Contour Quality Research Articles

ContourGAN: Image contour detection with generative adversarial network

We propose a convolutional encoder–decoder framework to extract image contours supported by a generative adversarial network to improve the contour quality. Traditional image-to-image models only consider the loss between prediction and ground truth, neglecting the similarity between the data distribution of the outcomes and ground truth. Based on this observation, the proposed generative adversarial network aims to increase the detection accuracy. The resulting method contains two models, namely an encoder–decoder model, whose weights are updated using binary cross-entropy loss with fine-tuning from the VGG16 pre-trained model, and a discriminator network that employs ground truths and predicted contours as input for discrimination. We evaluate the method on the Berkeley Segmentation DataSetand Benchmarks 500 and NYUD dataset, achieving state-of-the-art performance with an ODS F-measure of 0.810 and 0.715.

Evaluation of automatic contour propagation in T2-weighted 4DMRI for normal-tissue motion assessment using internal organ-at-risk volume (IRV).

PurposeThe purpose of this study was to evaluate the quality of automatically propagated contours of organs at risk (OARs) based on respiratory‐correlated navigator‐triggered four‐dimensional magnetic resonance imaging (RC‐4DMRI) for calculation of internal organ‐at‐risk volume (IRV) to account for intra‐fractional OAR motion.Methods and MaterialsT2‐weighted RC‐4DMRI images were of 10 volunteers acquired and reconstructed using an internal navigator‐echo surrogate and concurrent external bellows under an IRB‐approved protocol. Four major OARs (lungs, heart, liver, and stomach) were delineated in the 10‐phase 4DMRI. Two manual‐contour sets were delineated by two clinical personnel and two automatic‐contour sets were propagated using free‐form deformable image registration. The OAR volume variation within the 10‐phase cycle was assessed and the IRV was calculated as the union of all OAR contours. The OAR contour similarity between the navigator‐triggered and bellows‐rebinned 4DMRI was compared. A total of 2400 contours were compared to the most probable ground truth with a 95% confidence level (S95) in similarity, sensitivity, and specificity using the simultaneous truth and performance level estimation (STAPLE) algorithm.ResultsVisual inspection of automatically propagated contours finds that approximately 5–10% require manual correction. The similarity, sensitivity, and specificity between manual and automatic contours are indistinguishable (P > 0.05). The Jaccard similarity indexes are 0.92 ± 0.02 (lungs), 0.89 ± 0.03 (heart), 0.92 ± 0.02 (liver), and 0.83 ± 0.04 (stomach). Volume variations within the breathing cycle are small for the heart (2.6 ± 1.5%), liver (1.2 ± 0.6%), and stomach (2.6 ± 0.8%), whereas the IRV is much larger than the OAR volume by: 20.3 ± 8.6% (heart), 24.0 ± 8.6% (liver), and 47.6 ± 20.2% (stomach). The Jaccard index is higher in navigator‐triggered than bellows‐rebinned 4DMRI by 4% (P < 0.05), due to the higher image quality of navigator‐based 4DMRI.ConclusionAutomatic and manual OAR contours from Navigator‐triggered 4DMRI are not statistically distinguishable. The navigator‐triggered 4DMRI image provides higher contour quality than bellows‐rebinned 4DMRI. The IRVs are 20–50% larger than OAR volumes and should be considered in dose estimation.

Geometric and dosimetric evaluation of atlas based auto-segmentation of cardiac structures in breast cancer patients

