Maximum Intensity Projection Algorithm Research Articles

Pressure-Insensitive Epidermal Thickness of Fingertip Skin for Optical Image Encryption.

In this study, an internal fingerprint-guided epidermal thickness of fingertip skin is proposed for optical image encryption based on optical coherence tomography (OCT) combined with U-Net architecture of a convolutional neural network (CNN). The epidermal thickness of fingertip skin is calculated by the distance between the upper and lower boundaries of the epidermal layer in cross-sectional optical coherence tomography (OCT) images, which is segmented using CNN, and the internal fingerprint at the epidermis-dermis junction (DEJ) is extracted based on the maximum intensity projection (MIP) algorithm. The experimental results indicate that the internal fingerprint-guided epidermal thickness is insensitive to pressure due to normal correlation coefficients and the encryption process between epidermal thickness maps of fingertip skin under different pressures. In addition, the result of the numerical simulation demonstrates the feasibility and security of the encryption scheme by structural similarity index matrix (SSIM) analysis between the original image and the recovered image with the correct and error keys decryption, respectively. The robustness is analyzed based on the SSIM value in three aspects: different pressures, noise attacks, and data loss. Key randomness is valid by the gray histograms, and the average correlation coefficients of adjacent pixelated values in three directions and the average entropy were calculated. This study suggests that the epidermal thickness of fingertip skin could be seen as important biometric information for information encryption.

Diagnostic value of three-dimensional cube fluid attenuated inversion recovery imaging and its axial MIP reconstruction in multiple sclerosis

BackgroundMagnetic resonance imaging is regarded as one of the most important markers for multiple sclerosis. It can detect lesions in order to establish dissemination in time and space, which would aid in the diagnosis. Two-dimensional FLAIR is a standard sequence in MS routine imaging because it suppresses cerebrospinal fluid signal, increasing contrast between lesions and CSF and improving white matter lesion detection. Newer 3D FLAIR sequences are expected to offer even more benefits, such as improved MS lesions detection and higher resolution due to thinner slice thickness. We aimed to compare the role of 3D Cube FLAIR imaging (versus standard 2D FLAIR) in the assessment of white matter lesions in MS patients, as well as to test the convenience of using maximum intensity projection (MIP) on 3D FLAIR images for faster and easier evaluation.ResultsThis study included 160 MS patients. A 1.5 T routine brain MRI scan was performed, which included a 2D FLAIR sequence, followed by a 3D-FLAIR sequence. All images were analyzed after 3D-FLAIR images were reformatted into axial MIP images. Lesions were counted in each sequence and classified into supra-tentorial (periventricular, deep white matter, and juxta-cortical), and infra-tentorial lesions, with the relative comparison of lesions numbers on 3D-FLAIR and MIP versus 2D-FLAIR expressed as a percentage increase or decrease. 3D FLAIR can significantly improve MS lesion detection in all areas of the brain when compared with 2D FLAIR results. At 2 mm reformatting, there is no difference in MS lesion detection between sagittal 3D FLAIR and axial MIP reconstruction, implying that the MIP algorithm can be used to simplify lesion detection by reducing the number of images while maintaining the same level of reliability.Conclusion3D FLAIR sequences should be added to conventional 2D FLAIR sequences in the MRI protocol when MS is suspected.

A Novel Bio-Inspired Approach for High-Performance Management in Service-Oriented Networks

Service-continuity in distributed computing can be enhanced by designing self-organized systems, with a non-fixed structure, able to modify their structure and organization, as well as adaptively react to internal and external environment changes. In this paper, an architecture exploiting a bio-inspired management approach, i.e., the functioning of cell metabolism, for specialized computing environments in Service-Oriented Networks (SONs) is proposed. Similar to the processes acting in metabolic networks, the nodes communicate to each other by means of stimulation or suppression chains giving rise to emergent behaviors to defend against foreign invaders, attacks, and malfunctioning. The main contribution of this work is a novel bio-inspired methodology for SON analysis to improve the network reliability and robustness for maintaining service-continuity. To show the effectiveness of the proposed computational framework, an embedded Field-Programmable Gate Array (FPGA) prototyped SON for a relevant healthcare imaging application is also outlined. In particular, our case study extracts and analyzes the Cerebral Vascular Tree from Magnetic Resonance Angiography series <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">via</i> a Maximum Intensity Projection algorithm; the proposed solution addresses and implements some basic issues of an interesting diagnosis tool for cerebral aneurysm detection. The prototyped system was tested and evaluated in terms of execution time and used resource analysis, by achieving a 4× speed-up factor compared to the software counterpart.

