scholarly journals Development of Detection and Volumetric Methods for the Triceps of the Lower Leg Using Magnetic Resonance Images with Deep Learning

2021 ◽  
Vol 11 (24) ◽  
pp. 12006
Author(s):  
Yusuke Asami ◽  
Takaaki Yoshimura ◽  
Keisuke Manabe ◽  
Tomonari Yamada ◽  
Hiroyuki Sugimori

Purpose: A deep learning technique was used to analyze the triceps surae muscle. The devised interpolation method was used to determine muscle’s volume and verify the usefulness of the method. Materials and Methods: Thirty-eight T1-weighted cross-sectional magnetic resonance images of the triceps of the lower leg were divided into three classes, i.e., gastrocnemius lateralis (GL), gastrocnemius medialis (GM), and soleus (SOL), and the regions of interest (ROIs) were manually defined. The supervised images were classified as per each patient. A total of 1199 images were prepared. Six different datasets separated patient-wise were prepared for K-fold cross-validation. A network model of the DeepLabv3+ was used for training. The images generated by the created model were divided as per each patient and classified into each muscle types. The model performance and the interpolation method were evaluated by calculating the Dice similarity coefficient (DSC) and error rates of the volume of the predicted and interpolated images, respectively. Results: The mean DSCs for the predicted images were >0.81 for GM and SOL and 0.71 for GL. The mean error rates for volume were approximately 11% for GL, SOL, and total error and 23% for GL. DSCs in the interpolated images were >0.8 for all muscles. The mean error rates of volume were <10% for GL, SOL, and total error and 18% for GM. There was no significant difference between the volumes obtained from the supervised images and interpolated images. Conclusions: Using the semantic segmentation of the deep learning technique, the triceps were detected with high accuracy and the interpolation method used in this study to find the volume was useful.

2021 ◽  
Vol 77 ◽  
pp. 180-185
Author(s):  
Hiroyuki Sugimori ◽  
Hiroyuki Hamaguchi ◽  
Taro Fujiwara ◽  
Kinya Ishizaka

2020 ◽  
Vol 7 (1) ◽  
Author(s):  
Manan Binth Taj Noor ◽  
Nusrat Zerin Zenia ◽  
M Shamim Kaiser ◽  
Shamim Al Mamun ◽  
Mufti Mahmud

Abstract Neuroimaging, in particular magnetic resonance imaging (MRI), has been playing an important role in understanding brain functionalities and its disorders during the last couple of decades. These cutting-edge MRI scans, supported by high-performance computational tools and novel ML techniques, have opened up possibilities to unprecedentedly identify neurological disorders. However, similarities in disease phenotypes make it very difficult to detect such disorders accurately from the acquired neuroimaging data. This article critically examines and compares performances of the existing deep learning (DL)-based methods to detect neurological disorders—focusing on Alzheimer’s disease, Parkinson’s disease and schizophrenia—from MRI data acquired using different modalities including functional and structural MRI. The comparative performance analysis of various DL architectures across different disorders and imaging modalities suggests that the Convolutional Neural Network outperforms other methods in detecting neurological disorders. Towards the end, a number of current research challenges are indicated and some possible future research directions are provided.


2021 ◽  
Vol 20 (1) ◽  
pp. 50-54
Author(s):  
Thyago Guirelle Silva ◽  
Rodrigo Augusto do Amaral ◽  
Raphael Rezende Pratali ◽  
Luiz Pimenta

ABSTRACT Objective: To verify the effectiveness of indirect decompression after lateral access fusion in patients with high pelvic incidence. Methods: A retrospective, non-comparative, non-randomized analysis of 22 patients with high pelvic incidence who underwent lateral access fusion, 11 of whom were male and 11 female, with a mean age of 63 years (52-74), was conducted. Magnetic resonance exams were performed within one year after surgery. The cross-sectional area of the thecal sac, anterior and posterior disc heights, and bilateral foramen heights, measured pre- and postoperatively in axial and sagittal magnetic resonance images, were analyzed. The sagittal alignment parameters were measured using simple radiographs. The clinical results were evaluated using the ODI and VAS (back and lower limbs) questionnaires. Results: In all cases, the technique was performed successfully without neural complications. The mean cross-sectional area increased from 126.5 mm preoperatively to 174.3 mm postoperatively. The mean anterior disc height increased from 9.4 mm preoperatively to 12.8 mm postoperatively, while the posterior disc height increased from 6.3 mm preoperatively to 8.1 mm postoperatively. The mean height of the right foramen increased from 157.3 mm in the preoperative period to 171.2 mm in the postoperative period and that of the left foramen increased from 139.3 mm in the preoperative to 158.9 mm in the postoperative. Conclusions: This technique is capable of correcting misalignment in spinal deformity, achieving fusion and promoting the decompression of neural elements. Level of evidence III; Retrospective study.


2020 ◽  
Vol 10 (1) ◽  
Author(s):  
Aniket A. Tolpadi ◽  
Jinhee J. Lee ◽  
Valentina Pedoia ◽  
Sharmila Majumdar

Author(s):  
Vitoantonio Bevilacqua ◽  
Antonio Brunetti ◽  
Giacomo Donato Cascarano ◽  
Andrea Guerriero ◽  
Francesco Pesce ◽  
...  

Abstract Background The automatic segmentation of kidneys in medical images is not a trivial task when the subjects undergoing the medical examination are affected by Autosomal Dominant Polycystic Kidney Disease (ADPKD). Several works dealing with the segmentation of Computed Tomography images from pathological subjects were proposed, showing high invasiveness of the examination or requiring interaction by the user for performing the segmentation of the images. In this work, we propose a fully-automated approach for the segmentation of Magnetic Resonance images, both reducing the invasiveness of the acquisition device and not requiring any interaction by the users for the segmentation of the images. Methods Two different approaches are proposed based on Deep Learning architectures using Convolutional Neural Networks (CNN) for the semantic segmentation of images, without needing to extract any hand-crafted features. In details, the first approach performs the automatic segmentation of images without any procedure for pre-processing the input. Conversely, the second approach performs a two-steps classification strategy: a first CNN automatically detects Regions Of Interest (ROIs); a subsequent classifier performs the semantic segmentation on the ROIs previously extracted. Results Results show that even though the detection of ROIs shows an overall high number of false positives, the subsequent semantic segmentation on the extracted ROIs allows achieving high performance in terms of mean Accuracy. However, the segmentation of the entire images input to the network remains the most accurate and reliable approach showing better performance than the previous approach. Conclusion The obtained results show that both the investigated approaches are reliable for the semantic segmentation of polycystic kidneys since both the strategies reach an Accuracy higher than 85%. Also, both the investigated methodologies show performances comparable and consistent with other approaches found in literature working on images from different sources, reducing both the invasiveness of the analyses and the interaction needed by the users for performing the segmentation task.


PLoS ONE ◽  
2018 ◽  
Vol 13 (9) ◽  
pp. e0204071 ◽  
Author(s):  
Andrew T. Grainger ◽  
Nicholas J. Tustison ◽  
Kun Qing ◽  
Rene Roy ◽  
Stuart S. Berr ◽  
...  

Sign in / Sign up

Export Citation Format

Share Document