Background and purposeAuto-segmentation represents an efficient tool to segment organs on CT imaging. Primarily used in clinical setting, auto-segmentation plays an increasing role in research, particularly when analyzing thousands of images in the “big data” era. In this study we evaluate the accuracy of cardiac dosimetric endpoints derived from atlas based auto-segmentation compared to gold standard manual segmentation. Material and methodsHeart and cardiac substructures were manually delineated on 54 breast cancer patients. Twenty-seven patients were used to build the auto-segmentation atlas, the other 27 to validate performance. We evaluated accuracy of the auto-segmented contours with standard geometric indices and assessed dosimetric endpoints. ResultsAuto-segmented contours overlapped geometrically with manual contours of the heart and chambers with Dice-similarity coefficients of 0.93 ± 0.02 (mean ± standard deviation) and 0.79 ± 0.07 respectively. Similarly, there was a strong link between dosimetric parameters derived from auto-segmented and manual contours (R2 = 0.955–1.000). On the other hand, the left anterior descending artery had little geometric overlap (Dice-similarity coefficient 0.09 ± 0.07), though acceptable representation of dosimetric parameters (R2 = 0.646–0.992). ConclusionsThe atlas based auto-segmentation approach delineates heart structures with sufficient accuracy for research purposes. Our results indicate that quality of auto-segmented contours cannot be determined by geometric values only.

Bridging the Gap in Global Advanced Radiation Oncology Training: Impact of a Web-Based Open-Access Interactive Three-Dimensional Contouring Atlas on Radiation Oncologist Practice in Russia.

Radiation oncologists in Russia face a number of unique professional difficulties including lack of standardized training and continuing medical education. To combat this, under the auspices of the Russian Society of Clinical Oncology (RUSSCO), our group has developed a series of ongoing in-person interactive contouring workshops that are held during the major Russian oncology conferences in Moscow, Russia. Since November 2016 during each workshop, we utilized a web-based open-access interactive three-dimensional contouring atlas as part of our didactics. We sought to determine the impact of this resource on radiation oncology practice in Russia. We distributed an IRB-approved web-based survey to 172 practicing radiation oncologists in Russia. We inquired about practice demographics, RUSSCO contouring workshop attendance, and the clinical use of open-access English language interactive contouring atlas (eContour). The survey remained open for 2months until November 2017. Eighty radiation oncologists completed the survey with a 46.5% response rate. Mean number of years in practice was 13.7. Sixty respondents (75%) attended at least one RUSSCO contouring workshop. Of those who were aware of eContour, 76% were introduced during a RUSSCO contouring workshop, and 81% continue to use it in their daily practice. The greatest obstacles to using the program were language barrier (51%) and internet access (38%). Nearly 90% reported their contouring practices changed since they started using the program, particularly for delineation of clinical target volumes (57%) and/or organs at risk (46%). More than 97% found the clinical pearls/links to cooperative group protocols in the software helpful in their daily practice. The majority used the contouring program several times per month (43%) or several times per week (41%). Face-to-face contouring instruction in combination with open-access web-based interactive contouring resource had a meaningful impact on perceived quality of radiation oncology contours among Russian practitioners and has the potential to have applications worldwide.

An experimental comparison of superpixels detection methods for contour detection

Recently, many superpixels detection methods have been proposed and used in various applications. We are interested in which method is more suitable for the application of contour detection. In this paper, superpixels are evaluated on BSDS500 dataset in two different aspects. On the one hand, contours are directly provided by the boundaries of superpixels and experiments show that better results could be achieved by the superpixels with irregular shapes than those with regular shapes and similar sizes. On the other hand, contours are further detected from those candidate positions which are confirmed by the boundaries of superpixels through the operation of dilation. In this situation, experiments show that competitive results could also be achieved by some superpixels with regular shapes and similar sizes. Besides, we propose a superpixels detection method called watershed-based graph (WG), by which superpixels with irregular shapes could be produced. Firstly, a graph is constructed from an over-segmented map which is achieved through a watershed algorithm. Then, to get the desired superpixels, the graph is segmented by merging neighbor segments in an order of decreasing similarity. Experiments show that higher efficiency could be achieved by WG with a moderate worse contour quality than its original graph-based method.

Assessment of contour profile quality in D&B tunnelling

Contour profile quality affects tunnel excavation costs, in terms of operational safety, support materials and construction time. In drill and blast tunnelling, under/over-excavation and rock mass damage arising from excavation phase can be evaluated by means of the elaboration of survey data and geophysical testing or coring, before and after the blast. As far as the quality of the profile is concerned, some indices can be used to define the contour and for the rock mass in the boundary as well.This paper proposes a methodology well applicable to rock tunnelling, and a case study based analysis to correlate the over-excavation and the rock mass conditions is discussed to validate the procedure. Profiles and geological parameters have been processed with automatic code specifically developed for the study. Over-excavation distance and Tunnel Contour Quality Index are evaluated and compared with Q-system values. The results have been discussed, compared with other literature cases and validated for engineering applications.

