[Evaluation of Diffusional Kurtosis Inference Using Synthetic q-space Learning and Bias Correction].

PURPOSE: In synthetic q-space learning (synQSL), which uses deep learning to infer the diffusional kurtosis (K), a bias that depends on the noise level added to the synthetic training data occurs. The purpose of this study was to evaluate K inference using synQSL and bias correction. METHODS: Using the synthetic test data and the real image data, K was inferred by synQSL, and bias correction was performed. Then, those results were compared with K inferred by fitting by the least-squares fitting (LSF) method. At this time, the noise level of the training data was set to 3 types, the noise level of the synthesis test data was set to 5 types, and the number of excitation (NEX) of the real image data was set to 4 types. Robustness of inference was evaluated by the outlier rate, which is the ratio of K outliers to the whole brain. We also evaluated the root mean square error (RMSE) of the inferred K. RESULTS: The outlier rate inferred by synQSL without correction was significantly lower in the test data of each noise level than that by the LSF method and was further reduced by correction. In addition, the RMSE of NEX 1 with NEX 4 as the correct answer based on the real image data had the smallest correction result of K by synQSL. CONCLUSION: Inferring K using synQSL and bias correction is a robust and small error method compared to that using the LSF method.

Temporal Progression Patterns of Brain Atrophy in Corticobasal Syndrome and Progressive Supranuclear Palsy Revealed by Subtype and Stage Inference (SuStaIn).

Differentiating corticobasal degeneration presenting with corticobasal syndrome (CBD-CBS) from progressive supranuclear palsy with Richardson's syndrome (PSP-RS), particularly in early stages, is often challenging because the neurodegenerative conditions closely overlap in terms of clinical presentation and pathology. Although volumetry using brain magnetic resonance imaging (MRI) has been studied in patients with CBS and PSP-RS, studies assessing the progression of brain atrophy are limited. Therefore, we aimed to reveal the difference in the temporal progression patterns of brain atrophy between patients with CBS and those with PSP-RS purely based on cross-sectional data using Subtype and Stage Inference (SuStaIn)-a novel, unsupervised machine learning technique that integrates clustering and disease progression modeling. We applied SuStaIn to the cross-sectional regional brain volumes of 25 patients with CBS, 39 patients with typical PSP-RS, and 50 healthy controls to estimate the two disease subtypes and trajectories of CBS and PSP-RS, which have distinct atrophy patterns. The progression model and classification accuracy of CBS and PSP-RS were compared with those of previous studies to evaluate the performance of SuStaIn. SuStaIn identified distinct temporal progression patterns of brain atrophy for CBS and PSP-RS, which were largely consistent with previous evidence, with high reproducibility (99.7%) under cross-validation. We classified these diseases with high accuracy (0.875) and sensitivity (0.680 and 1.000, respectively) based on cross-sectional structural brain MRI data; the accuracy was higher than that reported in previous studies. Moreover, SuStaIn stage correctly reflected disease severity without the label of disease stage, such as disease duration. Furthermore, SuStaIn also showed the genialized performance of differentiation and reflection for CBS and PSP-RS. Thus, SuStaIn has potential for improving our understanding of disease mechanisms, accurately stratifying patients, and providing prognoses for patients with CBS and PSP-RS.

Advantages of Using Both Voxel- and Surface-based Morphometry in Cortical Morphology Analysis: A Review of Various Applications.

Surface-based morphometry (SBM) is extremely useful for estimating the indices of cortical morphology, such as volume, thickness, area, and gyrification, whereas voxel-based morphometry (VBM) is a typical method of gray matter (GM) volumetry that includes cortex measurement. In cases where SBM is used to estimate cortical morphology, it remains controversial as to whether VBM should be used in addition to estimate GM volume. Therefore, this review has two main goals. First, we summarize the differences between the two methods regarding preprocessing, statistical analysis, and reliability. Second, we review studies that estimate cortical morphological changes using VBM and/or SBM and discuss whether using VBM in conjunction with SBM produces additional values. We found cases in which detection of morphological change in either VBM or SBM was superior, and others that showed equivalent performance between the two methods. Therefore, we concluded that using VBM and SBM together can help researchers and clinicians obtain a better understanding of normal neurobiological processes of the brain. Moreover, the use of both methods may improve the accuracy of the detection of morphological changes when comparing the data of patients and controls.In addition, we introduce two other recent methods as future directions for estimating cortical morphological changes: a multi-modal parcellation method using structural and functional images, and a synthetic segmentation method using multi-contrast images (such as T1- and proton density-weighted images).

