Abstract
Median nerve swelling is one of the features of carpal tunnel syndrome (CTS), and
ultrasound measurement of maximum median nerve cross-sectional area is commonly used
to diagnose CTS. We hypothesized that volume might be a more sensitive measure than
cross-sectional area for CTS diagnosis. We therefore assessed the accuracy and reliability
of 3-D volume measurements of the median nerve in human cadavers, comparing direct
measurements with ultrasound images interpreted using deep learning algorithms. Ultrasound
images of a 10-cm segment of the median nerve were used to train the U-Net model,
which achieved an average volume similarity of 0.89 and area under the curve of 0.90
from the threefold cross-validation. Correlation coefficients were calculated using
the areas measured by each method. The intraclass correlation coefficient was 0.86.
Pearson's correlation coefficient R between the estimated volume from the manually measured cross-sectional area and
the estimated volume of deep learning was 0.85. In this study using deep learning
to segment the median nerve longitudinally, estimated volume had high reliability.
We plan to assess its clinical usefulness in future clinical studies. The volume of
the median nerve may provide useful additional information on disease severity, beyond
maximum cross-sectional area.
KeyWord
To read this article in full you will need to make a payment
Purchase one-time access:
Academic & Personal: 24 hour online accessCorporate R&D Professionals: 24 hour online accessOne-time access price info
- For academic or personal research use, select 'Academic and Personal'
- For corporate R&D use, select 'Corporate R&D Professionals'
Subscribe:
Subscribe to Ultrasound in Medicine and BiologyAlready a print subscriber? Claim online access
Already an online subscriber? Sign in
Register: Create an account
Institutional Access: Sign in to ScienceDirect
References
- Deep learning for brain MRI segmentation: State of the art and future directions.J Digit Imaging. 2017; 30: 449-459
- Extraction of brain tissue from CT head images using fully convolutional neural networks.Proc SPIE 10574, Medical Imaging 2018: Image Processing. 2018; (10574202 March)
- A survey of deep-learning applications in ultrasound: Artificial intelligence-powered ultrasound for improving clinical workflow.J Am Coll Radiol. 2019; 16: 1318-1328
- Robust brain extraction tool for CT head images.Neurocomputing. 2020; 392: 189-195
- Carpal tunnel syndrome: Role of carpal canal size.Neurology. 1985; 35 (1599-1599)
- Machine learning for medical ultrasound: status, methods, and future opportunities.Abdom Radiol (NY). 2018; 43: 786-799
- Validity and reliability of nerve and muscle ultrasound.Muscle Nerve. 2013; 47: 515-521
- The clinical and cost-effectiveness of corticosteroid injection versus night splints for carpal tunnel syndrome (INSTINCTS trial): An open-label, parallel group, randomised controlled trial.Lancet. 2018; 392: 1423-1433
- Carpal tunnel and median nerve volume changes after tunnel release in patients with the carpal tunnel syndrome: A magnetic resonance imaging (MRI) study.Int Orthop. 2016; 40: 981-987
- Measures of the amount of ecologic association between species.Ecology. 1945; 26: 297-302
- Ultrasound assessment of the median nerve: A biomarker that can help in setting a treat to target approach tailored for carpal tunnel syndrome patients.Springerplus. 2015; 4: 13
- Deep learning and machine learning in imaging: Basic principles.in: Ranschaert ER Morozov S Algra PR Artificial intelligence in medical imaging: Opportunities, applications and risks. Springer, Cham2019: 39-46
- Toolkits and libraries for deep learning.J Digit Imaging. 2017; 30: 400-405
- Machine learning for medical imaging.Radiographics. 2017; 37: 505-515
- Automated segmentation of the median nerve in the carpal tunnel using U-Net.Ultrasound Med Biol. 2021; 47: 1964-1969
- Neurophysiology not required before surgery for typical carpal tunnel syndrome.J Hand Surg Br. 2001; 26: 61-64
- The sensitivity and specificity of ultrasound for the diagnosis of carpal tunnel syndrome: A meta-analysis.Clin Orthop Relat Res. 2011; 469: 1089-1094
- The reliability of ultrasound measurements of the median nerve at the carpal tunnel inlet.J Hand Surg Am. 2015; 40: 1992-1995
- Color and power Doppler US for diagnosing carpal tunnel syndrome and determining its severity: A quantitative image processing method.Radiology. 2011; 261: 499-506
- Inter-rater and intra-rater reliability of sonographic median nerve and wrist measurements.J Med Ultrasound. 2018; 26: 14
- Deep learning techniques for medical image segmentation: Achievements and challenges.J Digit Imaging. 2019; 32: 582-596
- Long short-term memory.Neural Comput. 1997; 9: 1735-1780
- DeepNerve: A new convolutional neural network for the localization and segmentation of the median nerve in ultrasound image sequences.Ultrasound Med Biol. 2020; 46: 2439-2452
- Batch normalization: Accelerating deep network training by reducing internal covariate shift.PMLR. 2015; 37: 448-456