The Evidence Driven Dosimetric Constraints From Outcome Analysis of H&N Patients’ Data from NRG Oncology RTOG 0522 Trial

Clinical trial quality assurance (QA) programs have been shown to be vital in ensuring that inter-institutional differences do not dilute trial results. Most RTOG clinical trials have a radiation therapy quality assurance (RTQA) process that evaluates RT parameters (contour, dose distribution) retrospectively or prospectively. This research is to investigate the impact of radiation treatment clinical trial quality assurance (RTQA) on treatment outcome in a phase III trial for advanced head and neck cancer with development of a predictive model incorporating RTQA parameters. RTQA for RTOG 0522 included initial institution credentialing of RT technologies and individual RT case reviews. The case review processes (include contour and dosimetry evaluations) are performed by radiation oncology and radiation physics co-chairs. RTQA grades (per-protocol, variation acceptable, and deviation unacceptable) are given to contouring of target volume (TV), organ at risk (OAR) and dose-volume coverage of targets as defined in the protocol. The relationship between RTQA parameters and treatment outcome are analyzed with predictive modeling. Logistic regression model is established that includes RTQA parameters, age, T-stage, equivalent dose in fractions of 2 Gy (EQD2), tumor location, and hemoglobin levels. This model is compared to the one without incorporating RTQA parameters. The model prediction accuracy is validated by cross-validation and C-statistical methods. The contour (TV and OAR) quality grades did not correlate with two-year overall survival. The target dose-volume quality grade is slightly related to overall survival (P = 0.094), and the correlation test shows that target dose-volume quality grade is an independent predictive factor. The target contour quality grade (P = 0.019) and target dose-volume quality grade (P = 0.008) are related to local recurrence, while target dose-volume quality grade (P = 0.034) is related to distant metastasis. The model incorporating RTQA parameters shows that the two-year survival ratio is 86.2%, 83.6%, and 80.6% for the three scores of target dose-volume quality of per-protocol, variation acceptable, and deviation unacceptable. The area under the curve (AUC) indicator is 0.737 and 0.733 for the overall survival model with RTQA parameters and without RTQA parameters. The results demonstrate that it is feasible to incorporate RTQA parameters into outcome modeling. This capability is of critical importance for evaluating RTQA methods as related to outcome in head and neck cancer.

Knowledge-Based Auto Contouring for Radiation Therapy: Challenges in Standardizing Object Definitions, Ground Truth Delineations, Object Quality, and Image Quality

Manual contouring in RT planning of organs at risk (OARs) has been prone to significant variability due to a lack of standardized object definitions, leading to imprecision. Although clinical guidelines have been made available, ambiguities in object definitions still exist. Suboptimal object quality and image quality can also significantly impact the quality of object contouring. These issues are particularly problematic when creating robust object models for purposes of auto-contouring. As such, we present more precise definitions of selected OARs in the neck and thorax as an extension of existing guidelines. A prevailing issue is evaluation of auto-contouring methods as a function of input image quality. We propose a new approach to assess the impact of object and image quality upon auto-contouring results. We hypothesize that OAR manual contours still have significant variability from existing guidelines, and object and image quality impact auto-contouring. We first developed precise and computable definitions of the neck and thorax body regions based on anatomical considerations. We then developed precise standardized definitions of a set of key OARs in these body regions by extending object definitions available from recent guidelines. Next, we retrospectively created a database of CT images and manually drawn contours obtained from RT planning in 216 head and neck cancer patients and 200 thoracic cancer patients. A reader examined each image and assigned a quality grade to each 3D object in the image based on a set of pre-specified criteria which included: deviation in body posture and object intensity; presence of image noise, streak artifacts, object shape distortion, and pathology; and lack of object contrast. Using logical predicates, we designed an algorithm to map these quality grades to a numeric quality score for each object in each CT scan, as well as for the scan. Precise object definitions for 11 objects in the neck and 11 objects in the thorax were created. Per our standardized definitions, among all objects in the neck studies, 33.8% were acceptable without modification, 46.9% required minor changes, and 19.3% needed major editing. For the thorax, these rates were 10.5%, 56.17%, and 33.33%, respectively. ∼1% of the scans, considering all objects in each scan, were of high quality as per the above criteria. Results of analysis of an auto-contouring software as a function of the image quality numeric score will be presented at the conference. Precise standardized object definitions are essential to ensure high quality of OAR contours for RT planning. Yet, currently available guidelines are still ambiguous as borne out from our analysis. Suboptimal image quality may also significantly impact the quality of object contouring. We have arrived at more precise definitions of a set of objects in the neck and thorax, and developed a new approach to assess the impact of object and image quality upon auto-contouring results.

