Comparing of two dimensional and three dimensional fully convolutional networks for radiotherapy dose prediction in left-sided breast cancer

2021 ◽  
Vol 104 (3) ◽  
pp. 003685042110381
Author(s):  
Xue Bai ◽  
Ze Liu ◽  
Jie Zhang ◽  
Shengye Wang ◽  
Qing Hou ◽  
...  
Keyword(s):  
Three Dimensional ◽  
Region Of Interest ◽  
Dice Similarity Coefficient ◽  
Two Dimensional ◽  
Dimensional Model ◽  
Dose Distributions ◽  
Convolutional Networks ◽  
Dose Prediction ◽  
Fully Convolutional Networks ◽  
Percentage Difference

Fully convolutional networks were developed for predicting optimal dose distributions for patients with left-sided breast cancer and compared the prediction accuracy between two-dimensional and three-dimensional networks. Sixty cases treated with volumetric modulated arc radiotherapy were analyzed. Among them, 50 cases were randomly chosen to conform the training set, and the remaining 10 were to construct the test set. Two U-Net fully convolutional networks predicted the dose distributions, with two-dimensional and three-dimensional convolution kernels, respectively. Computed tomography images, delineated regions of interest, or their combination were considered as input data. The accuracy of predicted results was evaluated against the clinical dose. Most types of input data retrieved a similar dose to the ground truth for organs at risk ( p > 0.05). Overall, the two-dimensional model had higher performance than the three-dimensional model ( p < 0.05). Moreover, the two-dimensional region of interest input provided the best prediction results regarding the planning target volume mean percentage difference (2.40 ± 0.18%), heart mean percentage difference (4.28 ± 2.02%), and the gamma index at 80% of the prescription dose are with tolerances of 3 mm and 3% (0.85 ± 0.03), whereas the two-dimensional combined input provided the best prediction regarding ipsilateral lung mean percentage difference (4.16 ± 1.48%), lung mean percentage difference (2.41 ± 0.95%), spinal cord mean percentage difference (0.67 ± 0.40%), and 80% Dice similarity coefficient (0.94 ± 0.01). Statistically, the two-dimensional combined inputs achieved higher prediction accuracy regarding 80% Dice similarity coefficient than the two-dimensional region of interest input (0.94 ± 0.01 vs 0.92 ± 0.01, p < 0.05). The two-dimensional data model retrieves higher performance than its three-dimensional counterpart for dose prediction, especially when using region of interest and combined inputs.

Clinical experience of using virtual 3D modelling for pre and intraoperative guidance during robotic-assisted partial nephrectomy

2021 ◽  
pp. 205141582110002
Author(s):  
Lorenz Berger ◽  
Aziz Gulamhusein ◽  
Eoin Hyde ◽  
Matt Gibb ◽  
Teele Kuusk ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Partial Nephrectomy ◽  
Surgical Outcome ◽  
Prospective Cohort ◽  
Three Dimensional ◽  
Two Dimensional ◽  
Dimensional Model ◽  
Three Dimensional Model ◽  
Robotic Assisted ◽  
Robotic Assisted Partial Nephrectomy