Direct Three-Dimensional Diagnosis of Ex Vivo Facial Fractures.

The aim of the present study is to assess the reliability and accuracy of different 3-dimensional (3D) reconstruction algorithms in detecting undisplaced condylar, zygomatic arc, and orbital rim fractures based on cone-beam computed tomography data set. Twenty sheep heads were used in the present study. Sixty fractured and 60 nonfractured (control) zones were randomly allocated. Three groups consisting of nondisplaced fractures of condyle (CF, n = 20), orbital (OF, n = 20), and zygomatic arc (ZF, n = 20) were created by using a diamond cutting disc. Soft tissues were only dissected and no fractures were generated in the control group (n = 60). The 3D reconstructions were created by using multiplanar reconstruction (MPR), surface rendering (SR), volume rendering (VR), and maximum intensity projection (MIP) algorithms. Final 3D models were examined in Osirix software (Pixmeo SARL, Bernex, Switzerland) by 6 observers. Diagnostic accuracies of each algorithm were statistically compared by receiver operating characteristics (ROC) and area under the ROC curves (AUCs). For the detection of CF, AUC for VR algorithm was found to be statistically larger than that of MIP while AUCs for VR and MIP were larger than those of MPR and SR for OF detection. For the detection of ZF, AUCs for MPR and VR were significantly larger than those of MIP and SR (P < 0.05 for each). Within the limitations of this experimental study, it can be concluded that for maxillofacial surgeons, it is more likely to detect condylar, orbital, and zygomatic fractures by using VR algorithm in 3D reconstruction.

CT-calculometry (CT-CM): advanced NCCT post-processing to investigate urinary calculi.

This study aimed at evaluating the potential of CT-calculometry (CT-CM) as a novel method to determine mineralisation, composition, homogeneity and volume of urinary calculi based on preoperative non-contrast-enhanced computed tomography (NCCT) scans. CT-CM was performed in preoperative NCCTs of 25 patients treated for upper tract urinary calculi by ureterorenoscopy or percutaneous nephrolithotomy. Absolute mineralisation values were achieved by use of quantitative CT-osteoabsorptiometry and compared to Fourier infrared spectroscopy as a reference for stone composition. Homogeneity was assessed by advanced software-based NCCT post-processing and visualised by using a maximum intensity projection algorithm. Volumetric measurement was performed by software-based three-dimensional reconstruction. CT-CM was feasible in all of the 25 NCCTs. Absolute mineralisation values calculated by quantitative CT-OAM might be used to identify the most frequent stone types. High levels of inhomogeneity could be detected even in pure component stones. Volumetric measurement could be performed with minimal effort. CT-CM is based on advanced NCCT post-processing software and represents a novel and promising approach to determine mineralisation, composition, homogeneity and volume of urinary calculi based on preoperative NCCT. CT-CM could provide valuable information to predict outcome of different stone treatment methods.

Deep neural network-based computer-assisted detection of cerebral aneurysms in MR angiography.

