–5026
Estrada S et al (2019) FatSegNet: a fully automated deep learning pipeline for adipose tissue segmentation on abdominal dixon MRI. Magn Reson Med 83(4):6949410–6949410
Franke J (1993) System characterization of a highly integrated preclinical hybrid MPI-MRI scanner. IEEE Trans Med Imaging 35(9):1993–2004
Hussein S (2017) Automatic segmentation and quantification of white and brown adipose tissues from PET/CT scans. IEEE Trans Med Imaging 36(3):734–744
Hussein S et al (2016) Automatic segmentation and quantification of white and brown adipose tissues from PET/CT scans. IEEE Trans Med Imaging 36(3):6421081
Ibrahim SM et al (2019) A study on heart segmentation using deep learning algorithm for MRI scans. In: 2019 13th International conference on mathematics, actuarial science, pp 1–5
Jiang J et al (2019) Multi-spectral RGB-NIR image classification using double-channel CNN. IEEE Access 7:20607–20613
Khellal A, Ma H, Fei Q (2018) Convolutional neural network features comparison between back-propagation and extreme learning machine. 37th Chinese control conference (CCC), pp 9629–9634
Kitil AO, Kumar M, Gram-Lavale SI (2018) An IoT-based rain alerting and flood prediction using sensors and arduino for smart environment. Int J Pure Appl Math 118(24):1–12
Koitka S et al (2020) Fully automated body composition analysis in routine CT imaging using 3D semantic segmentation convolutional neural networks. Eur Radiol 31(4):7979624
Kucybała I et al (2020) A fast graph-based algorithm for automated segmentation of subcutaneous and visceral adipose tissue in 3D abdominal computed tomography images. Biocybern Biomed Eng 40(2):729–739
Kullberg J (2010) Whole-body MRI for analysis of body composition. In: 2010 IEEE international symposium on biomedical imaging: from nano to macro, pp 1065–1066
Lu L et al (2019) Integrating local CNN and global CNN for script identification in natural scene images. IEEE Access 7:52669–52679
Masoudi S et al (2020) Adipose tissue segmentation in unlabeled abdomen MRI using cross modality domain adaptation. Annu Int Conf IEEE Eng Med Biol Soc, p 33018306
Mizutani H (1994) The back propagation method for CNN. In: Proceedings of IEEE international symposium on circuits and systems—ISCAS ’94, vol 6, pp 463–466
Norlén A et al (2016) Automatic pericardium segmentation and quantification of epicardial fat from computed tomography angiography. J Med Imaging 3(3):34003–034003
Peng Y (2020) FB-CNN: feature fusion-based bilinear CNN for classification of fruit fly image. IEEE Access 8:3987–3995
Ramchoun H et al (2016) Multilayer perceptron: architecture optimization and training. IJIMAI 4(1):26–30
Ren P (2020) A comprehensive survey of neural architecture search: challenges and solutions. ACM Comput Surv 37(4)
Roy SK et al (2020) HybridSN: exploring 3-D-2-D CNN feature hierarchy for hyperspectral image classification. IEEE Geosci Remote Sens Lett 17:277–281
Shen J et al (2016) Automatic segmentation of abdominal organs and adipose tissue compartments in water-fat MRI: application to weight-loss in obesity. Eur J Radiol 85(9):27501897
Shen N et al (2019) Automated and accurate quantification of subcutaneous and visceral adipose tissue from magnetic resonance imaging based on machine learning. Magn Reson Imaging 64:31004712
Song M (2016) Bridging the semantic gaps of GPU acceleration for scale-out CNN-based big data processing: think big, see small. In: 2016 International conference on parallel architecture and compilation techniques (PACT), pp 315–326
Sussman DL, Yao J, Summers RM (2010) Automated measurement and segmentation of abdominal adipose tissue in MRI. In: 2010 IEEE international symposium on biomedical imaging: from nano to macro, pp 936–939
Wuerslin C et al (2009) Automated segmentation of adipose tissue in magnetic resonance images of the whole body. In: IEEE nuclear science symposium conference record (NSS/MIC), pp 2476–2481
Yan L et al (2015) Capsule robot for obesity treatment with wireless powering and communication. IEEE Trans Ind Electron 62:1125–1133
Yuan D et al (2021) Self-supervised deep correlation tracking. IEEE Trans Image Process 30:976–985
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Devi, B.S., Misbha, D.S. Hybrid convolutional neural network based segmentation of visceral and subcutaneous adipose tissue from abdominal magnetic resonance images. J Ambient Intell Human Comput 14, 13333–13347 (2023). https://doi.org/10.1007/s12652-022-03787-z
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12652-022-03787-z