Objective: Surgical planning for robotic-assisted partial nephrectomy is widely performed using two-dimensional computed tomography images. It is unclear to what extent two-dimensional images fully simulate surgical anatomy and case complexity. To overcome these limitations, software has been developed to reconstruct three-dimensional models from computed tomography data. We present the results of a feasibility study, to explore the role and practicality of virtual three-dimensional modelling (by Innersight Labs) in the context of surgical utility for preoperative and intraoperative use, as well as improving patient involvement. Methods: A prospective study was conducted on patients undergoing robotic-assisted partial nephrectomy at our high volume kidney cancer centre. Approval from a research ethics committee was obtained. Patient demographics and tumour characteristics were collected. Surgical outcome measures were recorded. The value of the three-dimensional model to the surgeon and patient was assessed using a survey. The prospective cohort was compared against a retrospective cohort and cases were individually matched using RENAL (radius, exophytic/endophytic, nearness to collecting system or sinus, anterior/posterior, location relative to polar lines) scores. Results: This study included 22 patients. Three-dimensional modelling was found to be safe for this prospective cohort and resulted in good surgical outcome measures. The mean (standard deviation) console time was 158.6 (35) min and warm ischaemia time was 17.3 (6.3) min. The median (interquartile range) estimated blood loss was 125 (50–237.5) ml. Two procedures were converted to radical nephrectomy due to the risk of positive margins during resection. The median (interquartile range) length of stay was 2 (2–3) days. No postoperative complications were noted and all patients had negative surgical margins. Patients reported improved understanding of their procedure using the three-dimensional model. Conclusion: This study shows the potential benefit of three-dimensional modelling technology with positive uptake from surgeons and patients. Benefits are improved perception of vascular anatomy and resection approach, and procedure understanding by patients. A randomised controlled trial is needed to evaluate the technology further. Level of evidence: 2b

Design of finite-time guidance law based on observer and head-pursuit theory

2021 ◽  
pp. 095441002098456
Author(s):  
Chenqi Zhu
Keyword(s):  
Finite Time ◽  
Three Dimensional ◽  
Hypersonic Vehicle ◽  
Two Dimensional ◽  
Dimensional Model ◽  
Reaching Law ◽  
Guidance Law ◽  
Fast Power ◽  
Simulation Results ◽  
Guidance Laws

In order to improve the guiding accuracy in intercepting the hypersonic vehicle, this article presents a finite-time guidance law based on the observer and head-pursuit theory. First, based on a two-dimensional model between the interceptor and target, this study applies the fast power reaching law to head-pursuit guidance law so that it can alleviate the chattering phenomenon and ensure the convergence speed. Second, target maneuvers are considered as system disturbances, and the head-pursuit guidance law based on an observer is proposed. Furthermore, this method is extended to a three-dimensional case. Finally, comparative simulation results further verify the superiority of the guidance laws designed in this article.

Mine Impact Burial Prediction From One to Three Dimensions

2008 ◽  
Vol 62 (1) ◽  
Author(s):  
Peter C. Chu
Keyword(s):  
Three Dimensional ◽  
Time History ◽  
Three Dimensions ◽  
Vertical Position ◽  
Burial Depth ◽  
Prediction Errors ◽  
Two Dimensional ◽  
Dimensional Model ◽  
One Dimensional ◽  
Dimensional Models

The Navy’s mine impact burial prediction model creates a time history of a cylindrical or a noncylindrical mine as it falls through air, water, and sediment. The output of the model is the predicted mine trajectory in air and water columns, burial depth/orientation in sediment, as well as height, area, and volume protruding. Model inputs consist of parameters of environment, mine characteristics, and initial release. This paper reviews near three decades’ effort on model development from one to three dimensions: (1) one-dimensional models predict the vertical position of the mine’s center of mass (COM) with the assumption of constant falling angle, (2) two-dimensional models predict the COM position in the (x,z) plane and the rotation around the y-axis, and (3) three-dimensional models predict the COM position in the (x,y,z) space and the rotation around the x-, y-, and z-axes. These models are verified using the data collected from mine impact burial experiments. The one-dimensional model only solves one momentum equation (in the z-direction). It cannot predict the mine trajectory and burial depth well. The two-dimensional model restricts the mine motion in the (x,z) plane (which requires motionless for the environmental fluids) and uses incorrect drag coefficients and inaccurate sediment dynamics. The prediction errors are large in the mine trajectory and burial depth prediction (six to ten times larger than the observed depth in sand bottom of the Monterey Bay). The three-dimensional model predicts the trajectory and burial depth relatively well for cylindrical, near-cylindrical mines, and operational mines such as Manta and Rockan mines.

