scholarly journals Automated Segmentation of Median Nerve in Dynamic Sonography Using Deep Learning: Evaluation of Model Performance

Diagnostics ◽  
2021 ◽  
Vol 11 (10) ◽  
pp. 1893
Author(s):  
Chueh-Hung Wu ◽  
Wei-Ting Syu ◽  
Meng-Ting Lin ◽  
Cheng-Liang Yeh ◽  
Mathieu Boudier-Revéret ◽  
...  
Keyword(s):  
Deep Learning ◽  
Median Nerve ◽  
Model Performance ◽  
Ultrasound Images ◽  
Learning Evaluation ◽  
Cross Section Area ◽  
Section Area ◽  
Dynamic Ultrasound ◽  
Tunnel Syndrome ◽  
And Training

There is an emerging trend to employ dynamic sonography in the diagnosis of entrapment neuropathy, which exhibits aberrant spatiotemporal characteristics of the entrapped nerve when adjacent tissues move. However, the manual tracking of the entrapped nerve in consecutive images demands tons of human labors and impedes its popularity clinically. Here we evaluated the performance of automated median nerve segmentation in dynamic sonography using a variety of deep learning models pretrained with ImageNet, including DeepLabV3+, U-Net, FPN, and Mask-R-CNN. Dynamic ultrasound images of the median nerve at across wrist level were acquired from 52 subjects diagnosed as carpal tunnel syndrome when they moved their fingers. The videos of 16 subjects exhibiting diverse appearance and that of the remaining 36 subjects were used for model test and training, respectively. The centroid, circularity, perimeter, and cross section area of the median nerve in individual frame were automatically determined from the inferred nerve. The model performance was evaluated by the score of intersection over union (IoU) between the annotated and model-predicted data. We found that both DeepLabV3+ and Mask R-CNN predicted median nerve the best with averaged IOU scores close to 0.83, which indicates the feasibility of automated median nerve segmentation in dynamic sonography using deep learning.

Usefulness of ultrasound assessment of median nerve mobility in carpal tunnel syndrome

2018 ◽  
Vol 59 (12) ◽  
pp. 1494-1499 ◽  
Author(s):  
Gi-young Park ◽  
Dong Rak Kwon ◽  
Jung Im Seok ◽  
Dong-Soon Park ◽  
Hee Kyung Cho
Keyword(s):  
Carpal Tunnel Syndrome ◽  
Median Nerve ◽  
Carpal Tunnel ◽  
Control Group ◽  
Healthy Individuals ◽  
Cross Sectional ◽  
Dynamic Ultrasound ◽  
Peripheral Compression ◽  
Ultrasound Assessment ◽  
Tunnel Syndrome

Background Carpal tunnel syndrome (CTS) is the most common peripheral compression neuropathy of the upper extremity. Recently, dynamic ultrasound (US) imaging has shown differences in median nerve mobility between the affected and unaffected sides in CTS. Purpose The present study was performed to compare the median nerve mobility between patients with CTS and healthy individuals, and to correlate median nerve mobility with the severity of CTS. Material and Methods A total of 101 patients (128 wrists) with CTS and 43 healthy individuals (70 wrists) were evaluated. Electrodiagnostic studies were initially conducted to determine the neurophysiological grading scale (NGS). The cross-sectional area (CSA) of the median nerve and the grade of median nerve mobility were measured using US. Results The mean grade of median nerve mobility in the CTS group (1.9) was significantly lower than that in the control group (2.6; P < 0.001). There were significant negative correlations between the grade of median nerve mobility and distal motor latency of the median nerve (r = –0.218, P = 0.015), NGS (r = –0.207, P = 0.020) and CSA of the median nerve (r = –0.196, P = 0.028). Conclusion The grade of median nerve mobility was negatively correlated with the severity of CTS. US assessment of median nerve mobility may be useful in diagnosing and determining the severity of CTS.

Learning-Based Median Nerve Segmentation From Ultrasound Images For Carpal Tunnel Syndrome Evaluation

2021 ◽  
Author(s):  
Mariachiara Di Cosmo ◽  
Maria Chiara Fiorentino ◽  
Francesca Pia Villani ◽  
Gianmarco Sartini ◽  
Gianluca Smerilli ◽  
...  
Keyword(s):  
Carpal Tunnel Syndrome ◽  
Median Nerve ◽  
Carpal Tunnel ◽  
Ultrasound Images ◽  
Tunnel Syndrome

scholarly journals DPlis: Boosting Utility of Differentially Private Deep Learning via Randomized Smoothing

2021 ◽  
Vol 2021 (4) ◽  
pp. 163-183
Author(s):  
Wenxiao Wang ◽  
Tianhao Wang ◽  
Lun Wang ◽  
Nanqing Luo ◽  
Pan Zhou ◽  
...  
Keyword(s):  
Deep Learning ◽  
Private Information ◽  
Model Performance ◽  
Training Data ◽  
Model Parameters ◽  
Model Quality ◽  
Learning Techniques ◽  
Core Idea ◽  
Privacy Budget ◽  
And Training

Abstract Deep learning techniques have achieved remarkable performance in wide-ranging tasks. However, when trained on privacy-sensitive datasets, the model parameters may expose private information in training data. Prior attempts for differentially private training, although offering rigorous privacy guarantees, lead to much lower model performance than the non-private ones. Besides, different runs of the same training algorithm produce models with large performance variance. To address these issues, we propose DPlis– Differentially Private Learning wIth Smoothing. The core idea of DPlis is to construct a smooth loss function that favors noise-resilient models lying in large flat regions of the loss landscape. We provide theoretical justification for the utility improvements of DPlis. Extensive experiments also demonstrate that DPlis can effectively boost model quality and training stability under a given privacy budget.

scholarly journals Schwannoma of the median nerve mimicking carpal tunnel syndrome in a pregnant patient. Case report.