The usefulness of computer-assisted detection (CAD) for detecting cerebral aneurysms has been reported; therefore, the improved performance of CAD will help to detect cerebral aneurysms. To develop a CAD system for intracranial aneurysms on unenhanced magnetic resonance angiography (MRA) images based on a deep convolutional neural network (CNN) and a maximum intensity projection (MIP) algorithm, and to demonstrate the usefulness of the system by training and evaluating it using a large dataset. Retrospective study. There were 450 cases with intracranial aneurysms. The diagnoses of brain aneurysms were made on the basis of MRA, which was performed as part of a brain screening program. Noncontrast-enhanced 3D time-of-flight (TOF) MRA on 3T MR scanners. In our CAD, we used a CNN classifier that predicts whether each voxel is inside or outside aneurysms by inputting MIP images generated from a volume of interest (VOI) around the voxel. The CNN was trained in advance using manually inputted labels. We evaluated our method using 450 cases with intracranial aneurysms, 300 of which were used for training, 50 for parameter tuning, and 100 for the final evaluation. Free-response receiver operating characteristic (FROC) analysis. Our CAD system detected 94.2% (98/104) of aneurysms with 2.9 false positives per case (FPs/case). At a sensitivity of 70%, the number of FPs/case was 0.26. We showed that the combination of a CNN and an MIP algorithm is useful for the detection of intracranial aneurysms. 4 Technical Efficacy: Stage 1 J. Magn. Reson. Imaging 2018;47:948-953.

The Efficacy of Non-Invasive Radiological Modalities in the Assessment of Carotid Artery Stenosis and Occlusions

Purpose: The aim of the study was to determine the efficacy of different radiologic modalities in detecting and grading of extracranial carotid artery stenosis and occlusion. Materials and Methods: Thirty seven patients who were detected to have severe stenosis or occlusion by initial color doppler sonography (CDS) underwent both Time-offlight (TOF) and 3 dimensional contrast enhanced Magnetic Resonance Angiography (3D-CEMRA) examinations. In both MRA techniques anjiograms were constituted with the maximum intensity projection (MIP) algorithm from the source images. Twenty five of the patients underwent digital subtraction angiography (DSA). The degree of stenoses measured on the angiograms were classified as follows: mild (0-39%), moderate (40-69%), severe (70-99%), and complete occlusion. DSA findings were compared to findings obtained from CDS, TOF and 3D-CEMRA techniques. Results: A total of 100 cervical carotid segments (bilateral ICA and ECA segments in 25 patients) were evaluated. The correlation of DSA and 3D-CEMRA was perfect, as r value was 0.99 and kapa was 0.98; where they were 0.91 and 0.87 for TOF, and 0.98 and 0.95 for CDS, respectively. CDS misdiagnosed one patent segment as occluded and 2 moderate stenoses were overestimated as severe. In 8 segments the degree of stenoses were overestimated by TOF and one patent segment was identified as occluded. Except one patient that bolus timing was incorrect, all findings in 3D-CEMRA were correlated with angiography. Two distal ICA tandem lesions accompanying bifurcation stenoses were correctly identified by 3D-CEMRA and correlated with angiography. One ulcerated plaque was visualised by 3D-CEMRA, but not with TOF. Conclusion: According to our findings we conclude that for the correlation of CDS findings in the patients who are under consideration for carotid endarterectomy, the most reliable modality is 3D-CEMRA. If the findings obtained from CDS and 3D-CEMRA are concordant, angiography is unnecessary. ( Sakarya Med J 2016, 6(2):81-85 )

3D Coronary Artery Reconstruction using SVM

Coronary arteries are the vascular structures that supply blood to the heart muscle. Identification of anomalies in the coronary arteries from the 2D slices is a challenging and time consuming process. Segmenting the coronary arteries from the 2D slices and reconstructing into 3D images help in analyzing abnormalities and enable easy diagnosis for medical experts. Hence in this research work, we propose an automated system that extracts coronary arteries from 2D slices of Computed Tomography Angiography (CTA) images using machine learning techniques and reconstruct them into 3D coronary artery tree. Our proposed method extracts statistical features from the arteries and classifies them as coronary and non-coronary arteries using Support Vector Machine (SVM). Classified coronary arteries from 2D slices are reconstructed into 3D coronary artery tree using Maximum Intensity Projection (MIP) algorithm. The performance of our proposed technique is evaluated using Structure Similarity Index Metric (SSIM).

Noncontrast Magnetic Resonance Lymphography.

