Contour Accuracy Research Articles

Numerical analysis of combustion noise in an atmospheric swirl-stabilized LDI burner through modal decomposition techniques

Combustion noise in gas turbine engines has recently become a relevant source of noise in the aircraft due to the appearance of new burner architectures that are intrinsically more unstable, and the optimization of other conventional noise sources in this mean of transport (e.g., jet, fan, airframe). In this work, a simulation setup for reactive conditions was prepared in the CONVERGE finite-volume package using the detailed chemistry SAGE solver to model the combustion of a benchmark case, which was solved using a LES approach with three different cell base sizes: 8,10,12 mm. A confined liquid-fueled swirl-stabilized burner located at the CORIA Laboratory, France, was used to validate the numerical results with the experimental measurements obtained at this facility. OH-PLIF measurements and PDA results for both phases were used to guarantee the accuracy of the numerical OH contours and the velocity profiles of both phases. These experimental measurements were collected at CORIA. After ensuring the stabilization of the numerical flame, the reactive simulations were extended with some adjustments in the time step to capture the acoustic motion. Several techniques like Fast Fourier Transform (FFT), Proper Orthogonal Decomposition (POD) and Dynamic Mode Decomposition (DMD) were used to analyze these results and confirm the presence of a Precessing Vortex Core (PVC) and a Vortex Breakdown Bubble (VBB) during the coupling of pressure, axial velocity and fuel mass fraction in reactive conditions. Furthermore, the acoustic analysis performed with a Helmholtz solver proved that the second longitudinal mode of the chamber (310 Hz) was present in the pressure signal (300 Hz in the LES calculations) and resonated with the Vortex Breakdown Bubble (VBB). However, this dominant frequency did not appear in the frequency distribution of the OH mass fraction and no feedback interaction between the acoustic and the combustion happened. Thus, only combustion noise was obtained.

Abstract P1-10-12: AI-based cardiac sub-structures segmentation for safer radiotherapy planning

Abstract Purpose: Whilst radiotherapy increases cure rates in breast cancer, lung cancer, among others, it may also involve some cardiac exposure, which in turn may increase the risk of different heart diseases. The heart is a complex anatomical organ that involves many different structures making it difficult to contour cardiac sub-structures reproducibly. Contouring, especially for these cases, suffers from inter- and intra-expert variability while being time consuming. Cardiac atlases have been developed to aid in the delineation of cardiac substructures. However, these methods have many shortcomings, including the inability to overcome variations in patient anatomy. In this study, a deep learning based commercial solution for automatic OAR delineation was trained following international guidelines for heart substructures delineation and tested on an unseen cohort of lung and breast patients to evaluate its clinical acceptability. Methods: ART-Net, a CE-marked, FDA-cleared anatomically preserving deep-learning ensemble architecture for automatic annotation of OAR was evaluated using data of 20 breast/lung patients from 2 centers. Automatic annotation of 27 different structures (Ventricles (left and right), atria (left and right), left ventricle (anterior, apical, inferior, lateral, septal), LAD coronary (mid, proximal, distal, total), circumflex coronary (distal, proximal, total), RCA (distal, mid, proximal, total), coronary sinus, left main coronary artery, ascending aorta, pulmonary arteries, vena cava inferior, vena cava superior and the heart) was performed and submitted to 2 experts across 2 centers for qualitative evaluation. Contours were scored as A/acceptable, B/acceptable after minor corrections, and C/not acceptable for clinical use. To avoid any bias, experts were blind to whether the contour were manually, or AI delineated. The DSC between automatic and manual (ground truth) contours of the heart sub-structures were evaluated and compared with interobserver variability from the literature [1,2] using average and min DSC scores. Results: Automatic contours were generated in a mean time of 0.5s per scan slice. Out of the 27 structures, 20 were considered clinically acceptable in the qualitative study. In the inter-expert variability study, 12 structures passed the test successfully using initial acceptance criterion over an acceptable sample size and 9 other structures demonstrated performances above the minimal threshold of inter-expert variability, sometimes on smaller datasets due to lack of manual data. Overall, 16 structures were included in the final model. 13 structures were considered clinically acceptable in 100% of the cases with AI contours rated at the same level as manual contours. For the other 3 structures (coronary sinus, left main coronary artery and vena cava inferior), the performance of the AI contours was slightly below that of the manual contours (within 3.4% difference), with the least performing structure being the coronary sinus (84% for AI vs 87% manual). Conclusion: We show first results for the evaluation of AI-based auto-contouring tool for annotation of the substructures of the heart. The results show very good clinical acceptance, highlighting the high usability of the commercial tool for cardiac cases and its clinical implementation feasibility. The use of this AI tool can facilitate and accelerate future research studies investigating relationships between substructure doses and cardiac outcomes. Future work will include improvement of the sub-structures (mid, proximal, distal) and a retrospective meta-analysis to assess heart sub-structures degree of importance in terms of toxicity.