Implementation of a Machine Learning–Based Automatic Contour Quality Assurance Tool for Online Adaptive Radiation Therapy of Prostate Cancer

Implementation of a Machine Learning–Based Automatic Contour Quality Assurance Tool for Online Adaptive Radiation Therapy of Prostate Cancer

SU‐C‐BRA‐03: An Automated and Quick Contour Errordetection for Auto Segmentation in Online Adaptive Radiotherapy

Purpose:To develop a tool that can quickly and automatically assess contour quality generated from auto segmentation during online adaptive replanning.Methods:Due to the strict time requirement of online replanning and lack of ‘ground truth’ contours in daily images, our method starts with assessing image registration accuracy focusing on the surface of the organ in question. Several metrics tightly related to registration accuracy including Jacobian maps, contours shell deformation, and voxel‐based root mean square (RMS) analysis were computed. To identify correct contours, additional metrics and an adaptive decision tree are introduced. To approve in principle, tests were performed with CT sets, planned and daily CTs acquired using a CT‐on‐rails during routine CT‐guided RT delivery for 20 prostate cancer patients. The contours generated on daily CTs using an auto‐segmentation tool (ADMIRE, Elekta, MIM) based on deformable image registration of the planning CT and daily CT were tested.Results:The deformed contours of 20 patients with total of 60 structures were manually checked as baselines. The incorrect rate of total contours is 49%. To evaluate the quality of local deformation, the Jacobian determinant (1.047±0.045) on contours has been analyzed. In an analysis of rectum contour shell deformed, the higher rate (0.41) of error contours detection was obtained compared to 0.32 with manual check. All automated detections took less than 5 seconds.Conclusion:The proposed method can effectively detect contour errors in micro and macro scope by evaluating multiple deformable registration metrics in a parallel computing process. Future work will focus on improving practicability and optimizing calculation algorithms and metric selection.

Analysis of fabric materials cut using ultraviolet laser ablation

Laser ablation technology has widely been applied in the clothing industry in recent years. However, the laser mechanism would affect the quality of fabric contours and its components. Hence, this study examined carbonization and oxidation conditions and contour variation in nonwoven, cotton, and composite leather fabrics cut by using an ultraviolet laser at a wavelength of 355 nm. Processing parameters such as laser power, pulse frequency, scanning speed, and number of pulses per spot were adjusted to investigate component variation of the materials and to determine suitable cutting parameters for the fabrics. The experimental results showed that the weights of the component changed substantially by pulse frequency but slightly by laser power, so pulse frequency of 100 kHz and laser power of 14 W were the approximate parameters for three fabrics for the smaller carbonization and a sufficient energy for rapidly cutting, which the pulse duration of laser system was fixed at 300 μs and laser irradiance was 0.98 J/mm2 simultaneously. In addition, the etiolate phenomenon of nonwoven was reduced, and the component weight of cotton and composite leather was closed to the value of knife-cut fabric as the scanning speed increased. The approximate scanning speed for nonwoven and composite leather was 200 mm/s, and one for cotton was 150 mm/s, respectively. The sharper and firmer edge is obtained by laser ablation mechanism in comparison with traditional knife cutting. Experimental results can serve as the reference for laser cutting in the clothing industry, for rapidly providing smoother patterns with lower carbonization and oxidation edge in the fashion industry.