Recent Advances in Parameter Inference for Diffusion MRI Signal Models.

In this paper, fundamentals and recent progress for obtaining biological features quantitatively by using diffusion MRI are reviewed. First, a brief description of diffusion MRI history, application, and development was presented. Then, well-known parametric models including diffusion tensor imaging (DTI), diffusional kurtosis imaging (DKI), and neurite orientation dispersion diffusion imaging (NODDI) are introduced with several classifications in various viewpoints with other modeling schemes. In addition, this review covers mathematical generalization and examples of methodologies for the model parameter inference from conventional fitting to recent machine learning approaches, which is called Q-space learning (QSL). Finally, future perspectives on diffusion MRI parameter inference are discussed with the aspects of imaging modeling and simulation.

Prospective Study of Spatial Distribution of Missed Lung Nodules by Readers in CT Lung Screening Using Computer-assisted Detection.

PURPOSE: To evaluate the spatial patterns of missed lung nodules in a real-life routine screening environment. MATERIALS AND METHODS: In a screening institute, 4,822 consecutive adults underwent chest CT, and each image set was independently interpreted by two radiologists in three steps: (1) independently interpreted without computer-assisted detection (CAD) software, (2) independently referred to the CAD results, (3) determined by the consensus of the two radiologists. The locations of nodules and the detection performance data were semi-automatically collected using a CAD server integrated into the reporting system. Fisher's exact test was employed for evaluating findings in different lung divisions. Probability maps were drawn to illustrate the spatial distribution of radiologists' missed nodules. RESULTS: Radiologists significantly tended to miss lung nodules in the bilateral hilar divisions (p < 0.01). Some radiologists had their own spatial pattern of missed lung nodules. CONCLUSION: Radiologists tend to miss lung nodules present in the hilar regions significantly more often than in the rest of the lung.

Novel platform for development, training, and validation of computer-assisted detection/diagnosis software.

PURPOSE: To build a novel, open-source, purely web-based platform system to address problems in the development and clinical use of computer-assisted detection/diagnosis (CAD) software. The new platform system will replace the existing system for the development and validation of CAD software, Clinical Infrastructure for Radiologic Computation of United Solutions (CIRCUS). METHODS: In our new system, the two top-level applications visible to users are the web-based image database (CIRCUS DB; database) and the Docker plug-in-based CAD execution platform (CIRCUS CS; clinical server). These applications are built on top of a shared application programming interface server, a three-dimensional image viewer component, and an image repository. RESULTS: We successfully installed our new system into a Linux server at two clinical sites. A total of 1954 cases were registered in CIRCUS DB. We have been utilizing CIRCUS CS with four Docker-based CAD plug-ins. CONCLUSIONS: We have successfully built a new version of the CIRCUS system. Our platform was successfully implemented at two clinical sites, and we plan to publish it as an open-source software project.

Pilot study of eruption forecasting with muography using convolutional neural network.

Muography is a novel method of visualizing the internal structures of active volcanoes by using high-energy near-horizontally arriving cosmic muons. The purpose of this study is to show the feasibility of muography to forecast the eruption event with the aid of the convolutional neural network (CNN). In this study, seven daily consecutive muographic images were fed into the CNN to compute the probability of eruptions on the eighth day, and our CNN model was trained by hyperparameter tuning with the Bayesian optimization algorithm. By using the data acquired in Sakurajima volcano, Japan, as an example, the forecasting performance achieved a value of 0.726 for the area under the receiver operating characteristic curve, showing the reasonable correlation between the muographic images and eruption events. Our result suggests that muography has the potential for eruption forecasting of volcanoes.

Cone-beam CT reconstruction for non-periodic organ motion using time-ordered chain graph model.