scholarly journals An experimental and theoretical investigation of particle–wall impacts in a T-junction

2013 ◽  
Vol 727 ◽  
pp. 236-255 ◽  
Author(s):  
D. Vigolo ◽  
I. M. Griffiths ◽  
S. Radl ◽  
H. A. Stone
Keyword(s):  
Flow Direction ◽  
Three Dimensional ◽  
Stokes Number ◽  
Two Dimensional ◽  
Dimensional Model ◽  
Viscous Boundary Layer ◽  
Particle Tracing ◽  
Two Dimensional Model ◽  
Viscous Boundary ◽  
The Impact

AbstractUnderstanding the behaviour of particles entrained in a fluid flow upon changes in flow direction is crucial in problems where particle inertia is important, such as the erosion process in pipe bends. We present results on the impact of particles in a T-shaped channel in the laminar–turbulent transitional regime. The impacting event for a given system is described in terms of the Reynolds number and the particle Stokes number. Experimental results for the impact are compared with the trajectories predicted by theoretical particle-tracing models for a range of configurations to determine the role of the viscous boundary layer in retarding the particles and reducing the rate of collision with the substrate. In particular, a two-dimensional model based on a stagnation-point flow is used together with three-dimensional numerical simulations. We show how the simple two-dimensional model provides a tractable way of understanding the general collision behaviour, while more advanced three-dimensional simulations can be helpful in understanding the details of the flow.

scholarly journals A computationally efficient approach to segmentation of the aorta and coronary arteries using deep learning

2021 ◽  
Author(s):  
Wing Keung Cheung ◽  
Robert Bell ◽  
Arjun Nair ◽  
Leon Menezies ◽  
Riyaz Patel ◽  
...  
Keyword(s):  
Deep Learning ◽  
Coronary Arteries ◽  
Automatic Segmentation ◽  
Three Dimensional ◽  
Regions Of Interest ◽  
Dice Similarity Coefficient ◽  
Processing Unit ◽  
Two Dimensional ◽  
Computed Tomography Images ◽  
Graphical Processing

AbstractA fully automatic two-dimensional Unet model is proposed to segment aorta and coronary arteries in computed tomography images. Two models are trained to segment two regions of interest, (1) the aorta and the coronary arteries or (2) the coronary arteries alone. Our method achieves 91.20% and 88.80% dice similarity coefficient accuracy on regions of interest 1 and 2 respectively. Compared with a semi-automatic segmentation method, our model performs better when segmenting the coronary arteries alone. The performance of the proposed method is comparable to existing published two-dimensional or three-dimensional deep learning models. Furthermore, the algorithmic and graphical processing unit memory efficiencies are maintained such that the model can be deployed within hospital computer networks where graphical processing units are typically not available.

Three-Dimensional Dynamic Modeling of the Human Knee to Include Tibio-Femoral and Patello-Femoral Joints

2001 ◽  
Author(s):  
Dumitru Caruntu ◽  
Mohamed Samir Hefzy
Keyword(s):  
Knee Joint ◽  
Mathematical Models ◽  
Dynamic Modeling ◽  
Three Dimensional ◽  
Two Dimensional ◽  
Dimensional Model ◽  
Human Knee ◽  
Two Dimensional Model ◽  
Rigid Contact ◽  
Locomotor Activities

Abstract Most of the anatomical mathematical models that have been developed to study the human knee are either for the tibio-femoral joint (TFJ) or patello-femoral joint (PFJ). Also, most of these models are static or quasistatic, and therefore do not predict the effects of dynamic inertial loads, which occur in many locomotor activities. The only dynamic anatomical model that includes both joints is a two-dimensional model by Tumer and Engin [1]. The model by Abdel-Rahman and Hefzy [2] is the only three dimensional dynamic model for the knee joint available in the literature; yet, it includes only the TFJ and allows only for rigid contact.

Sign in / Sign up

Export Citation Format

Share Document