Different imaging techniques have been used for the investigation of the lymphatic channels and lymph glands. Noncontrast magnetic resonance (MR) lymphography has significant advantages in comparison with other imaging modalities. Noncontrast MR lymphography uses very heavily T2-weighted fast spin echo sequences which obtain a nearly complete signal loss in tissue background and specific display of lymphatic vessels with a long T2 relaxation time. The raw data can be processed with different algorithms such as maximum intensity projection algorithm to obtain an anatomic representation. Standard T2-weighted MR images easily demonstrate the location of edema. It appears as subcutaneous infiltration of soft tissue with a classical honeycomb pattern. True collection around the muscular area may be demonstrated in case of severe lymphedema. Lymph nodes may be normal in size, number, and signal intensity; in other cases, lymph nodes may be smaller in size or number of lymph nodes may be restricted. MR lymphography allows a classification of lymphedema in aplasia (no collecting vessels demonstrated); hypoplasia (a small number of lymphatic vessels), and numerical hyperplasia or hyperplasia (with an increased number of lymphatic vessels of greater and abnormal diameter). Noncontrast MR lymphography is a unique noninvasive imaging modality for the diagnosis of lymphedema. It can be used for positive diagnosis, differential diagnosis, and specific evaluation of lymphedema severity. It may also be used for follow-up evaluation after treatment.

Role of 16-Multidetector Computerized Tomography in Evaluation of Graft Failure Risk in Patients with Pancreatic Graft Thrombosis After Simultaneous Pancreas and Kidney Transplantation

BackgroundThrombosis of the pancreatic graft vessels is a common complication in patients after pancreas transplantation. The aim of this study was to evaluate the usefulness of 16-multidetector computerized tomography (16-MDCT) with volume rendering (VR) and maximum intensity projection (MIP) reconstruction as a predicting method of pancreatic graft loss after simultaneous pancreas and kidney transplantation. MethodsWithin 6–8 days after combined kidney-pancreas transplantation, MDCT was performed in 60 patients. Secondary reconstructions were obtained with the application of VR and MIP algorithms. Vessel anastomosis, extra- and intrapancreatic arteries, venous morphology, and enhancement of graft's parenchyma were evaluated. The stenosis grade of pancreatic graft vessels and the areas of graft parenchyma malperfusion were evaluated. ResultsThrombosis of pancreatic graft vessels was recognized in 26 patients (43%), however only one-half of them required graftectomy. There were 17 cases of large vessel thrombosis and 9 cases of small intragraft vessel thrombosis. In 13 (86.6%) out of 15 recipients parenchymal malperfusion resulted in graft lost. It appeared that vessel narrowing >40% was a threshold for pancreatic graft loss with sensitivity and false positive values of 100% and 9%, respectively. For parenchyma nonenhancement >60% the sensitivity and false positive values of 100% and 0% were significantly associated with pancreatic graft loss (P < .0001). ConclusionsVascular occlusion of >40% and necrosis >60% of parenchymal volume correspond with pancreatic graft loss.

A comparative MRI study for white matter hyperintensities detection: 2D-FLAIR, FSE PD 2D, 3D-FLAIR and FLAIR MIP

The main objective of this study was to not only determine the most appropriate sequence for the analysis of white matter hyperintensities (WMH) on MRI but also to confirm the advantage of three-dimensional (3D) acquisition, as it has been suggested in previous studies, and to test the convenience of using maximum intensity projection (MIP) algorithms on 3D-fluid-attenuated inversion-recovery (FLAIR) images for a quicker evaluation of brain MR studies. The number of WMH was compared in 40 patients and a control group of 10 volunteers using 4 different imaging modalities: two dimensional (2D)-FLAIR, 2D fast spin echo proton density (FSE PD), 3D-FLAIR and FLAIR MIP. Four experienced radiologists took part in the imaging analysis. All studies were performed on a 1.5-T whole-body MR unit. A statistically significant difference between the number of lesions detected on 3D acquisitions (FLAIR CUBE® or FLAIR MIP sequences) compared with those on 2D-FLAIR or 2D FSE PD was demonstrated. There is no significant difference between 3D-FLAIR and FLAIR MIP, therefore both of them can be used with similar results. 3D-FLAIR sequences should replace conventional 2D-FLAIR and/or FSE PD sequences in the MR acquisition protocol when WMH are suspected. MIP reformat algorithms are less time consuming, therefore these can also be used to simplify the detection. 3D sequences are superior for WMH depiction. Moreover, MIP algorithms allow easier analyses with similar results.