PURPOSE: The purpose of this study is to introduce the new concept of a four-dimensional (4D) cone-beam computed tomography (CBCT) reconstruction approach for non-periodic organ motion in cooperation with the time-ordered chain graph model (TCGM) and to compare it with previously developed methods such as total variation-based compressed sensing (TVCS) and prior-image constrained compressed sensing (PICCS). MATERIALS AND METHODS: Our proposed reconstruction is based on a model including the constraint originating from the images of neighboring time phases. Namely, the reconstructed time-series images depend on each other in this TCGM scheme, and the time-ordered images are concurrently reconstructed in the iterative reconstruction approach. In this study, iterative reconstruction with the TCGM was carried out with 90° projection ranges. The images reconstructed by the TCGM were compared with the images reconstructed by TVCS (200° projection ranges) and PICCS (90° projection ranges). Two kinds of projection data sets-an elliptic-cylindrical digital phantom and two clinical patients' data-were used. For the digital phantom, an air sphere was contained and virtually moved along the longitudinal axis by 3 cm/30 s and 3 cm/60 s; the temporal resolution was evaluated by measuring the penumbral width of the air sphere. The clinical feasibility of the non-periodic time-ordered 4D CBCT image reconstruction was examined with the patient data in the pelvic region. RESULTS: In the evaluation of the digital-phantom reconstruction, the penumbral widths of the TCGM yielded the narrowest result; the results obtained by PICCS and TCGM using 90° projection ranges were 2.8% and 18.2% for 3 cm/30 s, and 5.0% and 23.1% for 3 cm/60 s narrower than that of TVCS using 200° projection ranges. This suggests that the TCGM has a better temporal resolution, whereas PICCS seems similar to TVCS. These reconstruction methods were also compared using patients' projection data sets. Although all three reconstruction results showed motion related to rectal gas or stool, the result obtained by the TCGM was visibly clearer with less blurring. CONCLUSION: The TCGM is a feasible approach to visualize non-periodic organ motion. The digital-phantom results demonstrated that the proposed method provides 4D image series with a better temporal resolution compared to TVCS and PICCS. The clinical patients' results also showed that the present method enables us to visualize motion related to rectal gas and flatus in the rectum.

Evaluation of glymphatic system activity with the diffusion MR technique: diffusion tensor image analysis along the perivascular space (DTI-ALPS) in Alzheimer's disease cases.

PURPOSE: The activity of the glymphatic system is impaired in animal models of Alzheimer's disease (AD). We evaluated the activity of the human glymphatic system in cases of AD with a diffusion-based technique called diffusion tensor image analysis along the perivascular space (DTI-ALPS). MATERIALS AND METHODS: Diffusion tensor images were acquired to calculate diffusivities in the x, y, and z axes of the plane of the lateral ventricle body in 31 patients. We evaluated the diffusivity along the perivascular spaces as well as projection fibers and association fibers separately, to acquire an index for diffusivity along the perivascular space (ALPS-index) and correlated them with the mini mental state examinations (MMSE) score. RESULTS: We found a significant negative correlation between diffusivity along the projection fibers and association fibers. We also observed a significant positive correlation between diffusivity along perivascular spaces shown as ALPS-index and the MMSE score, indicating lower water diffusivity along the perivascular space in relation to AD severity. CONCLUSION: Activity of the glymphatic system may be evaluated with diffusion images. Lower diffusivity along the perivascular space on DTI-APLS seems to reflect impairment of the glymphatic system. This method may be useful for evaluating the activity of the glymphatic system.

Automatic detection of vertebral number abnormalities in body CT images.

PURPOSE: The anatomical anomaly of the number of vertebral bones is one of the major anomalies in the human body, which can cause confusion of the spinal level in, for example, surgery. The aim of this study is to develop an automatic detection system for this type of anomaly. METHODS: We utilized our previously reported anatomical landmark detection system for this anomaly detection problem. This system uses a landmark point distribution model (L-PDM) to find multiple landmark positions. The L-PDM is a statistical probabilistic model of all landmark positions in the human body, including five landmarks for each vertebra. Given a new volume, the proposed algorithm applies five hypotheses (normal, 11 or 13 thoracic vertebrae, 4 or 6 lumbar vertebrae) to the given spine and attempts to detect all the landmarks. Then, the most plausible hypothesis with the largest posterior likelihood is selected as the anatomy detection result. RESULTS: The proposed method was evaluated using 300 neck-to-pelvis CT datasets. For normal subjects, the vertebrae of 211/217 (97.2%) of the subjects were successfully determined as normal. For subjects with 23 or 25 vertebrae without a transitional vertebra (TV), the vertebrae of 9/10 (90%) of the subjects were successfully determined. For subjects with TV, the vertebrae of 71/73 (97.3%) of subjects were judged as partially successfully determined. CONCLUSION: Our algorithm successfully determined the number of vertebrae, and the feasibility of our proposed system was validated.