- nnU-Net: A self-configuring method for deep learning-based biomedical image segmentation.Nat Methods. 2021; 18: 203-211
- Diagnosis of carpal tunnel syndrome: The gold standard.Am J Phys Med Rehabil. 1993; 72: 1
- A guideline of selecting and reporting intraclass correlation coefficients for reliability research.J Chiropr Med. 2016; 15: 155-163
- Clinical validation of a commercially available deep learning software for synthetic CT generation for brain.Radiat Oncol. 2021; 16: 66
- 2B-2 phase aberration correction on a 3D ultrasound scanner using RF speckle from moving targets.Proc IEEE Int Ultrason Symp. 2006; : 120-123
- A survey on deep learning in medical image analysis.Med Image Anal. 2017; 42: 60-88
- Deep learning in medical ultrasound analysis: A review.Engineering. 2019; 5: 261-275
- Median nerve ultrasound as a screening tool in carpal tunnel syndrome: Correlation of cross-sectional area measures with electrodiagnostic abnormality.Muscle Nerve. 2012; 46: 871-878
- Hough-CNN: Deep learning for segmentation of deep brain regions in MRI and ultrasound.Comput Vision Image Understanding. 2017; 164: 92-102
- Ultrasonography of carpal tunnel syndrome: Clinical significance and limitations in elderly patients.Intern Med. 2011; 50: 2157-2161
- Structural changes of the carpal tunnel, median nerve and flexor tendons in MRI before and after endoscopic carpal tunnel release.Hand Surg. 2014; 19: 193-198
- Quantitative MR imaging of carpal tunnel syndrome.AJR Am J Roentgenol. 1999; 172: 1581-1586
- Median nerve movement in the carpal tunnel before and after carpal tunnel release using transverse ultrasound.J Orthop Surg (Hong Kong). 2017; 252309499017730422
- Usefulness of clinical findings, nerve conduction studies and ultrasonography to predict response to surgical release in idiopathic carpal tunnel syndrome.Clin Exp Rheumatol. 2009; 27: 786-793
- Machine learning-based approach for disease severity classification of carpal tunnel syndrome.Sci Rep. 2021; 11: 17464
- Learning video object segmentation from static images.in: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2017: 3491-3500
- RIL-Contour: A medical imaging dataset annotation tool for and with deep learning.J Digit Imaging. 2019; 32: 571-581
- Carpal tunnel syndrome: Morphologic changes after release of the transverse carpal ligament.J Hand Surg Am. 1989; 14: 852-857
- 2015 U-Net: Convolutional networks for biomedical image segmentation.in: Navab N Hornegger J Wells WM Frangi AF Medical Image Computing and Computer-Assisted Intervention—MICCAI 2015. Springer, Cham2015: 234-241
- Relative motion of the connective tissue in carpal tunnel syndrome: The relation with disease severity and clinical outcome.Ultrasound Med Biol. 2020; 46: 2236-2244
- Nerve conduction studies and EMG in carpal tunnel syndrome: Do they add value?.Clin Neurophysiol Pract. 2018; 3: 78-88
- Dropout: A simple way to prevent neural networks from overfitting.J Mach Learn Res. 2014; 15: 1929-1958
- Metrics for evaluating 3D medical image segmentation: Analysis, selection, and tool.BMC Med Imaging. 2015; 15: 29
- The carpal-tunnel syndrome: A clinical and anatomical study.J Bone Joint Surg Am. 1959; (41-A:626–634)
- Interpretation of the correlation coefficient: A basic review.J Diagn Med Sonogr. 1990; 6: 35-39
- Recent advances in the understanding and management of carpal tunnel syndrome: A comprehensive review.Curr Pain Headache Rep. 2019; 23: 70
- A semi-automatic 3D ultrasound reconstruction method to assess the true severity of adolescent idiopathic scoliosis.Med Bio Eng Comput. 2019; 57: 2115-2128
- MNT-DeepSL: Median nerve tracking from carpal tunnel ultrasound images with deep similarity learning and analysis on continuous wrist motions.Comput Med Imaging Graph. 2020; 80101687
- Quantifying test-retest reliability using the intraclass correlation coefficient and the SEM.J Strength Cond Res. 2005; 19: 231-240
- Carpal tunnel syndrome: Pathophysiology and clinical neurophysiology.Clin Neurophysiol. 2002; 113: 1373-1381
- Automated abdominal segmentation of CT scans for body composition analysis using deep learning.Radiology. 2019; 290: 669-679
- Carpal tunnel syndrome: Diagnostic usefulness of sonography.Radiology. 2004; 232: 93-99
- Automated segmentation of median nerve in dynamic sonography using deep learning: Evaluation of model performance.Diagnostics (Basel). 2021; 11: 1893
- Deep convolutional neural network based medical image classification for disease diagnosis.J Big Data. 2019; 6: 113
- The role of ultrasonographic measurements of the median nerve in the diagnosis of carpal tunnel syndrome.Clin Radiol. 2004; 59: 910-915
- Diagnostic value of sonography in patients with suspected carpal tunnel syndrome: A prospective study.Arthritis Rheum. 2005; 52: 304-311
Article info
Publication history
Published online: August 10, 2022
Accepted:
June 15,
2022
Received in revised form:
June 14,
2022
Received:
December 27,
2021
Identification
Copyright
© 2022 World Federation for Ultrasound in Medicine & Biology. All rights reserved.
