Gold Standard Volume Research Articles

Comparison of three methods for nucleus pulposus volume measurement in rabbit lumbar spines: a preclinical model for measurement of the effectiveness of prophylactic intervertebral disk fenestration in dogs.

Compare 3 methods of nucleus pulposus (NP) volume measurement using the rabbit lumbar spines as a preclinical model to determine the effectiveness of prophylactic intervertebral disk fenestration in dogs. Twelve 9-month-old, skeletally mature female entire New Zealand White rabbits weighing between 3.5 to 4.5 kg. NP volume measurements of dissected rabbit lumber spines between L1 and L6 were made and compared using gross measurements, reconstructed MRI images, and water volumetry based on Archimedes' principle. Water volumetry was used as the true gold standard volume measurement in this study. The true volume (mean ± SD) of the nucleus pulposus NP as measured by water volumetry increased caudally from L1/L2 (16.26 ± 3.32 mm3) to L5/L6 (22.73 ± 6.09 mm3). Volume estimates made by MRI were significantly higher than those made using water volumetry at all sites (L1/L2 [P = .044], L2/L3 [P = .012], L3/L4 [P = .015], L4/L5 [P < .001], and L5/L6 [P < .001]). Gross measurements also significantly overestimated volume when compared to water volumetry at all sites; L1/L2 (P = .021), L2/L3 (P = .025), L3/L4 (P = .001), L4/L5 (P < .001), and L5/L6 (P < .001). MRI and gross volume estimates were significantly different at L4/L5 (P = .035) and L5/L6 (P = .030). The findings of this preclinical model might be relevant to veterinary surgeons who perform prophylactic fenestration for which there is no reliable method to determine the amount of NP to be removed. Preclinical ex vivo and in vivo fenestration studies with pre- and postoperative NP volume assessment are required.

Respiratory Pattern Analysis for Different Breathing Types and Recording Sensors in Healthy Subjects.

Accurate monitoring of respiratory activity can lead to early identification and treatment of possible respiratory failure. However, spontaneous breathing can vary considerably. To quantify this variability, this study aimed at comparing the breathing pattern characteristics obtained from several recording sensors during different breathing types. Respiratory activity was recorded with a pneumotachograph and two inductive plethysmographic bands, thoracic and abdominal, in 23 healthy volunteers (age 21.5±1.2 years, 13 females). The subjects were asked to breathe at their natural rate, in successive stages: first freely, then through their nose, nose and mouth, mouth alone, and finally deep and shallow. Both band signals were compared to the pneumotach-derived (gold standard) volume signal. The time series of inspiratory and expiratory duration, total cycle duration and tidal volume were estimated from each of these signals, and also from the sum of the thoracic and abdominal bands. This composite signal showed the highest correlation with the volume signal for almost all subjects, and also had a significantly higher correlation with those obtained from the gold standard volume, compared to either band. In general, breathing parameters increased from basal to nose-mouth breathing, had a minimum in shallow breathing and a maximum in deep breathing. Women exhibited a significantly longer exhalation phase than men during deep breathing, in the combined bands and the gold standard volume. In conclusion, variations in respiratory cycle morphology in different breathing types can be well captured by the simple addition of abdominal and thoracic band signals.Clinical Relevance- Breathing pattern variability can be identified by the combination of abdominal and thoracic bands.

Lung Volume Calculation in Preclinical MicroCT: A Fast Geometrical Approach.

In this study, we present a time-efficient protocol for thoracic volume calculation as a proxy for total lung volume. We hypothesize that lung volume can be calculated indirectly from this thoracic volume. We compared the measured thoracic volume with manually segmented and automatically thresholded lung volumes, with manual segmentation as the gold standard. A linear regression formula was obtained and used for calculating the theoretical lung volume. This volume was compared with the gold standard volumes. In healthy animals, thoracic volume was 887.45 mm3, manually delineated lung volume 554.33 mm3 and thresholded aerated lung volume 495.38 mm3 on average. Theoretical lung volume was 554.30 mm3. Finally, the protocol was applied to three animal models of lung pathology (lung metastasis and transgenic primary lung tumor and fungal infection). In confirmed pathologic animals, thoracic volumes were: 893.20 mm3, 860.12 and 1027.28 mm3. Manually delineated volumes were 640.58, 503.91 and 882.42 mm3, respectively. Thresholded lung volumes were 315.92 mm3, 408.72 and 236 mm3, respectively. Theoretical lung volume resulted in 635.28, 524.30 and 863.10.42 mm3. No significant differences were observed between volumes. This confirmed the potential use of this protocol for lung volume calculation in pathologic models.