Deep learning based automatic contour refinement for inaccurate auto-segmentation in MR-guided adaptive radiotherapy

Objective. Fast and accurate auto-segmentation is essential for magnetic resonance-guided adaptive radiation therapy (MRgART). Deep learning auto-segmentation (DLAS) is not always clinically acceptable, particularly for complex abdominal organs. We previously reported an automatic contour refinement (ACR) solution of using an active contour model (ACM) to partially correct the DLAS contours. This study aims to develop a DL-based ACR model to work in conjunction with ACM-ACR to further improve the contour accuracy. Approach. The DL-ACR model was trained and tested using bowel contours created by an in-house DLAS system from 160 MR sets (76 from MR-simulation and 84 from MR-Linac). The contours were classified into acceptable, minor-error and major-error groups using two approaches of contour quality classification (CQC), based on the AAPM TG-132 recommendation and an in-house classification model, respectively. For the major-error group, DL-ACR was applied subsequently after ACM-ACR to further refine the contours. For the minor-error group, contours were directly corrected by DL-ACR without applying an initial ACM-ACR. The ACR workflow was performed separately for the two CQC methods and was evaluated using contours from 25 image sets as independent testing data. Main results. The best ACR performance was observed in the MR-simulation testing set using CQC by TG-132: (1) for the major-error group, 44% (177/401) were improved to minor-error group and 5% (22/401) became acceptable by applying ACM-ACR; among these 177 contours that shifted from major-error to minor-error with ACM-ACR, DL-ACR further refined 49% (87/177) to acceptable; and overall, 36% (145/401) were improved to minor-error contours, and 30% (119/401) became acceptable after sequentially applying ACM-ACR and DL-ACR; (2) for the minor-error group, 43% (320/750) were improved to acceptable contours using DL-ACR. Significance. The obtained ACR workflow substantially improves the accuracy of DLAS bowel contours, minimizing the manual editing time and accelerating the segmentation process of MRgART.

Ultrasound Prostate Segmentation Using Adaptive Selection Principal Curve and Smooth Mathematical Model.

Accurate prostate segmentation in ultrasound images is crucial for the clinical diagnosis of prostate cancer and for performing image-guided prostate surgery. However, it is challenging to accurately segment the prostate in ultrasound images due to their low signal-to-noise ratio, the low contrast between the prostate and neighboring tissues, and the diffuse or invisible boundaries of the prostate. In this paper, we develop a novel hybrid method for segmentation of the prostate in ultrasound images that generates accurate contours of the prostate from a range of datasets. Our method involves three key steps: (1) application of a principal curve-based method to obtain a data sequence comprising data coordinates and their corresponding projection index; (2) use of the projection index as training input for a fractional-order-based neural network that increases the accuracy of results; and (3) generation of a smooth mathematical map (expressed via the parameters of the neural network) that affords a smooth prostate boundary, which represents the output of the neural network (i.e., optimized vertices) and matches the ground truth contour. Experimental evaluation of our method and several other state-of-the-art segmentation methods on datasets of prostate ultrasound images generated at multiple institutions demonstrated that our method exhibited the best capability. Furthermore, our method is robust as it can be applied to segment prostate ultrasound images obtained at multiple institutions based on various evaluation metrics.

Short communication: Part contour error prediction based on LSTM neural network

Abstract. Machine tools are subject to multiple sources of error during machining, resulting in deviations in the dimensions of the part and a reduction in contour accuracy. This paper proposes a contour error prediction model based on a long short-term memory (LSTM) neural network, taking hexagonal recess machining as an example and considering the power, vibration, and temperature signals that affect the contour error. The experimental data show that the model can accurately predict the contour error of the machined part. A more accurate and robust contour error prediction model can provide data support for online compensation of contour errors.