Automatic detection of over 100 anatomical landmarks in medical CT images: A framework with independent detectors and combinatorial optimization.

An automatic detection method for 197 anatomically defined landmarks in computed tomography (CT) volumes is presented. The proposed method can handle missed landmarks caused by detection failure, a limited imaging range and other problems using a novel combinatorial optimization framework with a two-stage sampling algorithm. After a list of candidates is generated by each landmark detector, the best combination of candidates is searched for by a combinatorial optimization algorithm using a landmark point distribution model (L-PDM) to provide prior knowledge. Optimization is performed by simulated annealing and iterative Gibbs sampling. Prior to each cycle of Gibbs sampling, another sampling algorithm is processed to estimate the spatial distribution of each target landmark, so that landmark positions without any correct detector-derived candidates can be estimated. The proposed method was evaluated using 104 CT volumes with various imaging ranges. The overall average detection distance error was 6.6mm, and 83.8, 93.2 and 96.5% of landmarks were detected within 10, 15 and 20mm from the ground truth, respectively. The proposed method worked even when most of the landmarks were outside of the imaging range. The identification accuracy of the vertebral centroid was also evaluated using public datasets and the proposed method could identify 70% of vertebrae including severely diseased ones. From these results, the feasibility of our framework in detecting multiple landmarks in various CT datasets was validated.

Alterations of the optic pathway between unilateral and bilateral optic nerve damage in multiple sclerosis as revealed by the combined use of advanced diffusion kurtosis imaging and visual evoked potentials.

OBJECTIVES: We investigated changes in the optic tract and optic radiation in patients with multiple sclerosis (MS) by comparing unilateral and bilateral optic nerve damage assessed based on visual evoked potentials (VEPs) using advanced diffusion MR metrics. METHODS: In 21 MS patients, diffusion MRI was performed. Maps of fractional anisotropy, apparent diffusion coefficient (ADC), and mean kurtosis (MK) were computed. On the basis of the P100 latency in VEPs, the MS patients were divided into three groups: bilateral (n=7), unilateral (n=7), and no abnormality (n=7). Their optic tracts and optic radiations were analyzed with diffusion MRI-based fiber tracking. We also investigated the correlations between diffusion parameters and VEPs (n=21). RESULTS: In the optic tract, the diffusion changes in each of the three groups showed step-like changes. The diffusion changes in the optic radiations of the unilateral group were similar to those in the normal VEP group. Only the bilateral group showed significantly higher ADC and lower MK relative to the other two groups (P<0.05, Steel-Dwass multiple-comparison test). A significant positive correlation between VEP latency and ADC and a significant negative correlation between VEP latency and MK were observed (P<0.01, Spearman's correction). CONCLUSIONS: We first evaluated the relationship between VEPs and DKI and concluded that the lateral geniculate nucleus may compensate for unilateral damage in the pre-geniculate optic pathway via neural plasticity.

Landmark-guided diffeomorphic demons algorithm and its application to automatic segmentation of the whole spine and pelvis in CT images.

PURPOSE: A fully automatic multiatlas-based method for segmentation of the spine and pelvis in a torso CT volume is proposed. A novel landmark-guided diffeomorphic demons algorithm is used to register a given CT image to multiple atlas volumes. This algorithm can utilize both grayscale image information and given landmark coordinate information optimally. METHODS: The segmentation has four steps. Firstly, 170 bony landmarks are detected in the given volume. Using these landmark positions, an atlas selection procedure is performed to reduce the computational cost of the following registration. Then the chosen atlas volumes are registered to the given CT image. Finally, voxelwise label voting is performed to determine the final segmentation result. RESULTS: The proposed method was evaluated using 50 torso CT datasets as well as the public SpineWeb dataset. As a result, a mean distance error of [Formula: see text] and a mean Dice coefficient of [Formula: see text] were achieved for the whole spine and the pelvic bones, which are competitive with other state-of-the-art methods. CONCLUSION: From the experimental results, the usefulness of the proposed segmentation method was validated.