Accuracy of two deep learning-based reconstruction methods compared with an adaptive statistical iterative reconstruction method for solid and ground-glass nodule volumetry on low-dose and ultra-low-dose chest computed tomography: A phantom study.

No published studies have evaluated the accuracy of volumetric measurement of solid nodules and ground-glass nodules on low-dose or ultra–low-dose chest computed tomography, reconstructed using deep learning–based algorithms. This is an important issue in lung cancer screening. Our study aimed to investigate the accuracy of semiautomatic volume measurement of solid nodules and ground-glass nodules, using two deep learning–based image reconstruction algorithms (Truefidelity and ClariCT.AI), compared with iterative reconstruction (ASiR-V) in low-dose and ultra–low-dose settings. We performed computed tomography scans of solid nodules and ground-glass nodules of different diameters placed in a phantom at four radiation doses (120 kVp/220 mA, 120 kVp/90 mA, 120 kVp/40 mA, and 80 kVp/40 mA). Each scan was reconstructed using Truefidelity, ClariCT.AI, and ASiR-V. The solid nodule and ground-glass nodule volumes were measured semiautomatically. The gold-standard volumes could be calculated using the diameter since all nodule phantoms are perfectly spherical. Subsequently, absolute percentage measurement errors of the measured volumes were calculated. Image noise was also calculated. Across all nodules at all dose settings, the absolute percentage measurement errors of Truefidelity and ClariCT.AI were less than 11%; they were significantly lower with Truefidelity or ClariCT.AI than with ASiR-V (all P<0.05). The absolute percentage measurement errors for the smallest solid nodule (3 mm) reconstructed by Truefidelity or ClariCT.AI at all dose settings were significantly lower than those of this nodule reconstructed by ASiR-V (all P<0.05). Furthermore, the lowest absolute percentage measurement errors for ground-glass nodules were observed with Truefidelity or ClariCT.AI at all dose settings. The absolute percentage measurement errors for ground-glass nodules reconstructed with Truefidelity at ultra–low-dose settings were significantly lower than those of all sizes of ground-glass nodules reconstructed with ASiR-V (all P<0.05). Image noise was lowest with Truefidelity (all P<0.05). In conclusion, the deep learning–based algorithms were more accurate for volume measurements of both solid nodules and ground-glass nodules than ASiR-V at both low-dose and ultra–low-dose settings.

Improving Interpretability in Machine Diagnosis: Detection of Geographic Atrophy in OCT Scans

PurposeManually identifying geographic atrophy (GA) presence and location on OCT volume scans can be challenging and time consuming. This study developed a deep learning model simultaneously (1) to perform automated detection of GA presence or absence from OCT volume scans and (2) to provide interpretability by demonstrating which regions of which B-scans show GA.DesignMed-XAI-Net, an interpretable deep learning model was developed to detect GA presence or absence from OCT volume scans using only volume scan labels, as well as to interpret the most relevant B-scans and B-scan regions.ParticipantsOne thousand two hundred eighty-four OCT volume scans (each containing 100 B-scans) from 311 participants, including 321 volumes with GA and 963 volumes without GA.MethodsMed-XAI-Net simulates the human diagnostic process by using a region-attention module to locate the most relevant region in each B-scan, followed by an image-attention module to select the most relevant B-scans for classifying GA presence or absence in each OCT volume scan. Med-XAI-Net was trained and tested (80% and 20% participants, respectively) using gold standard volume scan labels from human expert graders.Main Outcome MeasuresAccuracy, area under the receiver operating characteristic (ROC) curve, F1 score, sensitivity, and specificity.ResultsIn the detection of GA presence or absence, Med-XAI-Net obtained superior performance (91.5%, 93.5%, 82.3%, 82.8%, and 94.6% on accuracy, area under the ROC curve, F1 score, sensitivity, and specificity, respectively) to that of 2 other state-of-the-art deep learning methods. The performance of ophthalmologists grading only the 5 B-scans selected by Med-XAI-Net as most relevant (95.7%, 95.4%, 91.2%, and 100%, respectively) was almost identical to that of ophthalmologists grading all volume scans (96.0%, 95.7%, 91.8%, and 100%, respectively). Even grading only 1 region in 1 B-scan, the ophthalmologists demonstrated moderately high performance (89.0%, 87.4%, 77.6%, and 100%, respectively).ConclusionsDespite using ground truth labels during training at the volume scan level only, Med-XAI-Net was effective in locating GA in B-scans and selecting relevant B-scans within each volume scan for GA diagnosis. These results illustrate the strengths of Med-XAI-Net in interpreting which regions and B-scans contribute to GA detection in the volume scan.