Mechanism of ceramic slurry light scattering affecting contour accuracy and method of projection plane correction

Digital light processing (DLP) is a relatively mature ceramic additive manufacturing technology widely applied in medical bone implantation, electronic communication, and other fields. However, the size error caused by the light scattering from the complex contour can seriously affect the precision and forming quality of the printed green body. This paper studied the scattering behavior of complex structure ceramics under different exposure energies through two structural parameters: exposure width and contour shape. A formula for excess cure width is presented. The formula will be then applied to simplify the machine learning algorithms. Finally, by inverse compensation for the complex contour, we optimized some pore structures and complex shapes to greatly improve the fidelity of the structure.

Five-Axis Contour Error Control Based on Numerical Control Data

Improving contour accuracy is one of the significant goals of industrial machining. This paper proposes a contour error estimation and compensation algorithm for five-axis computer numerical control (CNC) machine tools based on modified numerical control (NC) codes. The expected path analyzed by NC data and the actual trajectory collected by sensors are spatially mapped by the hidden Markov model (HMM). Next, an evaluation function that hybrids the tool tip position and tool orientation change trend is proposed as the index of contour error estimation. Finally, spatial iterative learning control (ILC) is used to compensate the contour error, and high-precision machining instructions are obtained after multiple iterations. Experiments with different trajectories are performed on a five-axis platform to verify the proposed algorithm’s effectiveness. The results show that the proposed algorithm without using planned trajectories, has the same good control effect as traditional methods, which must know the planning trajectory for simple trajectories. At the same time, the method proposed in this paper has better performance than existing algorithms based on tool tip position nearest principle at sharp corners. In conclusion, on the basis of not depending on the planning trajectories, this method has a better compensation effect for the overall accuracy of trajectories and is easier to implement in industrial applications.

Efficient and Robust Video Object Segmentation Through Isogenous Memory Sampling and Frame Relation Mining.

Recently, memory-based methods have achieved remarkable progress in video object segmentation. However, the segmentation performance is still limited by error accumulation and redundant memory, primarily because of 1) the semantic gap caused by similarity matching and memory reading via heterogeneous key-value encoding; 2) the continuously growing and inaccurate memory through directly storing unreliable predictions of all previous frames. To address these issues, we propose an efficient, effective, and robust segmentation method based on Isogenous Memory Sampling and Frame-Relation mining (IMSFR). Specifically, by utilizing an isogenous memory sampling module, IMSFR consistently conducts memory matching and reading between sampled historical frames and the current frame in an isogenous space, minimizing the semantic gap while speeding up the model through an efficient random sampling. Furthermore, to avoid key information loss during the sampling process, we further design a frame-relation temporal memory module to mine inter-frame relations, thereby effectively preserving contextual information from the video sequence and alleviating error accumulation. Extensive experiments demonstrate the effectiveness and efficiency of the proposed IMSFR method. In particular, our IMSFR achieves state-of-the-art performance on six commonly used benchmarks in terms of region similarity & contour accuracy and speed. Our model also exhibits strong robustness against frame sampling due to its large receptive field.

Ship Contour Extraction From SAR Images Based on Faster R-CNN and Chan–Vese Model

Compared with most ship detection methods for synthetic aperture radar (SAR) images, ship contour extraction can provide more of the detailed shape and edge information of an observed ship and play a significant role in sea surface monitoring and marine transportation. In this study, a joint ship contour extraction method (Faster R-CNN, FNLM, and Chan-Vese model; FFCV method) was proposed to obtain detailed ship information from SAR images, including ship detection in complex scenes and contour extraction in target slices. First, Faster R-CNN was employed to slice ships from large-scene SAR images. Then, Fast non-local mean (FNLM) filtering was applied to denoise and enhance the structural information of the target slices. Finally, an optimized Chan-Vese model was proposed in this paper, which can not only accurately extract the contour of the observed ship, but also reduce the computation time of the model. The SAR ship detection dataset (SSDD) was selected and finely relabeled to evaluate the contour extraction performance. An evaluation index <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">R_N</i> , including quantitative value and offset direction, was developed to evaluate the extraction accuracy of the target contour from the SAR images. Compared with the Mask R-CNN network, the average contour extraction accuracy index <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">R_N</i> of the proposed FFCV method reached -0.002 on all the images in SSDD dataset, and its results were closer to the real ship contours while maintaining the applicability to complex scenes.