SU-E-J-24: Image-Guidance Using Cone-Beam CT for Stereotactic Body Radiotherapy (SBRT) of Lung Cancer Patients: Bony Alignment or Soft Tissue Alignment?

To assess the reliability of soft tissue alignment by comparing pre- and post-treatment cone-beam CT (CBCT) for image guidance in stereotactic body radiotherapy (SBRT) of lung cancers. Our lung SBRT procedures require all patients undergo 4D CT scan in order to obtain patient-specific target motion information through reconstructed 4D data using the maximum-intensity projection (MIP) algorithm. The internal target volume (ITV) was outlined directly from the MIP images and a 3-5 mm margin expansion was then applied to the ITV to create the PTV. Conformal treatment planning was performed on the helical images, to which the MIP images were fused. Prior to each treatment, CBCT was used for image guidance by comparing with the simulation CT and for patient relocalization based on the bony anatomy. Any displacement of the patient bony structure would be considered as setup errors and would be corrected by couch shifts. Theoretically, as the PTV definition included target internal motion, no further shifts other than setup corrections should be made. However, it is our practice to have treating physicians further check target localization within the PTV. Whenever the shifts based on the soft-tissue alignment (that is, target alignment) exceeded a certain value (e.g. 5 mm), a post-treatment CBCT was carried out to ensure that the tissue alignment is reliable by comparing between pre- and post-treatment CBCT. Pre- and post-CBCT has been performed for 7 patients so far who had shifts beyond 5 mm despite bony alignment. For all patients, post CBCT confirmed that the visualized target position was kept in the same position as before treatment after adjusting for soft-tissue alignment. For the patient population studied, it is shown that soft-tissue alignment is necessary and reliable in the lung SBRT for individual cases.

High-throughput screening of large volumes of whole blood using structured illumination and fluorescent on-chip imaging.

Undiluted blood samples are difficult to image in large volumes since blood constitutes a highly absorbing and scattering medium. As a result of this limitation, optical imaging of rare cells (e.g., circulating tumour cells) within unprocessed whole blood remains a challenge, demanding the use of special microfluidic technologies. Here we demonstrate a new fluorescent on-chip imaging modality that can rapidly screen large volumes of absorbing and scattering media, such as undiluted whole blood samples, for detection of fluorescent micro-objects at low concentrations (for example ≤50-100 particles/mL). In this high-throughput imaging modality, a large area microfluidic device (e.g., 7-18 cm(2)), which contains for example ~0.3-0.7 mL of undiluted whole blood sample, is directly positioned onto a wide-field opto-electronic sensor-array such that the fluorescent emission within the microchannel can be detected without the use of any imaging lenses. This microfluidic device is then illuminated and laterally scanned with an array of Gaussian excitation spots, which is generated through a spatial light modulator. For each scanning position of this excitation array, a lensfree fluorescent image of the blood sample is captured using the opto-electronic sensor-array, resulting in a sequence of images (e.g., 144 lensfree frames captured in ~36 s) for the same sample chip. Digitally merging these lensfree fluorescent images based on a maximum intensity projection (MIP) algorithm enabled us to significantly boost the signal-to-noise ratio (SNR) and contrast of the fluorescent micro-objects within whole blood, which normally remain undetected (i.e., hidden) using conventional uniform excitation schemes, involving plane wave illumination. This high-throughput on-chip imaging platform based on structured excitation could be useful for rare cell research by enabling rapid screening of large volume microfluidic devices that process whole blood and other optically dense media.