Consistency of Organ Geometries during Prostate Radiotherapy with Two Different Bladder and Bowel Regimens

BackgroundThe majority of Ontario cancer centres incorporate bladder and bowel preparation protocols for the treatment of prostate cancer with radical radiotherapy. Differing methods are used to achieve a full bladder and empty rectum for planning and treatment. We compared the effects of two different bladder and bowel preparation regimens on bladder, rectum, and prostate +/− seminal vesicle geometries through a course of radiotherapy. An optimal preparation would achieve reliable spatial arrangements and a high therapeutic ratio. MethodsThis prospective longitudinal study involved 59 prostate cancer patients treated with radical radiotherapy, of which half followed cohort 1 (laxative cohort) and the other cohort 2 (consistent timing cohort) bladder and bowel preparation regimen. Participants were asked to maintain an empty rectum for both planning and daily treatment appointments in cohort 1 through a fleet enema the morning of the planning appointment, and intake milk of magnesium during daily treatments. No specific bowel preparation was provided to cohort 2 patients. Instead, their appointment times were aligned with their natural bowel habits. This information was collected through a prescreening tool before treatment booking. All cohort 1 and 2 participants were asked to drink 250 mL of water 1 hour before planning and daily treatment appointments. Cohort 2 participants who identified no pre-existing urinary conditions were also asked to drink 2 L of water within 24 hours before the planning session and to continue this during treatment trajectory unless unable to do so because of treatment-induced bladder toxicities later in the treatment. A total of 1,335 structures (bladder, rectum +/− gas, and prostate +/− seminal vesicles) were contoured on the cone beam computerized tomography scans by three radiotherapists. A stringent quality assurance process was performed to assure quality and consistency of contours. Organ volumes were measured and evaluated for consistency over time from planning to completion of radiotherapy. Data analysis included the Fischer exact test and mixed effect modelling for total and subvolumes for bladder, rectum, rectal gas, and prostate +/− seminal vesicles. ResultsBaseline total volumes for bladder ranged from 132 mL to 501 mL with means of 325 mL and 315 mL in cohorts 1 and 2, respectively. Bladder volume declined 3.6 mL per fraction and 2.4 mL per fraction in cohorts 1 and 2, respectively. The volume of the bladder structure inside the planning target volume (PTV) on simulation showed no difference by cohort (P = .095) but there was an effect of time (linear P < .0005). Baseline total volumes for rectum ranged from 19.2 mL to 106.3 mL with means of 52.0 mL and 54.7 mL in cohorts 1 and 2, respectively. The volume of the rectum inside the PTV on simulation showed no difference by cohort (P = .12) or time (P = .30) during the treatment course. Volume of gas in the rectum did not vary by cohort (P = .6) or time (P = .08). Baseline total volumes for the clinical prostate +/− seminal vesicles target ranged from 37.1 mL to 167.5 mL with means of 76.2 mL and 66.0 mL in cohorts 1 and 2, respectively. The clinical target decreased by 3% in total volume during the course of radiotherapy in both cohorts, with similar rates of the target falling outside the planned PTV structure. ConclusionsNo significant difference was found between cohorts for rectal volume, gas volume, target coverage, and rectal and bladder volumes in the PTV. Hence, patients should be offered a choice between cohort 1 and 2 bowel preparation regimens to allow for patient preference customization. Cohort 2 bladder preparation regimen was shown to be superior for consistency with slightly larger volume over time.

A study on the establishment of Tunnel Contour Quality Index considering construction cost