A clinical and time savings evaluation of a deep learning automatic contouring algorithm.

Automatic contouring algorithms may streamline clinical workflows by reducing normal organ-at-risk (OAR) contouring time. Here we report the first comprehensive quantitative and qualitative evaluation, along with time savings assessment for a prototype deep learning segmentation algorithm from Siemens Healthineers. The accuracy of contours generated by the prototype were evaluated quantitatively using the Sorensen-Dice coefficient (Dice), Jaccard index (JC), and Hausdorff distance (Haus). Normal pelvic and head and neck OAR contours were evaluated retrospectively comparing the automatic and manual clinical contours in 100 patient cases. Contouring performance outliers were investigated. To quantify the time savings, a certified medical dosimetrist manually contoured de novo and, separately, edited the generated OARs for 10 head and neck and 10 pelvic patients. The automatic, edited, and manually generated contours were visually evaluated and scored by a practicing radiation oncologist on a scale of 1-4, where a higher score indicated better performance. The quantitative comparison revealed high (> 0.8) Dice and JC performance for relatively large organs such as the lungs, brain, femurs, and kidneys. Smaller elongated structures that had relatively low Dice and JC values tended to have low Hausdorff distances. Poor performing outlier cases revealed common anatomical inconsistencies including overestimation of the bladder and incorrect superior-inferior truncation of the spinal cord and femur contours. In all cases, editing contours was faster than manual contouring with an average time saving of 43.4% or 11.8 minutes per patient. The physician scored 240 structures with > 95% of structures receiving a score of 3 or 4. Of the structures reviewed, only 11 structures needed major revision or to be redone entirely. Our results indicate the evaluated auto-contouring solution has the potential to reduce clinical contouring time. The algorithm's performance is promising, but human review and some editing is required prior to clinical use.

Новый оператор выделения контуров объектов железнодорожной инфраструктуры

Presents a new operator for selecting the contours of railway infrastructure objects in images. The study focuses on the development of successive stages of the algorithm, including smoothing with a Gaussian filter, adaptive partitioning to regions and image binarization, as well as suppression of non-contour elements. The operator demonstrates efficiency in improving the accuracy of contouring of railway objects such as trains, cars, platforms and signals. The proposed methods to process images, remove noise and select objects, thus increasing the overall quality of recognition. The simulation results show that the new operator contributes to a significant improvement in the accuracy of fault detection compared to alternative methods.

Abdomen CT multi-organ segmentation using token-based MLP-Mixer.

Manual contouring is very labor-intensive, time-consuming, and subject to intra- and inter-observer variability. An automated deep learning approach to fast and accurate contouring and segmentation is desirable during radiotherapy treatment planning. This work investigates an efficient deep-learning-based segmentation algorithm in abdomen computed tomography (CT) to facilitate radiation treatment planning. In this work, we propose a novel deep-learning model utilizing U-shaped multi-layer perceptron mixer (MLP-Mixer) and convolutional neural network (CNN) for multi-organ segmentation in abdomen CT images. The proposed model has a similar structure to V-net, while a proposed MLP-Convolutional block replaces each convolutional block. The MLP-Convolutional block consists of three components: an early convolutional block for local features extraction and feature resampling, a token-based MLP-Mixer layer for capturing global features with high efficiency, and a token projector for pixel-level detail recovery. We evaluate our proposed network using: (1) an institutional dataset with 60 patient cases and (2) a public dataset (BCTV) with 30 patient cases. The network performance was quantitatively evaluated in three domains: (1) volume similarity between the ground truth contours and the network predictions using the Dice score coefficient (DSC), sensitivity, and precision; (2) surface similarity using Hausdorff distance (HD), mean surface distance (MSD) and residual mean square distance (RMS); and (3) the computational complexity reported by the number of network parameters, training time, and inference time. The performance of the proposed network is compared with other state-of-the-art networks. In the institutional dataset, the proposed network achieved the following volume similarity measures when averaged over all organs: DSC = 0.912, sensitivity = 0.917, precision = 0.917, average surface similarities were HD = 11.95 mm, MSD = 1.90 mm, RMS = 3.86 mm. The proposed network achieved DSC = 0.786 and HD = 9.04 mm on the public dataset. The network also shows statistically significant improvement, which is evaluated by a two-tailed Wilcoxon Mann-Whitney U test, on right lung (MSD where the maximum p-value is 0.001), spinal cord (sensitivity, precision, HD, RMSD where p-value ranges from 0.001 to 0.039), and stomach (DSC where the maximum p-value is 0.01) over all other competing networks. On the public dataset, the network report statistically significant improvement, which is shown by the Wilcoxon Mann-Whitney test, on pancreas (HD where the maximum p-value is 0.006), left (HD where the maximum p-value is 0.022) and right adrenal glands (DSC where the maximum p-value is 0.026). In both datasets, the proposed method can generate contours in less than 5 s. Overall, the proposed MLP-Vnet demonstrates comparable or better performance than competing methods with much lower memory complexity and higher speed. The proposed MLP-Vnet demonstrates superior segmentation performance, in terms of accuracy and efficiency, relative to state-of-the-art methods. This reliable and efficient method demonstrates potential to streamline clinical workflows in abdominal radiotherapy, which may be especially important for online adaptive treatments.