Semiautomatic technique for defining the internal gross tumor volume of lung tumors close to liver/spleen cupola by 4D-CTa)

It has been shown that in cases of lung tumors close to the liver cupola, the four dimensional (4D)-CT postprocessing maximum intensity projection (MIP) algorithm does not fully recover the radiotherapy internal gross tumor volume (IGTV). In this work, a semiautomatic technique was evaluated by which the residual IGTV that was not included into the IGTV by MIP algorithm was actually added. A moving phantom and five selected patients were considered. The various IGTVs produced by the semiautomatic approach were compared to those generated by 4D-CT manual contouring. In all cases, the radiation oncologist qualitatively concurred with the semiautomatic IGTV. A quantitative difference in volume of 2.6% was found in the phantom study, whereas a mean difference of 0.1 +/- 4.6% was obtained in the patient studies. A semiautomatic technique to include the residual part of IGTV covered by liver/spleen cupola when using MIP algorithm was validated on phantom and on selected patients, revealing the possibility of defining the IGTV for patients with lesions located near liver/spleen cupola by performing only the contours on the MIP series.

Iliac crest angle: a novel sonographic parameter for the prediction of Down syndrome risk during the second trimester of pregnancy

To validate a new sonographic technique for the display and measurement of fetal iliac crest angle (ICA), and to determine the relative contribution of standardized fetal pelvic measurements for estimating risk of Down syndrome during the second trimester of pregnancy. Three-dimensional ultrasonography of the fetal pelvis was performed during a second-trimester scan for genetic amniocentesis. A coronal view of the 'virtual pelvis' was obtained by applying a maximum intensity projection algorithm to visualize both iliac wings. Two different ICAs were measured from a coronal projection of the fetal pelvis (ICA-coronal 1 and ICA-coronal 2); the latter approach relied on using ischial tuberosities as reference landmarks. Next, a coronal projection of the fetal pelvis was rotated to demonstrate a rendered view of the axial fetal pelvis. Axial inner (ICA-inner), middle (ICA-middle) and outer (ICA-outer) ICAs were measured. The maximum anterior iliac distance measurement was defined as the widest separation between the most anterior points of the rendered iliac crests. The reproducibility of each ICA method was analyzed using Bland-Altman statistics. Simple and multiple logistic regression analysis determined the relative contribution of each parameter as a prenatal predictor of Down syndrome. Ninety-four normal fetuses and 19 fetuses with Down syndrome were examined. The ICA-middle and ICA-coronal 2 parameters were the most reproducible angle measurements. The mean +/- SD ICA-middle measurement for fetuses with Down syndrome was significantly greater than that for normal subjects (94.5 +/- 9.8 degrees vs. 83.1 +/- 6.5 degrees ; P < 0.001). The mean +/- SD ICA-coronal 2 angle measurement for fetuses with Down syndrome was slightly greater than that for normal subjects (57.9 +/- 5.5 degrees vs. 51.9 +/- 7.1 degrees ; P = 0.0014). A multiple logistic regression model including ICA-middle and ICA-coronal 2 provided a predictive ability of 88.1% based on the area under the receiver-operating characteristics curve. This combination had a sensitivity of 94.4% for a false-positive rate of 5% in the detection of Down syndrome. Standardized iliac crest measurements of the fetal pelvis can be used to identify some fetuses at risk for trisomy 21 during the second trimester of pregnancy.

Usefulness of 16-Row Multidetector Computed Tomography With Volume Rendering and Maximum Intensity Projection Reconstruction as a Monitor Pancreatic Graft Vessel Patency During the Early Postoperative Period

Thrombosis of the pancreatic graft vessels is the most common complication after transplantation. It leads to loss of 5% to 8% of grafts during the early postoperative period. The aims of this study were to evaluate the usefulness of 16-row multidetector computed tomography (16-MDCT) with volume rendering (VR) and maximum intensity projection (MIP) reconstruction to monitor pancreatic graft vessel patency during the early postoperative period and the efficacy of a heparin infusion as a treatment for graft thrombosis. Among 40 consecutive simultaneous pancreas-kidney transplant recipients, 16-MDCT was performed at 6 to 8 days after the operation. Secondary reconstructions were obtained with VR and MIP algorithms to evaluate the morphology and patency of the extra- and intrapancreatic arteries and veins. In cases of thrombosis, every patient was treated with an infusion of unfractionated heparin. In 15 recipients, thrombosis of the large vessels was detected by 16-MDCT. Heparin infusions saved five pancreatic grafts (5/15; 33.3%), but the other 10 pancreatic grafts were removed. In another four recipients (4/40; 10%) the thrombi were localized only in small intrapancreatic vessels. Treatment with heparin infusion was successful in 3/4 (75%) cases with patent vessels upon control computed tomography examination. We compared the efficacy of heparin treatment depending on the diameter of the thrombosed vessel, observing a significant difference (5/15 vs 3/4; P < .01; chi-square). 16-MDCT with secondary reconstruction by application of VR and MIP algorithms was an efficient method to visualize not only large pancreatic graft arteries and veins but also intrapancreatic parenchymal vessels. In cases of thrombosis of small intrapancreatic vessels, unfractionated heparin infusion significantly decreased graft loss.