Contribution of spectroscopic magnetic resonance imaging to target volume delineation in Gamma Knife Radiosurgery: Myth or reality?

Brain malignancies are still associated with poor prognosis despite multimodal radiosurgical therapeutic approach using Gamma Knife (GK), CyberKnife (CK), and linear accelerator-based technologies [1]. These advances have significantly improved the treatment outcome. However, the surgical and radiosurgical concept is still “image-guided”, and the success is closely related to precise tumor volume definition. The gross tumor volume (GTV) is defined as the visible contrast- enhancing lesion on magnetic resonance (MR) images with high three-dimensional spatial accuracy. Target delineation requires always both T2-weighted and volumetric T1-weighted sequences. T2-weighted fluid attenuated inversion recovery (FLAIR) sequences analyze the lesions surrounding brain tissues [2,3]. Objective assessment of apparently healthy tissue surrounding brain tumors seems to be a considerable factor interfering not only with the radiosurgical procedure, but also with the recurrence rate and overall survival. Several studies identified infiltrative spectroscopic pattern of the perilesional edema in more than 96% of high-grade gliomas cases and in 11,5% of patients with brain metastasis [4]. Moreover, some autopsy series of brain metastases confirmed infiltrative growth in radiologically healthy surrounding tissues in more than 60% of cases. This unseen malignant component is responsible of 80 % of “early recurrence” which should be considered as natural evolution of the main tumor [5]. In the management of high grade gliomas, the radiosurgeons are faced either to carcinologic incomplete procedures or to overestimated target irradiation with unbalanced benefit/risk action mostly related to radiation-induced brain necrosis [6]. The delineation of clinical target volume (CTV) which is defined as the volume of tissue that contains the GTV and any microscopic tumor or paths of spread, became a standard for any radio-surgical planning. Since a decade, the magnetic resonance spectroscopy (MRS) was standardized in the target volume assessment. The aim is to establish a metabolic lesional cartography. It had been reported that choline/ N- acetylaspartate (NAA) multivoxel MR spectroscopy index higher that 2,5 is in favor of malignancy in glioma with sensitivity of 90 % and specificity of 85 % [7]. However, NAA/Creatine (Cr) and Choline/Cr ratios are more relevant in the analysis of perilesional edema in brain metastasis cases. The introduction of MRS metabolic cartography concept, the use of relevant metabolite and adapted metabolites ratio estimation contributed to precision in radiosurgery. However, MRS is not used for target delineation for Gamma Knife radiosurgical treatment because of its incompatibility with the Leksell Gamma Knife planning software. Recently, we described the development of the first software allowing the integration of metabolic cartography based on multivoxel spectroscopic MRI in the radiosurgical planning for Leksell Gamma Knife Radiosurgery. The few existing meta-analysis could not lead to gold standard volume delineation techniques despite objective advance in imaging assessment [8,9]. Prospective studies using multimodal imaging data will help to overcome this insufficiency for target delineation in radiosurgery

An integrated multi-objective whale optimized support vector machine and local texture feature model for severity prediction in subjects with cardiovascular disorder.

The left ventricle (LV) myocardium undergoes deterioration with the reduction in ejection fraction (EF). The analysis of its texture pattern plays a major role in diagnosis of heart muscle disease severity. Hence, a classification framework with co-occurrence of local ternary pattern feature (COALTP) and whale optimization algorithm has been attempted to improve the prediction accuracy of disease severity level. This analysis is carried out on 600 slices of 76 participants from Kaggle challenge that include subjects with normal and reduced EF. The myocardium of LV is segmented using optimized edge-based local Gaussian distribution energy (LGE)-based level set, and end-diastolic and end-systolic volumes were calculated. COALTP is extracted for two distance levels (d = 1 and 2). The t-test has been performed between the features of individual binary classes. The features are ranked using feature ranking methods. The experiments have been performed to analyze the performance of various percentages of features in each combination of bin for fivefold cross-validation. An integrated whale optimized feature selection and multi-classification framework is developed to classify the normal and pathological subjects using CMR images, and DeLong test has been performed to compare the ROCs. The optimized edge embedded to level set has produced better segmented myocardium that correlates with R = 0.98 with gold standard volume. The t-test shows that texture features extracted from severe subjects with distance level "1" are more statistically significant with a p value (< 0.00004) compared to other pathologies. This approach has produced an overall multi-class accuracy of 75% [confidence interval (CI) 63.74-84.23%] and effective subclass specificity of 70% (CI 55.90-81.22%). The obtained results show that the multi-objective whale optimized multi-class support vector machine framework can effectively discriminate the healthy and patients with reduced ejection fraction and potentially support the treatment process.