2016 ◽  
Vol 18 (4) ◽  
pp. 521
Author(s):  
Giokits-Kakavouli Giasna ◽  
Mihaela C. Micu ◽  
Romeo Micu
Keyword(s):  
Median Nerve ◽  
Carpal Tunnel ◽  
Imaging Techniques ◽  
Pregnant Patient ◽  
Nerve Lesion ◽  
Cross Section Area ◽  
Section Area ◽  
The Us ◽  
Magnetic Resonance Imaging Mri ◽  
The Right

In patients with symptoms of a peripheral neuropathy especially during pregnancy, use of imaging techniques such as Ultrasound (US) and Magnetic Resonance Imaging (MRI) may be essential for the diagnostic accomplishment. A 30-weekspregnant diabetic female attending US evaluation due to intermittent hand pain, numbness, and weakness bilaterally. Although, the US evaluation revealed the median nerve (MN) normal size, echogenicity and echo-texture within the right carpal tunnel; the US assessment applied proximally to the carpal tunnel, revealed a hypoechoic tumor-like mass and increased MN cross section area. In transverse view, the MN was detected as an eccentric, hypoechoic structure compressed by the aforementioned mass. A presence of MN schwannoma or neurofibroma was suspected. US has been proved to be extremely useful to determine location, extent as well as the type of nerve lesion. 

scholarly journals Quantitative Evaluation of the Echo Intensity of Paraneural Area and Myofascial Structure around Median Nerve in Carpal Tunnel Syndrome

Diagnostics ◽  
2020 ◽  
Vol 10 (11) ◽  
pp. 914
Author(s):  
Chenglei Fan ◽  
Caterina Fede ◽  
Carmelo Pirri ◽  
Diego Guidolin ◽  
Carlo Biz ◽  
...  
Keyword(s):  
Carpal Tunnel Syndrome ◽  
Median Nerve ◽  
Carpal Tunnel ◽  
Ultrasound Images ◽  
Echo Intensity ◽  
Rater Reliability ◽  
Transversal Displacement ◽  
Significant Difference ◽  
Tunnel Syndrome ◽  
And Control

The aim of this study was to investigate whether the echo intensity (EI) of the paraneural area (PA), the median nerve (MN) at the carpal tunnel, the EI of the myofascial structure (MS) around MN, the ‘PA and MN’ at the mid-forearm, and the MN transversal displacement at both sites differs between persons with carpal tunnel syndrome (CTS) and control subjects. Methods: In total, 16 CTS patients and 16 controls, age- and gender-matched, were recruited. Cross-sectional ultrasound images of MN were obtained to evaluate the EI of the PA, the MN at carpal tunnel, the EI of MS, and the ‘PA and MN’ at the mid-forearm in a natural position, then images were taken after a whole-hand grasp movement, to evaluate MN transversal displacement. Inter-rater and intra-rater reliability in control, and differences in the EI and MN displacement between CTS and control, were analyzed. In addition, the correlations between ultrasound parameters and MN displacement were evaluated. Results: The quantitative EI of PA, MN, EI of MS, ‘PA and MN’ had high inter-rater and intra-rater reliability in the control. The EI of PA, MS and ‘PA and MN’ were significantly higher in CTS subjects (p < 0.01), whilst there was no significant difference in the EI of MN at the carpal tunnel. MN displacement was significantly decreased both at the carpal tunnel and the mid-forearm in CTS subjects (p < 0.01). In addition, there were negative correlations among the EI of PA (rs = −0.484, p = 0.004), EI of MS (rs = −0.479, p = 0.002), EI of ‘PA and MN’ (rs = −0.605, p < 0.001) and MN transversal displacement. Conclusions: The higher EI of PA and MS around MN in CTS may indicate greater fibrosis along the course of MN, reducing fascial adaptability, influencing the synergy and coordination of the MS, and increasing the shear stress between MS and MN, and it may further increase the abnormal pressure on the MN not only at the carpal tunnel, but also at the mid-forearm. These results may partly explain the role of PA and MS in CTS pathogenesis.

EFFECT OF BEDLOAD SEDIMENT NATURAL COMPOSITION ON GEOMETRIC AND DINAMIC CHARACTERISTICS OF CHANNEL FORMS

2018 ◽  
pp. 36-39
Author(s):  
N Ikramov ◽  
T Majidov
Keyword(s):  
Research Work ◽  
Laboratory Data ◽  
Hydroelectric Station ◽  
Negative Consequences ◽  
Movement Velocity ◽  
Reservoir Volume ◽  
Pump Station ◽  
Cross Section Area ◽  
Section Area ◽  
Bedload Sediment

The article brings up data on sediment diversity at watercourse bed and on their movement in the form of ridges. The ridge form movement of sediment leads to the reduction of reservoir volume and canal cross section area, which has an effect on their carrying capacity, filling of pump station forechambers and hydroelectric station pressure basins with sediment. The presence of sediment in flow leads to abrasive deterioration of pumps, water motors and pressure pipes and to other negative consequences. Research work tasks on the study of these effects have been examined with the purpose of preventing such negative consequences. On the basis of laboratory data diagrams and relationships were obtained for ridge length, height and movement velocity vs. sediment hydraulic and geometric sizes.

Sign in / Sign up

Export Citation Format

Share Document