SU-FF-J-90: Target Relocalization Accuracy and PTV Margin Verification Using Three-Dimensional Cone-Beam Computed Tomography (CBCT) in Stereotactic Body Radiotherapy (SBRT) of Lung Cancers

Purpose: To assess the target relocalization accuracy in respect to the bony structures using daily CBCT and thus to validate the planning target volume (PTV) margin used in the lung SBRT. Methods and Materials: All patients underwent 4D CT scanning in preparation for lung SBRT. The internal target volume (ITV) is outlined from the reconstructed 4D data using the maximum-intensity projection (MIP) algorithm. The clinical target volume (CTV) is defined as the ITV plus 3 mm margin and an additional 3 mm margin is added to the CTV to make the PTV. Conformal treatment planning is performed on the helical images, to which the MIP images were fused. Prior to each treatment, CBCT is used to align with the simulation CT (helical) based on bony anatomy. The necessary shifts are recorded. The treating physician then checks and modifies the alignment based on target relocalization within the PTV (soft tissue alignment). The final shifts are derived based on the soft tissue alignment. Two sets of shifts are compared here for the purpose of the study. Results: For 8 consecutive patients, treating 9 targets for a total of 36 fractions, it was found that the treatment setup errors (based on bony anatomy alignment) are 0.23±0.47 cm in the anterior-posterior (AP), 0.06±0.5 cm in the lateral (Lat), and 0.18±0.39 cm in the superior-inferior (S-I) directions, respectively. After the setup error correction, the targets were found to be within the PTV margin with the average shifts of 0.064±0.64 cm in AP, 0.089±0.41 cm in Lat, and 0.21±0.38 cm in S-I directions, respectively. Conclusions: For this patient population, the target relocation accuracy is generally satisfactory after patient setup error correction. This demonstrates that the PTV margin designed based on the ITV outlined on MIP images is appropriate to account for intra and inter-fractional tumor motions.

Internal carotid artery stenosis measurements from 3D reconstructed multi-directional views using phantom data set on MRA image sequence

Purpose To evaluate the inter-observer variability of stenosis measurements by using multi-directional 3D reconstructed projection view of internal carotid artery (ICA) that simulate the pre-determined stenosis degree on magnetic resonance angiography (MRA) image sequence of phantom data set and to compare the difference in measured percentages of maximum ICA stenosis between projection and axial images. Materials and methods By using adaptive region growing and circumscribed quadrangle on the clinical data set, the cylinder shape of a mathematical phantom was modeled and implemented. The maximum ICA stenosis was categorized as mild (30%), moderate (50%), and severe (70%) stenosis those simulated on the tomographic image sequence of ICA only and synthesized phantom. The 36 maximum intensity projection (MIP) images were radially projected at 10° increments that were rotated about the long axis of the body by using the simulated stenosis degree on the tomographic image sequence of phantom data sets. The six different projection image data sets were used to measure the minimum residual lumen and reference diameter by three blinded observers. The percentage of maximum ICA stenosis was calculated as the following. The ICA stenosis grading was [1 − (minimum residual lumen/averaged reference diameter)] × 100%. Results The percentage of maximum ICA stenosis degree measured on projection image was underestimated on ICA only phantom and overestimated on synthesized phantom compared to the simulated ICA stenosis degree on the axial image. In addition, the synthesized phantom provided the less projection image for stenosis measurement than ICA only phantom. These results attributed to the nature of MIP algorithm and the overlapping effects of surrounded anatomic structures such as other ICA, pair of external carotid artery (ECA), pair of VA, and background tissues. Furthermore, the inter-observer variability was also introduced by manual measurement. Conclusion The automated scheme is recommended to measure the ICA stenosis by using axial image. This technique is not only accurate as possible but also robust, simple to handle, and less time consuming compared with manual measurements. A computerized ICA stenosis measuring method, which applied the image processing technique on the axial image, is necessary to overcome the drawbacks introduced by using the projection image and manual measurement.

