scholarly journals Fully Automatic Segmentation and Landmarking of Hip CT Images

10.29007/ds5r ◽  
2020 ◽  
Author(s):  
Guoyan Zheng
Keyword(s):  
Automatic Segmentation ◽  
Ct Images ◽  
Generic Models ◽  
Mean Error ◽  
Segmentation Accuracy ◽  
Fully Automatic ◽  
Ct Data ◽  
Two Stages ◽  
Atlas Segmentation ◽  
Acetabular Rim

We present a fully automatic method of segmenting and landmarking hip CT images for planning of Total Hip Arthroplasty (THA). Our method consists of two stages, i.e., the segmentation stage and the landmarking stage. At the segmentation stage, a multi-atlas segmentation constrained graph method is employed to fully automatically segment both the pelvis and the bilateral proximal femurs from the input CT data. The segmentation stage is followed by the landmarking stage, where a set of pre-defined landmarks are transferred from generic models of the associated hip structures to the input CT space via non-rigid registrations in order to compute a set of functional parameters that are relevant to planning of THA. Evaluated on 20 hip patients, we computed both the segmentation accuracy and the landmarking accuracy. An average segmentation error of 0.38 ± 0.25 mm and 0.49 ± 0.22 mm was found for the hemi-pelvis and for the proximal femurs, respectively. For 3D landmarking, a mean error of 1.58 ± 0.87 mm and 0.46 ± 0.39 mm was found for the acetabular rim center and the acetabular rim radius, respectively; a mean error of 0.74±0.45o was found for the orientation of the anterior pelvic plane; and a mean error of 3.14 ± 1.90 mm and 2.04 ± 1.61 mm was found for the femoral head center and the femoral offset, respectively.

SU-E-J-50: Fully Automatic Segmentation of Brain Tumor in CT Images

2011 ◽  
Vol 38 (6Part8) ◽  
pp. 3453-3453
Author(s):  
M Gao ◽  
D Wei ◽  
S Chen
Keyword(s):  
Brain Tumor ◽  
Automatic Segmentation ◽  
Ct Images ◽  
Fully Automatic

scholarly journals A Multi-atlas Approach for the Automatic Segmentation of Multiple Structures in Head and Neck CT Images

2016 ◽  
Author(s):  
Antong Chen ◽  
Benoit Dawant
Keyword(s):  
Head And Neck ◽  
Local Level ◽  
Automatic Segmentation ◽  
Ct Images ◽  
Training Dataset ◽  
Parotid Glands ◽  
Label Fusion ◽  
Qualitative And Quantitative ◽  
Segmentation Accuracy ◽  
Multiple Structures

A multi-atlas approach is proposed for the automatic segmentation of nine different structures in a set of head and neck CT images for radiotherapy. The approach takes advantage of a training dataset of 25 images to build average head and neck atlases of high-quality. By registering patient images with the atlases at the global level, structures of interest are aligned approximately in space, which allowed multi-atlas-based segmentations and correlation-based label fusion to be performed at the local level in the following steps. Qualitative and quantitative evaluations are performed on a set of 15 testing images. As shown by the results, mandible, brainstem and parotid glands are segmented accurately (mean volume DSC>0.8). The segmentation accuracy for the optic nerves is also improved over previously reported results (mean DSC above 0.61 compared with 0.52 for previous results).

scholarly journals Automated incisional hernia characterization by non-rigid registration of CT images – a pilot study

2020 ◽  
Vol 6 (3) ◽  
pp. 91-94
Author(s):  
Samuel Voß ◽  
Philipp D. Lösel ◽  
Vincent Heuveline ◽  
Sylvia Saalfeld ◽  
Philipp Berg ◽  
...  
Keyword(s):  
Incisional Hernia ◽  
Clinical Observation ◽  
Valsalva Maneuver ◽  
Incisional Hernia Repair ◽  
Ct Images ◽  
Software Framework ◽  
Maximum Displacement ◽  
Data Set ◽  
Fully Automatic ◽  
Ct Data

AbstractIncisional hernia repair makes use of prosthetic meshes to re-establish a biomechanically stable abdominal wall. Mesh sizing and fixation have been found to be essential for the clinical outcome. Comparative CT images a) under rest versus b) under Valsalva maneuver (exhalation against closed airways) provide useful information for hernia characterization. However, this process incorporates several manual measurements, which led to observer variability. The present study suggests using an image registration approach of the CT data to reliably and reproducibly extract hernia quantities. The routine is implemented in the software framework MATLAB and works fully automatic. After CT data import, slice by slice undergo non-rigid B-spline grid registration. Local displacement and strain are extracted from the transformation field. The qualitative results correspond to the clinical observation. Maximum displacement of 3.5 cm and maximum strain of 25 % are calculated for one patient’s data set. Current approaches do not provide this type of information. Further research will focus on validation and possibilities to include this new kind of knowledge into the design process of prosthetic meshes.

scholarly journals Segmentation of Lung Ct Images using Cascaded Fully Convolutional Neural Networks

2019 ◽  
Vol 8 (2) ◽  
pp. 5472-5474
Keyword(s):  
Neural Networks ◽  
Image Segmentation ◽  
Convolutional Neural Networks ◽  
Automatic Segmentation ◽  
Ct Images ◽  
Fully Convolutional Neural Networks ◽  
Proposal A ◽  
Fully Automatic ◽  
Lung Ct

Interpretation of CT Lung images by the radiologist can be enhanced to a greater extent by automatic segmentation of nodules. The efficiency of this interpretation depends on the completeness and non-ambiguousness of the CT Lung images. Here, a fully automatic cascaded basis was proposed for CT Lung image segmentation. In this proposal a customized FCN was used feature extractions exploration from many visual scales and differentiate anatomy with a thick forecast map. Widespread experimental outcomes demonstrate that this technique can address the incompleteness in boundary and this technique can achieve best accuracy in segmentation of Lung CT Images when compared to other techniques which address the same area

scholarly journals A Context Based Automated System for Lung Nodule Detection in CT Images

2021 ◽  
Author(s):  
Maciej Dajnowiec
Keyword(s):  
Lung Nodule ◽  
Region Growing ◽  
Ct Images ◽  
Automated System ◽  
Data Sets ◽  
Lung Nodule Detection ◽  
Nodule Detection ◽  
Testing Data ◽  
Fully Automatic ◽  
Ct Data

This thesis is focused on automatic lung nodule detection in CT images. CAD systems are suited for this tak because the sheer volume of information present in CT data sets is overwhelming for radiologists to process. The system developed in this thesis presents a fully automatic solution that applies a sequential algoriths which strongly focuses on nodule context. The system operates at a rate of 80% sensitivity with 3.05 FPs per slice. Our testing data, consisting of 19 CTdata sets containing239 lung nodules, is extremely robust when compared with other documented systems. In addition it introduces many new approaches such as a tight bounding, vessel connectivity, perimeter analysis, adaptive MLT and region growing based lung segmentation. The experimental results produced by this systemare an affirmation of the competitiveness of its performance when compared to other documented approaches.

Sign in / Sign up

Export Citation Format

Share Document