Our study investigated the impact of irregular respiration on correlation between surrogate respiratory signal and internal organ motion in 4DCT. Based on the proposed quantitative index of respiratory irregularity, the study cases were divided into two: the case with regular breathing and the case with irregular breathing. By groups, internal organ movements were measured from 4CT images, and correlation between external surrogate signal and internal organ motion was analyzed. The results showed that the correlation with internal organ motion was obviously deteriorated in the case with irregular breathing, rather than the case with regular breathing. As observed that the irregular respiration may lead to such errors on gating interval up to 20% either phase or amplitude level, it would be necessary to employ the respiratory guidance and feedback system that can minimize the respiratory irregularity.
Prediction of survival period for patients with hepatocellular carcinoma (HCC) provides important information for treatment planning such as radiotherapy. However, the task is known to be challenging due to the similarity of tumor imaging characteristics from patients with different survival periods. In this paper, we propose a survival prediction method using deep learning and radiomic features from CT images with support vector machine (SVM) regression. First, to extract the deep features, the convolutional neural network (CNN) is trained for the task of classifying the patients for 24-month survival. Second, the radiomic features including texture and shape are extracted from the patient images. After concatenating the radiomic features and the deep features, the SVM regressor is trained to predict the survival period of the patients. The experiment was performed on the CT scans of 171 HCC patients with 5-fold cross validation. In the experiments, the proposed method showed an accuracy of 86.5%, a root-mean-squared-error (RMSE) of 11.6, and a Spearman rank coefficient of 0.11. In comparisons with the deep feature-only- and radiomic feature-only regression results, the proposed method showed improved accuracy and RMSE than both, but lower rank coefficient than the radiomic feature-only regression. It can be observed that (1) the deep learning of CT images has a promising potential for predicting the survival period of HCC patients, and (2) the radiomic feature analysis provides useful information to strengthen the power of deep learning-based survival prediction.
Liver segmentation is a prerequisite for measuring hepatic volume in liver transplantation, modeling of the liver anatomy in hepatic surgery planning, and contouring in radiotherapy treatment planning. The main challenges of liver segmentation are the appearance similarity of liver and surrounding stomach, heart, and spleen in 2D images and are the large shape variations of liver in 3D volume. Therefore, we propose a deep learning-based liver segmentation method by using global context of three orthogonal planes to localize the liver in whole abdomen and by using local context of targeted liver bounding volume and high-score shape prior to delineate the liver without leakage to the surrounding structures. To localize the liver within the whole abdomen and exclude outliers through the global context, three 2D segmentation networks are learned on each axial, coronal, and sagittal planes. To consider the shape information obtained from the 2D segmentation network in the next 3D segmentation network, the high-score shape prior is generated by a weighted fusion of three score maps. To correct the fine details of the liver in the targeted liver bounding volume and to be less affected by shape variation, the 3D segmentation network is learned based on 3D U-Net with highscore shape prior. Experimental results show that the DSC of the proposed segmentation network with high-score shape prior (LiverNet-WS) was 94.3%, which is 5.4% higher than LiverNet without high-score shape prior. The proposed method accurately localized the liver within the whole abdomen by using global contexts of three orthogonal planes. Moreover, segmentation accuracy improved fine details considering local context and avoided over-segmentation considering high-score shape prior.
Dual-energy imaging can enhance lesion conspicuity. However, the conventional (fast kilovoltage switching)
dual-shot dual-energy imaging is vulnerable to patient motion. The single-shot method requires a special design
of detector system. Alternatively, single-shot bone-suppressed imaging is possible using post-image processing
combined with a filter obtained from training an artificial neural network. In this study, the authors investigate
the general properties of artificial neural network filters for bone-suppressed digital radiography. The filter
properties are characterized in terms of various parameters such as the size of input vector, the number of hidden
units, the learning rate, and so on. The preliminary result shows that the bone-suppressed image obtained from
the filter, which is designed with 5,000 teaching images from a single radiograph, results in about 95% similarity
with a commercial bone-enhanced image.
In this study, we fabricated the ultra-thin fiber-optic dosimeter (UTFOD) for high energy photon beam therapy dosimetry. The UTFOD has high spatial resolution due to the relatively small volume compared to conventional dosimeters therefore the UTFOD can measure depth doses precisely in build-up regions of therapeutic radiation beams. For 10 MV photon beams, we measured the scintillation signal generated from the UTFOD according to monitor units (MUs) and dose rates of the clinical linear accelerator (CLINAC). Also, we measured percentage depth doses (PDDs) at different depths of solid water phantoms using the UTFOD and the GAFCHROMIC® EBT films, and the results were compared with those using the Monte Carlo N-Particle eXtended (MCNPX) code.