Current Volumetric Models Overestimate Vestibular Schwannoma Size Following Stereotactic Radiosurgery.

Accurate volume assessment is essential for the management of vestibular schwannoma after stereotactic radiosurgery (SRS). A cuboidal approximation for volume is the standard surveillance method; however, this may overestimate tumor volume. We sought to evaluate several volumetric models and their suitability for post-SRS surveillance. Retrospective cohort study. Tertiary referral center. We evaluated 54 patients with vestibular schwannoma before and after SRS. Gold-standard volumes were obtained by a radiation oncologist using contouring software. Volume was also calculated by cuboidal, ellipsoidal, and spherical formulae using tumor diameters obtained by a neuroradiologist. Percent error (PE) and absolute percent error (APE) were calculated. Paired t test evaluated bias, and the Bland-Altman method evaluated reproducibility. Linear regression evaluated predictors of model error. All models overestimated volume compared with the gold standard. The cuboidal model was not reproducible before SRS (p < 0.001), and no model was reproducible after SRS (cuboidal p < 0.001; ellipsoidal p = 0.02; spherical p = 0.02). Significant bias was present before SRS for the cuboidal model (p < 0.001), and post-SRS for all models [cuboidal (p < 0.001), ellipsoidal (p < 0.02), and spherical (p = 0.005)]. Model error was negatively associated with pretreatment volume for the cuboidal (PE p = 0.03; APE p = 0.03), ellipsoidal (PE p = 0.03; APE p = 0.04), and spherical (PE p = 0.02; APE p = 0.03) methods and lost linearity post-SRS. The standard cuboidal practice for following vestibular schwannoma tumor volume after SRS overestimates size. Ellipsoidal and spherical estimations have improved performance but also overestimate volume and lack reliability post-SRS. The development of other volumetric models or application of contouring software should be investigated.

Correlations between contouring similarity metrics and simulated treatment outcome for prostate radiotherapy

Many similarity metrics exist for inter-observer contouring variation studies, however no correlation between metric choice and prostate cancer radiotherapy dosimetry has been explored. These correlations were investigated in this study. Two separate trials were undertaken, the first a thirty-five patient cohort with three observers, the second a five patient dataset with ten observers. Clinical and planning target volumes (CTV and PTV), rectum, and bladder were independently contoured by all observers in each trial. Structures were contoured on T2-weighted MRI and transferred onto CT following rigid registration for treatment planning in the first trial. Structures were contoured directly on CT in the second trial. STAPLE and majority voting volumes were generated as reference gold standard volumes for each structure for the two trials respectively. VMAT treatment plans (78 Gy to PTV) were simulated for observer and gold standard volumes, and dosimetry assessed using multiple radiobiological metrics. Correlations between contouring similarity metrics and dosimetry were calculated using Spearman’s rank correlation coefficient.No correlations were observed between contouring similarity metrics and dosimetry for CTV within either trial. Volume similarity correlated most strongly with radiobiological metrics for PTV in both trials, including TCPPoisson (ρ = 0.57, 0.65), TCPLogit (ρ = 0.39, 0.62), and EUD (ρ = 0.43, 0.61) for each respective trial. Rectum and bladder metric correlations displayed no consistency for the two trials. PTV volume similarity was found to significantly correlate with rectum normal tissue complication probability (ρ = 0.33, 0.48). Minimal to no correlations with dosimetry were observed for overlap or boundary contouring metrics. Future inter-observer contouring variation studies for prostate cancer should incorporate volume similarity to provide additional insights into dosimetry during analysis.

Non-invasive volumetric analysis of asymptomatic hands using a 3-D scanner.