Condition monitoring and mechanism analysis of belt wear in robotic grinding of TC4 workpiece using acoustic emissions

Sharp abrasion of the abrasive belt always causes surface quality to deteriorate in the robotic grinding of thin-walled blades, which has stymied the further advancement of robot machining technology. A novel belt wear prediction approach based on the acoustic emissions (AE) signal is presented in this study to monitor belt wear conditions to achieve the expected machining result. A time-varying model of belt wear height and tangential force is initially established by employing the force analysis for various belt wear states. Next, a relationship between AE signal power and belt wear height is established utilizing the energy principle. Finally, robotic machining experiments on Ti-6Al-4V alloy workpieces are performed to achieve accurate belt wear prediction of an average error of approximately 10%, and the influence analysis of belt wear on the surface quality is further investigated to extend effective service life of abrasive belt by about 20% and realize the desired blade’s machining quality of surface roughness Ra<0.4 μm and contour accuracy ≤±0.17mm.

Toolpath interpolation with novel corner smoothing technique

• Propose a novel filter-based corner smoothing with double filter technique for CNC machining. • Derive analytical formulations for contour errors with satisfying kinematic constraints. • Design filter order and corner speed to ensure that acceleration, jerk and contour error. • Achieve higher machining efficiency and contour accuracy as compared to the state-of-art technique. In this study, a novel filter-based corner smoothing with double filter (FCSDF) technique is proposed to generate interpolated trajectories for high-speed machine tools. The technique uses a double filter to smooth segment junctions and limit axial jerk. The novelty of this study is to derive an analytical formulation for corner contour error and to find the relationship between corner speed and contour error. The FCSDF algorithm provides a systematic way to design filter order and to determine the corner speed to ensure that acceleration, jerk and contour error are confined. As a result, the technology can achieve both high machining accuracy and efficiency and is easy to implement in real time. Finally, the FCSDF is validated using a multisegment trajectory, and the experimental results verify the effectiveness in improving the contouring performance and cycle time.

Study on the Control Method of Sidewall Taper in Electrolytic Broaching of Micro Multi-Grooves.

Micro multi-grooves are important functional structures widely used in new heat exchanger types, chemical reactors, and other applications. Electrolytic broaching is an efficient and low-cost technology for processing micro multi-grooves. In the conventional electrolytic broaching of multi-grooves, the cathode tools are usually designed as a wedge-shaped tooth structure array with a constant tooth width, and the sidewalls are covered with insulating layers. The machined groove sidewall is always tapered because of stray current corrosion, which strongly affects the groove contour accuracy. Cathode tools with variable tooth width structures are proposed to solve this problem. Based on the simulation results of the electrolytic broaching anode forming process, the optimal front tooth width is obtained through the golden section optimization method, and comparative tests of the conventional and optimized cathode tools were carried out. At an electrochemical broaching feed rate of 120 mm/min, array microgrooves with widths of about 550 μm and depths of about 520 μm were processed. With the optimized variable tooth width tool, the sidewall tapers of the grooves were reduced from 7.254° to 0.268°. The experimental results verify the effectiveness of the simulation and cathode structure optimization.