Tunnel Contour Quality Index (TCI) has been proposed in previous studies to quantify a contour quality resulting from blasting in a tunnel or underground space, and to effectively compare the contour quality with other cases. TCI evaluates general geometric features of a tunnel contour with three parameters regarding overbreak, three weights, three correction factors and one range adjustment constant.This paper proposes another concept of TCI to consider the effect of a contour quality on construction cost. This index is termed TCIC (Tunnel Contour Quality Index for construction cost) to be distinguished from the existing TCI. TCIC consists of two parameters, two weights and one constant for range adjustment. The two parameters are defined by three elements, i.e., average depth of overbreak, the ratio of actual contour length to planned length, and longitudinal overbreak difference, representing increased tunnel volume and increased contour area for a given contour condition. This analysis can be made by simplifying the tunnel into a semi-circular truncated cone having the same three elements. The two weights are constants represented by the unit cost ratio between two operation groups, i.e., one group for blasting, muckpile and concrete lining costs, and the other group for shotcrete and waterproof membrane.To check the feasibility of the index, the contour quality of two tunnels (5 detailed tunnels) in Norway is evaluated by TCIC, based upon the unit construction costs for other two Norwegian tunnels.The result shows that the range of TCIC is between 34 and 55. It is thought that the difference % in TCIC is similar to the difference % in the unit construction cost increased due to the contour quality. In addition, a statistical analysis shows that a usual range of TCIC is between 30 and 61.Finally, it is thought that TCIC can be used as a tool not only to evaluate contour quality, but also to estimate the construction cost increased due to a contour quality in a tunnel project.

A framework for automated contour quality assurance in radiation therapy including adaptive techniques

Contouring of targets and normal tissues is one of the largest sources of variability in radiation therapy treatment plans. Contours thus require a time intensive and error-prone quality assurance (QA) evaluation, limitations which also impair the facilitation of adaptive radiotherapy (ART). Here, an automated system for contour QA is developed using historical data (the ‘knowledge base’). A pilot study was performed with a knowledge base derived from 9 contours each from 29 head-and-neck treatment plans. Size, shape, relative position, and other clinically-relevant metrics and heuristically derived rules are determined. Metrics are extracted from input patient data and compared against rules determined from the knowledge base; a computer-learning component allows metrics to evolve with more input data, including patient specific data for ART. Nine additional plans containing 42 unique contouring errors were analyzed. 40/42 errors were detected as were 9 false positives. The results of this study imply knowledge-based contour QA could potentially enhance the safety and effectiveness of RT treatment plans as well as increase the efficiency of the treatment planning process, reducing labor and the cost of therapy for patients.

A Precision-Recall Criterion Based Consensus Model For Fusing Multiple Segmentations

This paper presents a general framework for seamlessly combining multiple low cost and inaccurate estimated segmentation maps (with an arbitrary number of regions) of the same scene to achieve a final improved segmentation. The proposed fusion model is derived from the well-known precision-recall criterion, specially dedicated to the specific clustering problem of any spatially indexed data and which is also efficient and widely used in the vision community for evaluating both a region-based segmentation and the quality of contours produced by this segmentation map compared to one or multiple ground-truth segmentations of the same image. The proposed combination framework is here specifically designed to be robust with respect to outlier segmentations (that appear to be inconsistent with the remainder of the segmentation ensemble) and includes an explicit internal regularization factor reflecting the inherent ill-posed nature of the segmentation problem. We propose also a hierarchical and efficient way to optimize the consensus energy function related to this fusion model that exploits a simple and deterministic iterative relaxation strategy combining the different segments or individual regions belonging to the segmentation ensemble in the final solution. The experimental results on the Berkeley database with manual ground truth segmentations show the effectiveness of our combination model.

Arterial contour detectability in head CT angiography.

Arterial contour extraction is essential for visualization and analysis of vasculature in CT angiography (CTA). A means for evaluating the detectability of artery contours CTA images is required. We developed and tested a new method for this purpose based on phase information from two-dimensional Fourier transforms of CTA images. The relationship between arterial contour detectability and a patient's ocular lens dose was evaluated in CTA images obtained with various tube voltages and currents. A head phantom was designed for use as a target object containing a simulated vascular tree, filled with dilute contrast medium (10 mg iodine/ml). The head phantom was scanned using a 64-multidetector CT scanner with tube voltages of 80-140 kV and tube currents corresponding to volume CT dose index [Formula: see text] ranging from 24.4 to 72.8 mGy. Lens doses were measured using the planar silicon PIN-photodiode system. The quality of artery contours in the CTA source images was assessed using a computed detectability index. Lens dose increased proportionally with tube voltage and current. The use of 80 kV provided the highest contour detectability. However, for each tube voltage, the detectability of artery contours was almost constant across the CTDI(vol) values. These results were mostly consistent with the subjective recognition of artery contours on CTA images. A CTA protocol using 80 kV and 420 mA can reduce the radiation exposure to ocular lens by approximately 40 %, and improve the artery contour detectability compared with a routine protocol.