Hand swelling is one of the symptoms often seen in practice, but none of the available morphometric methods can quickly and efficiently quantify hand volume in an objective manner, and the current gold-standard volume measurement requires immersion in water, which can be difficult to use. Therefore, we aimed to analyze the accuracy of using 3-dimensional (3-D) scanning to measure hand volume. First, we compared the hand volume calculated using the 3-D scanner to that calculated from the conventional method among 109 volunteers to determine the reliability of 3-D measurements. We defined the beginning of the hand as the distal wrist crease, and 3-D forms of the hands were captured by the 3-D scanning system. Second, 238 volunteers (87 men, 151 women) with no disease or history of hand surgery underwent 3-D scanning. Data collected included age, height, weight, and shoe size. The wrist circumference (WC) and the distance between distal wrist crease and tip of middle finger (DDT) were measured. Statistical analyses were performed using linear regression to investigate the relationship between the hand volume and these parameters. In the first study, a significantly strong positive correlation was observed [R = 0.98] between the hand volume calculated via 3-D scanning and that calculated via the conventional method. In the second study, no significant differences between the volumes, WC or DDT of right and left hands were found. The correlations of hand volume with weight, WC, and DDT were strong. We created a formula to predict the hand volume using these parameters; these variables explained approximately 80% of the predicted volume. We confirmed that the new 3-D scanning method, which is performed without touching the hand and can record the form of the hand, yields an accurate volumetric analysis of an asymptomatic hand.

Validation of the Semiquantitative Static SUVR Method for 18F-AV45 PET by Pharmacokinetic Modeling with an Arterial Input Function.

Increased brain uptake of 18F-AV45 visualized by PET is a key biomarker for Alzheimer disease (AD). The SUV ratio (SUVR) is widely used for quantification, but is subject to variability based on choice of reference region and changes in cerebral blood flow. Here we validate the SUVR method against the gold standard volume of distribution (VT) to assess cross-sectional differences in plaque load. Methods: Dynamic 60-min 18F-AV45 (291 ± 67 MBq) and 1-min 15O-H2O (370 MBq) scans were obtained in 35 age-matched elderly subjects, including 10 probable AD, 15 amnestic mild cognitive impairment (aMCI), and 10 cognitively healthy controls (HCs). 18F-AV45 VT was determined from 2-tissue-compartment modeling using a metabolite-corrected plasma input function. Static SUVR was calculated at 50-60 min after injection, using either cerebellar gray matter (SUVRCB) or whole subcortical white matter (SUVRWM) as the reference. Additionally, whole cerebellum, pons, centrum semiovale, and a composite region were examined as alternative references. Blood flow was quantified by 15O-H2O SUV. Data are presented as mean ± SEM. Results: There was rapid metabolization of 18F-AV45, with only 35% of unchanged parent remaining at 10 min. Compared with VT, differences in cortical Aβ load between aMCI and AD were overestimated by SUVRWM (+4% ± 2%) and underestimated by SUVRCB (-10% ± 2%). VT correlated better with SUVRWM (Pearson r: from 0.63 for posterior cingulate to 0.89 for precuneus, P < 0.0001) than with SUVRCB (Pearson r: from 0.51 for temporal lobe [P = 0.002] to 0.82 for precuneus [P < 0.0001]) in all tested regions. Correlation results for the alternative references were in between those for CB and WM. 15O-H2O data showed that blood flow was decreased in AD compared with aMCI in cortical regions (-5% ± 1%) and in the reference regions (CB, -9% ± 8%; WM, -8% ± 8%). Conclusion: Increased brain uptake of 18F-AV45 assessed by the simplified static SUVR protocol does not truly reflect Aβ load. However, SUVRWM is better correlated with VT and more closely reflects VT differences between aMCI and AD than SUVRCB.

Reference standard space hippocampus labels according to the European Alzheimer's Disease Consortium–Alzheimer's Disease Neuroimaging Initiative harmonized protocol: Utility in automated volumetry

IntroductionA harmonized protocol (HarP) for manual hippocampal segmentation on magnetic resonance imaging (MRI) has recently been developed by an international European Alzheimer's Disease Consortium–Alzheimer's Disease Neuroimaging Initiative project. We aimed at providing consensual certified HarP hippocampal labels in Montreal Neurological Institute (MNI) standard space to serve as reference in automated image analyses. MethodsManual HarP tracings on the high-resolution MNI152 standard space template of four expert certified HarP tracers were combined to obtain consensual bilateral hippocampus labels. Utility and validity of these reference labels is demonstrated in a simple atlas-based morphometry approach for automated calculation of HarP-compliant hippocampal volumes within SPM software. ResultsIndividual tracings showed very high agreement among the four expert tracers (pairwise Jaccard indices 0.82–0.87). Automatically calculated hippocampal volumes were highly correlated (rL/R = 0.89/0.91) with gold standard volumes in the HarP benchmark data set (N = 135 MRIs), with a mean volume difference of 9% (standard deviation 7%). ConclusionThe consensual HarP hippocampus labels in the MNI152 template can serve as a reference standard for automated image analyses involving MNI standard space normalization.