Research on Single Point Incremental Forming Characteristics of Perforated TA1 Sheet

In view of its high compactibility, flexibility, reduced compacting pressure, and other superiorities, the single point incremental forming (SPIF) technique has been applied increasingly in the cranial prosthesis forming of perforated TA1 sheet with irregular surface features. Whereas the SPIF of perforated sheet, which seems to meet many challenges in the biomedicine field, where control of component geometric dimensions are qualification requirements of cranial prosthesis. The mechanism of incremental forming and characteristics of perforated TA1 sheet become more complicated because of the mesh apertures. Hence, certain path and forming technique parameters were set to conduct finite element modelling on the truncated right angle cone components of perforated TA1 sheet and titanium plate respectively on the basis of the ANSYS/LS-DYNA platforms, in light of the constructed model, the distribution and variation law of displacement field coupled with the contour accuracy, wall thickness, and strain capacity for different areas of components in the forming process were obtained by researches, bringing up to light the forming mechanism of conical perforated TA1 sheet components. The research findings indicate that the aperture structure of perforated TA1 sheet allows the material elements to accomplish deformation in the surface where the aperture is located, and the strain rate and radial aperture growth rate of perforated titanium sheet are relatively high, the maximum radial aperture growth rate value is as high as 78.53 percent, the maximum circumferential aperture growth rate value is only 10.84 percent, the extension-thinning of forming section for perforated TA1 sheet is higher than that of the titanium plate, and perforated TA1 sheet possesses higher geometric accuracy than titanium plate.

2D NURBS Curve discretization method with double constraints of area and chord error and contour accuracy prediction modeling

The short segment discretization algorithm for complex curves is one of the critical factors in improving the quality and efficiency of economical CNC machine tools. However, traditional single constraint methods such as equal error and equal chord length algorithms cannot achieve the requirements of both quality and efficiency, and intelligent methods with high algorithmic complexity are unsuitable for traditional machining. To this end, a simple and effective method for generating short segment trajectories called area-chord error double constraint is presented in this paper. The area enclosed by the trace of the short line segment and the original curve is defined as area error, and the semicircle with chord length as the diameter is defined as constrained area. By controlling the ratio of the area error to the constrained area, short line segment separation with variable chord length and error can be achieved according to the curvature characteristics. The influence mechanism of short line segmentation on curves is investigated and the contour accuracy prediction model is established through 70 irregular curves from simple to complex. The experimental results show that this method can retain more geometric features of the original curve on the premise of fewer feature points so as to achieve both processing precision and efficiency.

Accuracy of the Contour® Next Link 2.4 blood glucose monitoring system in Japanese individuals: a single-center study.

Self-monitoring of blood glucose is a useful method for monitoring blood glucose. It is often a key role of a management plan to reduce glycemic variability and diabetic complications. Wireless monitoring systems to connect blood glucose and insulin pumps can facilitate glycemic control. In this study, we evaluated the accuracy of Contour® Next Link 2.4, a blood glucose monitoring system that cooperates wirelessly with most insulin pumps, in Japanese individuals. In this study, finger-stick samples from 59 individuals were collected at the Tokyo Saiseikai Central Hospital. Blood glucose concentrations were measured with the monitoring systems against an available reference. We evaluated the accuracy of the system based on the ISO 15197:2013 Section 6.3 accuracy criteria. In the present study, 100% of the results fulfilled the ISO 15197:2013 Section 6.3 accuracy criteria (95% within ± 15mg/dL or ± 15% of reference for glucose < 100 and ≥ 100mg/dL, respectively). The Parkes-Consensus Error Grid analysis showed that 100% of the results fulfilled within Zone A. The Contour® Next Link 2.4 blood glucose monitoring system fulfilled the ISO 15197:2013 accuracy criteria limit and the consensus error grid criterion. Therefore, this monitoring system for observing blood glucose levels is accurate for Japanese individuals.