In computed tomography (CT) imaging, radiat ion dose delivered to the patient is one of the major concerns. Many CT
developers and researchers have been making efforts to reduce radiat ion dose. Spars e-view CT takes project ions at
sparser view-angles and provides a viable option to reducing radiation dose. Sparse-view CT inspired by a compressive
sensing (CS) theory, which acquires sparsely sampled data in projection angles to reconstruct volumetric images of the
scanned object, is under active research for low-dose imaging. Also, region of interest (ROI) imaging method is one of
the reasonable approaches to reducing the integral dose to the patient and the risk of overdose. In this study, we
combined the two approaches together to achieve an ultra-low-dose imaging: a sparse-view imaging and the intensityweighted
region-of-interest (IWROI) imaging. IWROI imaging technique is particularly interesting because it can reduce
the imaging radiation dose substantially to the structures away from the imaging target, while allowing a stable solution
of the reconstruction problem in comparison with the interior problem. We used a total-variation (TV) minimization
algorithm that exploits the sparseness of the image derivative magnitude and can reconstruct images from sparse-view
data. In this study, we implemented an imaging mode that combines a sparse-view imaging and an ROI imaging. We
obtained promising results and believe that the proposed scanning approach can help reduce radiation dose to the patients
while preserving good quality images for applications such as image-guided radiation therapy. We are in progress of
applying the method to the real data.
For image-guided proton therapy, we investigated the feasibility of CBCT (cone-beam computed tomography) and
CBDT (cone-beam digital tomosynthesis) technologies in the gantry treatment room. A fully equipped x-ray projection
system, which was originally operated for patient alignment, in parallel to proton-beam direction was utilized for
acquiring CBCT/CBDT. The performance of the imaging detector was analyzed in terms of MTF (modulation-transfer
function), NPS (noise-power spectrum) and DQE (detective quantum efficiency). Tomographic imaging performances,
such as spatial resolving power, linearity of CT numbers, SNR (signal-to-noise ratio), and CNR (contrast-to-noise ratio),
were analyzed by using the AAPM (American Association of Physicists in Medicine) CT QC phantom. Geometric
alignment of CBCT/CBDT system was analyzed by using a calibration phantom, which consists of steal ball bearings.
The determined calibration parameters were applied to the image reconstruction procedures. The overall CBCT
performances of the system were demonstrated with reconstructed humanoid phantom images. In addition, we
implemented the CBDT with a selected number of projection views acquired for CBCT in limited angle ranges. From the
reconstructed phantom images, the CBCT system in the gantry treatment room will be very useful as a primary patient
alignment system for image-guided proton therapy. The CBDT may provide fast patient positioning with less motion
artifact and patient doses.
A colon polyp phantom, 28 cm long and 5 cm in diameter, was constructed by inflating a latex ultrasound transducer cover. Four round pieces of ham (3, 6, 9, 12 mm diameter) were imbedded in the outer membrane surface of the phantom and then were tied by string at the base to simulate pedunculated polyps. Three more pieces of ham (3, 6, 9 mm) were impressed and taped on the outer surface to simulate sessile polyps. The circumference of the phantom was constricted by string at four evenly spaced locations to simulate haustral folds. The phantom was placed in a water bath and was modified by infusing water into the lumen or by partially deflating the lumen, and then rescanned. CT images were obtained in a multi-slice CT (4x1 mm collimation, 0.5s scan, 120 Kvp, 90 mAs, 1 mm slice thickness). CT images were processed with our computer-aided detection program. First, the three-dimensional colonic boundary and inner structure were segmented. From this segmented region, soft-tissue structures were extracted and labeled to generate candidates. Shape features were evaluated along with geometric constraints. Three-dimensional region-growing and morphologic matching processes were applied to refine and classify the candidates. The detected polyps were compared with the true polyps in the phantom or known polyps in clinical cases to calculate the sensitivity and false positives.
Multi-row detector CT (MDCT) gated with ECG-tracing allows continuous image acquisition of the heart during a breath-hold with a high spatial and temporal resolution. Dynamic segmentation and display of CT images, especially short- and long-axis view, is important in functional analysis of cardiac morphology. The size of dynamic MDCT cardiac images, however, is typically very large involving several hundred CT images and thus a manual analysis of these images can be time-consuming and tedious. In this paper, an automatic scheme was proposed to segment and reorient the left ventricular images in MDCT. Two segmentation techniques, deformable model and region-growing methods, were developed and tested. The contour of the ventricular cavity was segmented iteratively from a set of initial coarse boundary points placed on a transaxial CT image and was propagated to adjacent CT images. Segmented transaxial diastolic cardiac phase MDCT images were reoriented along the long- and short-axis of the left ventricle. The axes were estimated by calculating the principal components of the ventricular boundary points and then confirmed or adjusted by an operator. The reorientation of the coordinates was applied to other transaxial MDCT image sets reconstructed at different cardiac phases. Estimated short-axes of the left ventricle were in a close agreement with the qualitative assessment by a radiologist. Preliminary results from our methods were promising, with a considerable reduction in analysis time and manual operations.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.