Accuracy of virtual models in the assessment of maxillary defects.

PurposeThis study aimed to assess the reliability of measurements performed on three-dimensional (3D) virtual models of maxillary defects obtained using cone-beam computed tomography (CBCT) and 3D optical scanning.Materials and MethodsMechanical cavities simulating maxillary defects were prepared on the hard palate of nine cadavers. Images were obtained using a CBCT unit at three different fields-of-views (FOVs) and voxel sizes: 1) 60×60 mm FOV, 0.125 mm3 (FOV60); 2) 80×80 mm FOV, 0.160 mm3 (FOV80); and 3) 100×100 mm FOV, 0.250 mm3 (FOV100). Superimposition of the images was performed using software called VRMesh Design. Automated volume measurements were conducted, and differences between surfaces were demonstrated. Silicon impressions obtained from the defects were also scanned with a 3D optical scanner. Virtual models obtained using VRMesh Design were compared with impressions obtained by scanning silicon models. Gold standard volumes of the impression models were then compared with CBCT and 3D scanner measurements. Further, the general linear model was used, and the significance was set to p=0.05.ResultsA comparison of the results obtained by the observers and methods revealed the p values to be smaller than 0.05, suggesting that the measurement variations were caused by both methods and observers along with the different cadaver specimens used. Further, the 3D scanner measurements were closer to the gold standard measurements when compared to the CBCT measurements.ConclusionIn the assessment of artificially created maxillary defects, the 3D scanner measurements were more accurate than the CBCT measurements.

Measurement of uterine fibroid volume: a comparative accuracy and validation of methods study

ObjectiveA range of measurement techniques have been described which may be used to calculate uterine fibroid volume. A commonly-reported method involves application of a formula for the volume of an ellipsoid sphere to three orthogonal axes of a fibroid as measured on cross-sectional images. We aimed to compare this method and a second method, that of software-computed parallel planimetric uterine fibroid computation on MRI images, to a gold standard: the volume of objects measured by water displacement. We also compared these methods in volume estimation of patient fibroids using MRI data. Study designMixed observational study and blinded cross-sectional analysis of imaging data. ResultsLarge inter-observer variability was noted when using the ellipsoid formula method, which was also inaccurate when compared to the gold standard. Conversely, the parallel planimetric method showed excellent interobserver correlation and a high degree of correlation with gold standard volume measurements. ConclusionWe conclude that the parallel planimetric method, although a more complex and time consuming technique, is the more accurate and therefore preferred method for measuring uterine fibroid volume.

A Semi-Automated Method for Measuring White Matter Hyperintensity Volume

YongSoo Shim, M.D.*, Bora Yoon, M.D.†, Yun Jeong Hong, M.D.‡, A-Hyun Cho, M.D.*, Dong-Won Yang, M.D.* Department of Neurology*, The Catholic University of Korea, Seoul; Department of Neurology†, Konyang University College of Medicine, Daejeon; ‡Hyoja Geriatric Hospital, Yongin, Korea Background: White matter hyperintensities (WMHs) on magnetic resonance imaging (MRI) have been considered as a reliable biomarker of small vessel damages. To evaluate the severity of WMHs, it is vital to develop reliable methods to measure the volume of WMHs. We applied open source software to measure WMH volume in the semi-automated way, and tested the reliability and validity by comparing with the commonly used qualitative rating scale. Methods: Twenty five subjects with variable WMHs were recruited. ANALYZE 10.0 was used for the image processing and volumetric measurement of WMHs. The inhomogeneity and artifacts of signal were corrected with Insight Segmentation and Registration Toolkit in ANALYZE. For the gold standard of the WMH volumetric measurement, threshold method was applied with consensus of manual editing on each slice of the MRI images by two raters. Histogram of the all slices of the Fluid Attenuated Inversion Recovery (FLAIR) MRI was generated to calculate the optimal voxel intensity of threshold, and the lowest voxel threshold was decided as the mean+1.4 SD. The volumes of WMHs were generated by multiplying the area and the thickness of each slice. Interand intrarater reliability of the semi-automated volumetric and Scheltens’methods, and the association between the individual methods were analyzed. Results: The semi-automated WMH volume at the threshold of 1.4 SD as well as the gold standard volume was well correlated with the Scheltens’ visual scale (r= 0.75, p< 0.001). The semi-automated volumetry showed the excellent intra-rater (ICC= 0.9929; 95% CI, 0.98400.9968) and inter-rater reliability (ICC= 0.9830; 95% CI, 0.9620-0.9925), superior to the Scheltens’ visual rating scale. Conclusions: The semi-automated volume measurement of the WMHs with Analyze was a valid and a reliable method to quantify subcortical white matter damages of various etiologies. Received: November 15, 2012 Revision received: March 29, 2013 Accepted: March 29, 2013