Comparison of maximum intensity projection and digitally reconstructed radiographic projection for carotid artery stenosis measurement

Digital subtraction angiography is being supplanted by three-dimensional imaging techniques in many clinical applications, leading to extensive use of maximum intensity projection (MIP) images to depict volumetric vascular data. The MIP algorithm produces intensity profiles that are different than conventional angiograms, and can also increase the vessel-to-tissue contrast-to-noise ratio. We evaluated the effect of the MIP algorithm in a clinical application where quantitative vessel measurement is important: internal carotid artery stenosis grading. Three-dimensional computed rotational angiography (CRA) was performed on 26 consecutive symptomatic patients to verify an internal carotid artery stenosis originally found using duplex ultrasound. These volumes of data were visualized using two different postprocessing projection techniques: MIP and digitally reconstructed radiographic (DRR) projection. A DRR is a radiographic image simulating a conventional digitally subtracted angiogram, but it is derived computationally from the same CRA dataset as the MIP. By visualizing a single volume with two different projection techniques, the postprocessing effect of the MIP algorithm is isolated. Vessel measurements were made, according to the NASCET guidelines, and percentage stenosis grades were calculated. The paired t-test was used to determine if the measurement difference between the two techniques was statistically significant. The CRA technique provided an isotropic voxel spacing of 0.38 mm. The MIPs and DRRs had a mean signal-difference-to-noise-ratio of 30:1 and 26:1, respectively. Vessel measurements from MIPs were, on average, 0.17 mm larger than those from DRRs (P < 0.0001). The NASCET-type stenosis grades tended to be underestimated on average by 2.4% with the MIP algorithm, although this was not statistically significant (P=0.09). The mean interobserver variability (standard deviation) of both the MIP and DRR images was 0.35 mm. It was concluded that the MIP algorithm slightly increased the apparent dimensions of the arteries, when applied to these intra-arterial CRA images. This subpixel increase was smaller than both the voxel size and interobserver variability, and was therefore not clinically relevant.

Initial evaluation of a four‐dimensional computed tomography system, using a programmable motor

A dynamic lung tumor phantom was used to investigate the geometric reconstruction accuracy of a commercial four‐dimensional computed tomography (4D‐CT) system. A ball filled with resin, embedded in a cork cube, was placed on a moving platform. Various realistic antero‐posterior (AP) motions were programmed to reproduce the respiratory motion of a lung tumor. Several three‐dimensional (3D) CT and 4D‐CT images of this moving object were acquired and compared using different acquisition parameters. Apparent volume and diameter of the ball were measured and compared to the real values. The position of two points (the AP limits of the ball) during the motion in the coordinate system of the CT scanner were also compared with the expected values. Volume error was shown to increase with object speed. However, although the volume error was associated with intraslice artifacts, it did not reflect large interslice inconstancies in object position and should not be used as an indicator of image accuracy. The 3D‐CT gave a random position of the tumor along the phantom excursion; accuracy in the assessment of position by 4D‐CT ranged from 0.4 mm to 2.6 mm during extreme phases of breathing. We used average projection (AVE) and maximum intensity projection (MIP) algorithms available on the commercial software to create internal target volumes (ITVs) by merging gross tumor volume (GTV) images at various respiratory phases. The ITVs were compared to a theoretical value computed from the programmed ball excursion. The ITVs created from the MIP algorithm were closer to the theoretical value (within 12%) than were those created from the AVE algorithm (within 40%).PACS numbers: 87.53.Xd, 87.56.Mp, 87.57.Ce, 87.59.Fm