Segmentation of Targets and Organs at Risk for CBCT-Based Online Adaptive Radiotherapy Using Recurrent Neural Networks: A Clinical Evaluation

<h3>Purpose/Objective(s)</h3> Accurate and automated segmentations of targets and organs at risk (OARs) are critical for the successful clinical implementation of online adaptive radiotherapy (oART). Deforming contours from an initial planning CT to a newly acquired CBCT is one popular approach utilized by major oART platforms to facilitate online delineation. However, this can be challenging due to poor quality CBCT imaging or large anatomical changes from simulation to treatment. In this work, we developed a framework to utilize contours from previous adaptive treatments (per patient), to further improve the accuracy of the auto segmentations for the upcoming fractions (FXs). <h3>Materials/Methods</h3> Based on a 3D U-Net, we built a Recurrent Neural Networks (RNNs) with Long Short-Term Memory (LSTM). The most recent prior FX contours and the base U-Net were used as performance benchmarks. The model inputs included an image of the most recent FX, its clinician-edited contours, as well as an image of the current FX. The label was the clinician-edited contours for the current FX. For each patient, we trained the model one FX at a time via sequential FXs and reset the memory after the last FX. The memory of LSTMs within each down-sampled block preserves a multi-scale memory about the current patient. The models were pre-trained using simulated data. The simulated data was generated from 95 Head & Neck patients at our institution, by interpolating and extrapolating the deformation vector field from planning CT to CBCT image. We fine-tuned and tested our models using a Head & Neck dataset from a clinical oART system. Contour accuracy was measured using DICE. <h3>Results</h3> The clinical oART dataset included 10 patients and each patient had 6 FXs. Four patients were used for fine-tuning, 2 for evaluation, and 4 for testing. Structures of 4 OARs and GTV-nodal are tested. RNN with prior FX knowledge improved the quality of contours as shown by the average DICE scores in Table 1. "Input" column represented using the prior FX edited contours as the current FX contours. CNN represents the base U-Net model. <h3>Conclusion</h3> Automatic segmentation of OARs and targets on CBCT images can be challenging. By utilizing the contours from previous FXs as prior knowledge through an RNN architecture, we can achieve higher accuracy for segmenting OARs and targets in CBCT images for online adaptive radiotherapy.

Automated Brain Metastasis Detection and Segmentation Using Deep-Learning Method

<h3>Purpose/Objective(s)</h3> Brain metastases (BM) detection and segmentation can be difficult and time-consuming during stereotactic radiosurgery (SRS) planning. This work aims to develop deep learning models to automatically detect and segment BM on MRI. <h3>Materials/Methods</h3> Two volumetric deep convolutional neural network (V-net) models, general (GM) and small lesion models (SM), were developed using T1-weighted gadolinium contrast enhanced and T2-weighted FLAIR MRIs along with ground truth contours of all BMs collected for 53 patients from our own institution. The GM was developed by transfer learning based on a model previously trained with multi-institutional datasets. The SM was a new model presently trained using images slices with small lesions (volumes < 1 cc). The training sets for both models were the data from 43 patients, a total of 5,204 image sets extracted with three different patch sizes (16 × 16 × 16, 32 × 32 × 32, and 64 × 64 × 64). The obtained models were evaluated using a test set of 10 patients in terms of the Dice similarity coefficient (DSC), sensitivity, specificity, and area under the receiver operating characteristic curve (AUC). <h3>Results</h3> For the GM, the voxel-level segmentation AUCs were 0.90 +/- 0.10, 0.88 +/- 0.10, and 0.87 +/- 0.11 for patch sizes of 16 × 16 × 16, 32 × 32 × 32, and 64 × 64 × 64, respectively, and the detection sensitivity and specificity were 0.86 and 0.94, respectively, for all test datasets (all lesions). For the test data excluding small lesions, the sensitivity and specificity became 0.93 and 0.95, respectively. For SM, a sensitivity of 0.76 and a specificity of 0.91 were achieved for all small lesions in the test datasets. The DSC was 0.92 for GM for all lesions and was 0.82 for SM for all the small BMs. <h3>Conclusion</h3> The two deep learning models, SM and GM, can automatically detect and segment brain metastases with reasonable accuracy, sensitivity, and specificity on the MRIs commonly used for SRS planning. For BMs with volumes > 1 cc, the GM achieved clinically relevant results. With further development using larger datasets, the models can be implemented to improve the clinical practice of SRS planning.