Determination of internal target volume using selective phases of a 4-dimensional computed tomography scan

PurposeInternal target volume (ITV) is frequently determined by contouring of gross tumor volumes (GTV) on 10 phases of a 4-dimensional computed tomography (4DCT) study set for lung cancer radiotherapy. This study aimed to investigate the possibility of generating ITV by using selective phases of a 4DCT scan. Methods and MaterialsThe 4DCT scans of 20 patients with lung cancer were included in this study. GTVs were contoured on 10 phases in Focal4D (CMS, St Louis, MO). Different ITVs were derived by encompassing volumes of contours from selective phases. ITV10 was the combination of GTVs on all of the 10 phases and served as the gold standard volume. All of the other ITVs were smaller and within ITV10. The ratios of the volumes of these ITVs to ITV10 were calculated and used as a criterion to determine the similarity of different ITVs to ITV10. ITV2 represented the ITV derived by using end-inhalation and end-exhalation (0% + 50%). ITV3E was derived from contouring the 3 phases at end-inhalation, mid-exhalation, and end-exhalation (0% + 20% + 50%). ITV3I was derived from contouring the 3 phases at end-inhalation, mid-inhalation, and end-exhalation (0% + 70% + 50%). ITV4 was derived by contouring the 4 phases at end-inhalation, mid-inhalation, end-exhalation, and mid-exhalation (0% + 20% + 50% + 70%). ITV6E was derived from contouring the 6 consecutive phases during exhalation (0% + 10% + 20% + 30% + 40% + 50%). ITV6I was derived from contouring the 6 consecutive phases during inhalation (50% + 60% + 70% + 80% + 90% + 0%). The volumes of ITVs were calculated and compared. ResultsITV6I showed excellent agreement with ITV10 (volume ratio ITV6I/ITV10 = 0.975). ITV4 and ITV6E showed good agreement with ITV10 (ITV6E/ITV10 = 0.939, ITV4/ITV10 = 0.944). The volume ratios ITV3I/ITV10 and ITV3E/ITV10 were 0.927 and 0.906, respectively. ITV2 did not agree well with ITV10 (ITV2/ITV10 = 0.888). ConclusionsContouring all phases during inhalation provides a good estimate of the ITV. However, the ITV may be underestimated if only contouring on 2 extreme phases.

WE‐G‐214‐02: Utility of 18F‐FDOPA PET for Radiotherapy Target Delineation in Glioma Patients

Purpose: To evaluate the impact of adding 18F‐FDOPA‐PET, an amino acid tracer, to standard of care MRI for radiotherapy target volume delineation for gliomas. Methods: 6 patients with newly diagnosed brain tumors were given 5.0 mCi +/− 10% of 3,4‐dihydroxy‐6‐[18F]‐fluoro‐L‐phenylalanine (18F‐FDOPA), and underwent a 10‐minute PET/CT acquisition. Overall pathology confirmed WHO grade 3 astrocytoma (n=3), grade 3 oligoastrocytoma (n=1), and glioblastoma (n=2). T1‐weighted post gadolinium (T1‐gad) and T2‐weighted/FLAIR MRI sequences were rigidly aligned to the 18F‐FDOPA‐PET/CT. Three experienced radiation oncologists contoured the T2/FLAIR abnormality, T1‐gad enhancement, and areas of 18F‐FDOPA uptake separately. An experienced nuclear medicine physician contoured 18F‐FDOPA uptake, which was considered as the gold standard for positive disease on PET imaging. Intersection, union, and exclusion boolean operators were performed to determine discordance and concordance between MRI and 18F‐FDOPA‐PET, along with interobserver variability defining 18F‐FDOPA‐PET volume. In addition, uniform expansions of the T1‐gad contours were performed in increments of 0.5 cm until 100% of the 18F‐FDOPA‐PET volume was covered. Results: 18F‐FDOPA‐PET uptake was observed in all patients, with interobserver delineation variability between radiation oncologists (RadOnc1‐RadOnc3) and the gold standard ranging from 47% less to 191% more volume. On average 9.5% (0.3%–29.1%) of the 18F‐FDOPA‐PET gold standard volume extended beyond the RadOnc T2/FLAIR volumes. Half of the cases had no enhancement, and of the remainder on average 77.8% (61.3%–93.1%) of the 18F‐FDOPA‐PET volume extended outside the T1‐gad enhancement. An average expansion of 2.0cm (1.0cm–5.0cm) beyond the T1‐gad contours was necessary to cover 100% of the 18F‐FDOPA‐PET volume. Conclusions: This work introduces the potential utilization of 18F‐FDOPA‐PET for identifying positive disease not visible with conventional MRI and the potential to customize radiotherapy target volumes to the active disease. Future studies will correlate pathology with concordant and discordant regions, and compare biopsy confirmed results with automatic segmentation techniques for 18F‐FDOPA‐PET.Funding has been provided by Brains Together for a Cure