Time Course of Diffusion Kurtosis in Cerebral Infarctions of Transient Middle Cerebral Artery Occlusion Rat Model.

OBJECTIVE: To evaluate the relationship between fiber bundle direction and changes in diffusion kurtosis, we evaluated the apparent diffusion kurtosis coefficients (AKCs) that were perpendicular to and parallel to the principal diffusion tensor eigenvector. MATERIALS AND METHODS: Adult male Wistar rats were subjected to 30 or 60 minutes of middle cerebral artery occlusion and imaged with a 7T Magnetic Resonance Imager System (Varian MRI System 7T/210: Agilent Technologies, CA). Diffusion kurtosis images were obtained before middle cerebral artery (MCA) reperfusion and 3, 6, and 24 hours after reperfusion to generate the apparent diffusion coefficient (ADC), fractional anisotropy (FA), mean apparent diffusion kurtosis coefficient (mAKC), AKC axial to the eigenvector (axAKC), and AKC radial to the eigenvector (radAKC) images. The time course of the region/normal ratio was evaluated for the above parameters in the caudoputamen and white matter. RESULTS: Relative FA and relative ADC values decreased 3 hours after MCA reperfusion and remained decreased until 24 hours. Relative mAKC, axAKC, and radAKC values were increased 3 hours after MCA reperfusion, peaked after 6 hours, and slightly decreased after 24 hours. In the white matter, axAKC showed larger changes than radAKC. CONCLUSION: The time course of the diffusion kurtosis value showed earlier pseudonormalization than the ADC value of the lesions. For white matter lesions, the increase in axAKC was larger than that in radAKC, suggesting that the tissue changes after infarction mainly produce reduced diffusivity along the fibers and lead to increased inhomogeneity of the diffusion.

Machine Learning for Computer-aided Diagnosis.

Machine learning algorithms are to analyze any dataset to extract data-driven model, prediction rule, or decision rule from the dataset. Various machine learning algorithms are now used to develop high-performance medical image processing systems such as computer-aided detection (CADe) system which detects clinically significant objects from medical images and computer-aided diagnosis (CADx) system which quantifies malignancy of manually or automatically detected clinical objects. In this paper, we introduce some applications of machine learning algorithms to the development of medical image processing system.

Reconstruction of the treatment area by use of sinogram in helical tomotherapy.

BACKGROUND: TomoTherapy (Accuray, USA) has an image-guided radiotherapy system with a megavoltage (MV) X-ray source and an on-board imaging device. This system allows one to acquire the delivery sinogram during the actual treatment, which partly includes information from the irradiated object. In this study, we try to develop image reconstruction during treatment with helical tomotherapy. FINDINGS: Sinogram data were acquired during helical tomotherapy delivery using an arc-shaped detector array that consists of 576 xenon-gas filled detector cells. In preprocessing, these were normalized with full air-scan data. A software program was developed that reconstructs 3D images during treatment with corrections as; (1) the regions outside the field were masked not to be added in the backprojection (a masking correction), and (2) each voxel of the reconstructed image was divided by the number of the beamlets passing through its voxel (a ray-passing correction). The masking correction produced a reconstructed image, however, it contained streak artifacts. The ray-passing correction reduced this artifact. Although the SNR (the ratio of mean to standard deviation in a homogeneous region) and the contrast of the reconstructed image were slightly improved with the ray-passing correction, use of only the masking correction was sufficient for the visualization purpose. CONCLUSIONS: The visualization of the treatment area was feasible by using the sinogram in helical tomotherapy. This proposed method would be useful in the treatment verification.

Corticospinal tract-sparing intensity-modulated radiotherapy treatment planning.