Noise Adaptation and Threshold Determination in Image Contour Recognition Method Based on Complex Network

In practice of image contour recognition, the precision of shape contour extraction is affected by lots of factors, such as noise, shelter and parameters. That will affect the shape contour quality and reduce the recognition effect. To solve these problems, a Shape Contour Recognition Method Based on Complex Network is discussed in this study. The main idea of the approach is to use complex network methodology to extract a feature vector for shape contour recognition under rotation, noise and shelter. An approximation method for Distance Threshold Determining (DTD) is presented to help modeling the complex networks. Experiments show that the proposed method and the DTD method have efficient power in shape recognition. It is also proved to be scale invariant, rotation invariant and partially overcome noise-sensitive and shelter.

Groupwise Conditional Random Forests for Automatic Shape Classification and Contour Quality Assessment in Radiotherapy Planning

Radiation therapy is used to treat cancer patients around the world. High quality treatment plans maximally radiate the targets while minimally radiating healthy organs at risk. In order to judge plan quality and safety, segmentations of the targets and organs at risk are created, and the amount of radiation that will be delivered to each structure is estimated prior to treatment. If the targets or organs at risk are mislabelled, or the segmentations are of poor quality, the safety of the radiation doses will be erroneously reviewed and an unsafe plan could proceed. We propose a technique to automatically label groups of segmentations of different structures from a radiation therapy plan for the joint purposes of providing quality assurance and data mining. Given one or more segmentations and an associated image we seek to assign medically meaningful labels to each segmentation and report the confidence of that label. Our method uses random forests to learn joint distributions over the training features, and then exploits a set of learned potential group configurations to build a conditional random field (CRF) that ensures the assignment of labels is consistent across the group of segmentations. The CRF is then solved via a constrained assignment problem. We validate our method on 1574 plans, consisting of 17[Formula: see text] 579 segmentations, demonstrating an overall classification accuracy of 91.58%. Our results also demonstrate the stability of RF with respect to tree depth and the number of splitting variables in large data sets.

Iris center location using Hough transform with two-dimensional parameter space

An algorithm of approximate location of iris center in image is presented. It is based on Hough transformation for circles. Problem is set in a way to detect only iris center position without evaluating its size. Such formulation allows to reduce the dimension of parameter space compared to commonly used approaches that detect center position and radius simultaneously. Apart, omitting radius estimation gives an opportunity to use points of both pupil and iris circles in Hough transform, increasing the stability of method especially for images with poor pupil contour quality. The algorithm was tested for more than 95000 iris images from public domain databases.

A study on micro machining of e-glass–fibre–epoxy composite by ECSM process

The paper presents the result of an experimental investigation on the micro machining of electrically non-conductive e-glass–fibre–epoxy composite during electrochemical spark machining using specially designed square cross section with centrally micro hole brass tool and different diameter round-shaped micro tools made of IS-3748 steel. A micro electrochemical spark machining (ECSM) setup has been designed, fabricated and used for conducting the experiments. According to the Taguchi method-based design, the specific numbers of experiments have been carried out to investigate the influence of the fabricated ECSM parameters on the material removal rate and overcut on generated hole radius. Test results show that the material removal rate is maximum when machining was performed at higher setting value of D.C. supply voltage (e.g. 70 V), moderate setting value of electrolytic concentration (e.g. 80 g/l) and 180-mm gap between electrodes. Taking significant machining parameters into consideration and using multiple linear regression, mathematical modes for material removal rate and overcut on hole radius are established to investigate the influence of cutting parameters during micro-ECSM. The influence of machining parameters on machined hole and special shape contour quality are also analysed through different scanning electron micrographs. Confirmation test results established the fact that the developed mathematical models are appropriate for effectively representing the machining performance criteria.