Development and Validation of a Heart Atlas to Study Cardiac Exposure to Radiation Following Treatment for Breast Cancer

Cardiac toxicity is an important sequela of breast radiotherapy. However, the relationship between dose to cardiac structures and subsequent toxicity has not been well defined, partially due to variations in substructure delineation, which can lead to inconsistent dose reporting and the failure to detect potential correlations. Here we have developed a heart atlas and evaluated its effect on contour accuracy and concordance. A detailed cardiac computed tomography scan atlas was developed jointly by cardiology, cardiac radiology, and radiation oncology. Seven radiation oncologists were recruited to delineate the whole heart, left main and left anterior descending interventricular branches, and right coronary arteries on four cases before and after studying the atlas. Contour accuracy was assessed by percent overlap with gold standard atlas volumes. The concordance index was also calculated. Standard radiation fields were applied. Doses to observer-contoured cardiac structures were calculated and compared with gold standard contour doses. Pre- and post-atlas values were analyzed using a paired t test. The cardiac atlas significantly improved contour accuracy and concordance. Percent overlap and concordance index of observer-contoured cardiac and gold standard volumes were 2.3-fold improved for all structures (p < 0.002). After application of the atlas, reported mean doses to the whole heart, left main artery, left anterior descending interventricular branch, and right coronary artery were within 0.1, 0.9, 2.6, and 0.6 Gy, respectively, of gold standard doses. This validated University of Michigan cardiac atlas may serve as a useful tool in future studies assessing cardiac toxicity and in clinical trials which include dose volume constraints to the heart.

Probabilistic Brain Tissue Segmentation in Neonatal Magnetic Resonance Imaging

A fully automated method has been developed for segmentation of four different structures in the neonatal brain: white matter (WM), central gray matter (CEGM), cortical gray matter (COGM), and cerebrospinal fluid (CSF). The segmentation algorithm is based on information from T2-weighted (T2-w) and inversion recovery (IR) scans. The method uses a K nearest neighbor (KNN) classification technique with features derived from spatial information and voxel intensities. Probabilistic segmentations of each tissue type were generated. By applying thresholds on these probability maps, binary segmentations were obtained. These final segmentations were evaluated by comparison with a gold standard. The sensitivity, specificity, and Dice similarity index (SI) were calculated for quantitative validation of the results. High sensitivity and specificity with respect to the gold standard were reached: sensitivity >0.82 and specificity >0.9 for all tissue types. Tissue volumes were calculated from the binary and probabilistic segmentations. The probabilistic segmentation volumes of all tissue types accurately estimated the gold standard volumes. The KNN approach offers valuable ways for neonatal brain segmentation. The probabilistic outcomes provide a useful tool for accurate volume measurements. The described method is based on routine diagnostic magnetic resonance imaging (MRI) and is suitable for large population studies.

A study of prostate delineation referenced against a gold standard created from the visible human data

Purpose To measure inter- and intra-observer variation and systematic error in CT based prostate delineation, where individual delineations are referenced against a gold standard produced from photographic anatomical images from the Visible Human Project (VHP). Materials and methods The CT and anatomical images of the VHP male form the basic data set for this study. The gold standard was established based on 1 mm thick anatomical photographic images. These were registered against the 3 mm thick CT images that were used for target delineation. A total of 120 organ delineations were performed by six radiation oncologists. Results The physician delineated prostate volume was on average 30% larger than the “true” prostate volume, but on average included only 84% of the gold standard volume. Our study found a systematic delineation error such that posterior portions of the prostate were always missed while anteriorly some normal tissue was always defined as target. Conclusions Our data suggest that radiation oncologists are more concerned with the unintentional inclusion of rectal tissue than they are in missing prostate volume. In contrast, they are likely to overextend the anterior boundary of the prostate to encompass normal tissue such as the bladder.