AIM: To establish intensity-modulated radiotherapy (IMRT) planning procedures that spare the corticospinal tract by integrating diffusion tensor tractography into the treatment planning software. BACKGROUND: Organs at risk are generally contoured according to the outline of the organ as demonstrated by CT or MRI. But a part of the organ with specific function is difficult to protect, because such functional part of the organ cannot be delineated on CT or conventional sequence of MRI. METHODS: Diagnostic and treatment planning images of glioblastoma patients who had been treated by conventional 3-dimensional conformal radiotherapy were used for re-planning of IMRT. Three-dimensional fiber maps of the corticospinal tracts were created from the diffusion tensors obtained from the patients before the surgery, and were blended with the anatomical MR images (i.e. gadolinium-enhanced T1-weighted images or T2-weighted images). DICOM-formatted blended images were transferred and fused to the planning CT images. Then, IMRT plans were attempted. RESULTS: The corticospinal tracts could be contoured as organs at risk (OARs), because the blended images contained both anatomical information and fiber-tract maps. Other OARs were contoured in a way similar to that of ordinary IMRT planning. Gross tumor volumes, clinical target volumes, planning target volumes, and other OARs were contoured on the treatment planning software, and IMRT plans were made. CONCLUSIONS: IMRT plans with diminished doses to the corticospinal tract were attained. This technique enabled us to spare specific neuron fibers as OARs which were formerly "invisible" and to reduce the probability of late morbidities.

Development of automatic visceral fat volume calculation software for CT volume data.

OBJECTIVE: To develop automatic visceral fat volume calculation software for computed tomography (CT) volume data and to evaluate its feasibility. METHODS: A total of 24 sets of whole-body CT volume data and anthropometric measurements were obtained, with three sets for each of four BMI categories (under 20, 20 to 25, 25 to 30, and over 30) in both sexes. True visceral fat volumes were defined on the basis of manual segmentation of the whole-body CT volume data by an experienced radiologist. Software to automatically calculate visceral fat volumes was developed using a region segmentation technique based on morphological analysis with CT value threshold. Automatically calculated visceral fat volumes were evaluated in terms of the correlation coefficient with the true volumes and the error relative to the true volume. RESULTS: Automatic visceral fat volume calculation results of all 24 data sets were obtained successfully and the average calculation time was 252.7 seconds/case. The correlation coefficients between the true visceral fat volume and the automatically calculated visceral fat volume were over 0.999. CONCLUSIONS: The newly developed software is feasible for calculating visceral fat volumes in a reasonable time and was proved to have high accuracy.

Fast and robust estimation of diffusional kurtosis imaging (DKI) parameters by general closed-form expressions and their extensions.

Diffusional kurtosis imaging (DKI) for clinical imaging involves time-consuming computation and demonstrates low robustness. Standard estimation of DKI parameters is based on an extension of Stejskal-Tanner's signal model with squared b-value term and is a least-squares fitting problem. The use of numerical methods for computation requires time, and estimation of DKI parameters is noise sensitive and often produces noisy results, such as images with pepper noise.In this study, we propose general closed-form solutions for DKI parameters to avoid numerical computation for least-squares fitting, solutions that can be applied to diffusion weighted imaging (DWI) datasets with any number of b-values more than three. Solutions are obtained through stationary-point conditions of an objective function that are minimized for fitting. We use 3 techniques to extend the solutions to increase robustness-b-value-dependent weighting in fitting, removal of outliers, and addition of neighbor sampling. Based on synthetic datasets and clinical datasets that both consist of 6 b-value and 3 b-value datasets, we detail and compare the 3 methods including a method by Jensen et al. are compared and investigated in detail. The synthetic data consist of several combinations of DKI parameters and some Rician noise. In addition to visually assessing result images, we also performed quantitative evaluation using a range of estimated parameters, positive-definiteness of the objective function for fitting, and root-mean-square error including estimation bias from the true value (synthetic data only). Methods that added neighbor sampling outperformed others in terms of low errors and visual smoothness. Though the solution by our method is to estimate DKI parameters in a single MPG direction, it can contribute to anisotropic analysis of diffusional kurtosis such as kurtosis tensor. More robust estimation is expected by combining techniques.