Background
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.
...Purpose
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.
Study Type
Retrospective study.
Subjects
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.
Field Strength/Sequence
Noncontrast‐enhanced 3D time‐of‐flight (TOF) MRA on 3T MR scanners.
Assessment
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.
Statistical Tests
Free‐response receiver operating characteristic (FROC) analysis.
Results
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.
Data Conclusion
We showed that the combination of a CNN and an MIP algorithm is useful for the detection of intracranial aneurysms.
Level of Evidence: 4
Technical Efficacy: Stage 1
J. Magn. Reson. Imaging 2018;47:948–953.
The purposes of this study are to propose an unsupervised anomaly detection method based on a deep neural network (DNN) model, which requires only normal images for training, and to evaluate its ...performance with a large chest radiograph dataset. We used the auto-encoding generative adversarial network (α-GAN) framework, which is a combination of a GAN and a variational autoencoder, as a DNN model. A total of 29,684 frontal chest radiographs from the Radiological Society of North America Pneumonia Detection Challenge dataset were used for this study (16,880 male and 12,804 female patients; average age, 47.0 years). All these images were labeled as “Normal,” “No Opacity/Not Normal,” or “Opacity” by board-certified radiologists. About 70% (6,853/9,790) of the Normal images were randomly sampled as the training dataset, and the rest were randomly split into the validation and test datasets in a ratio of 1:2 (7,610 and 15,221). Our anomaly detection system could correctly visualize various lesions including a lung mass, cardiomegaly, pleural effusion, bilateral hilar lymphadenopathy, and even dextrocardia. Our system detected the abnormal images with an area under the receiver operating characteristic curve (AUROC) of 0.752. The AUROCs for the abnormal labels Opacity and No Opacity/Not Normal were 0.838 and 0.704, respectively. Our DNN-based unsupervised anomaly detection method could successfully detect various diseases or anomalies in chest radiographs by training with only the normal images.
Display omitted
•The chemical structure of photodegradation product of azilsartan was determined by alternative synthesis and single-crystal X-ray diffraction.•The photodegradation product was a ...phenanthridine derivative.•The photolysis mechanism via an imidoylnitrene intermediate has been proposed.
Photodegradation of azilsartan produces a phenanthridine derivative, with its molecular structure determined by 1H and 13C NMR spectroscopy. This structure is confirmed by single-crystal X-ray diffraction and alternative synthesis. The phenanthridine ring formation is explained through the ring closure of an imidoylnitrene intermediate produced by decarboxylation of the 5-oxo-1,2,4-oxadiazole ring (oxadiazolone).
Purpose
To develop an anomaly detection system in PET/CT with the tracer
18
F-fluorodeoxyglucose (FDG) that requires only normal PET/CT images for training and can detect abnormal FDG uptake at any ...location in the chest region.
Materials and methods
We trained our model based on a Bayesian deep learning framework using 1878 PET/CT scans with no abnormal findings. Our model learns the distribution of standard uptake values in these normal training images and detects out-of-normal uptake regions. We evaluated this model using 34 scans showing focal abnormal FDG uptake in the chest region. This evaluation dataset includes 28 pulmonary and 17 extrapulmonary abnormal FDG uptake foci. We performed per-voxel and per-slice receiver operating characteristic (ROC) analyses and per-lesion free-response receiver operating characteristic analysis.
Results
Our model showed an area under the ROC curve of 0.992 on discriminating abnormal voxels and 0.852 on abnormal slices. Our model detected 41 of 45 (91.1%) of the abnormal FDG uptake foci with 12.8 false positives per scan (FPs/scan), which include 26 of 28 pulmonary and 15 of 17 extrapulmonary abnormalities. The sensitivity at 3.0 FPs/scan was 82.2% (37/45).
Conclusion
Our model trained only with normal PET/CT images successfully detected both pulmonary and extrapulmonary abnormal FDG uptake in the chest region.
Interest has recently grown in multi-center studies, which have more power than smaller studies in conducting sophisticated evaluations of basic neuroanatomy and neurodegenerative disorders. The ...large number of subjects that result from pooling multi-center datasets increases sensitivity, but also introduces a between-center variance component. Taking sex differences as an example, we examined the effects of different ratios of cases to controls (males to females) between scanners in multi-scanner morphometric studies, using voxel-based morphometry and data obtained on two scanners of the exact same model. Each subject was scanned twice with both scanners. Using the image obtained on either of the two scanners for each subject, voxel-based analyses were repeated with different ratios of males to females for each scanner. As the ratio of males to females became more imbalanced between the scanners, the differences between the two scanners more strongly affected the results of analyses of sex differences. When the ratio of males to females was balanced, the inclusion of scanner as a covariate in the statistical analysis had almost no influence on the results of analyses of sex differences. When the ratio of males to females was ill-balanced, the inclusion of scanner as a covariate suppressed scanner effects on the results, but made sex differences less likely to become significant. The present results suggest that as long as the ratio of cases to controls is well-balanced across different scanners, it is not always necessary to include scanner as a covariate in the statistical analysis, and that when the ratio of cases to controls is ill-balanced across scanners, the inclusion of scanner as a covariate in the statistical analysis can suppress scanner effects, but may make differences less likely to be detected.
•We examined the effects of different ratios of cases to controls between scanners.•We used VBM and MRI data obtained on two scanners of the exact same model.•The more imbalanced the sex ratio, the stronger scanner differences affected results.•Including scanner as a covariate suppressed the effects of scanner variability.•But it made differences less likely to be detected when the sex ratio was imbalanced.
Purpose
The performance of computer-aided detection (CAD) software depends on the quality and quantity of the dataset used for machine learning. If the data characteristics in development and ...practical use are different, the performance of CAD software degrades. In this study, we investigated changes in detection performance due to differences in training data for cerebral aneurysm detection software in head magnetic resonance angiography images.
Materials and methods
We utilized three types of CAD software for cerebral aneurysm detection in MRA images, which were based on 3D local intensity structure analysis, graph-based features, and convolutional neural network. For each type of CAD software, we compared three types of training pattern, which were two types of training using single-site data and one type of training using multisite data. We also carried out internal and external evaluations.
Results
In training using single-site data, the performance of CAD software largely and unpredictably fluctuated when the training dataset was changed. Training using multisite data did not show the lowest performance among the three training patterns for any CAD software and dataset.
Conclusion
The training of cerebral aneurysm detection software using data collected from multiple sites is desirable to ensure the stable performance of the software.
On account of their novel properties, bimetallic nanoparticles and nanoclusters (NCs) are strong potential candidates for optical, magnetic, and catalytic functional materials. These properties ...depend on the chemical composition and size (number of constituent atoms) of the NCs. Control of size, structure, and composition is particularly important for fabricating highly functional materials based on bimetallic NCs. Size- and structure-controlled synthesis of two-element alloys can reveal their intrinsic electronic synergistic effects. However, because synergistic enhancement of activity is strongly affected by composition as well as by size and structure, controlled synthesis is a challenging task, particularly in catalytic applications. To investigate catalytic synergistic effects, we have synthesized highly monodisperse, sub-2 nm, solid-solution AuPd NCs stabilized with poly(N-vinylpyrrolidone) (AuPd:PVP) using a newly developed ultrafine microfluidic mixing device with 15 μm wide multiple lamination channels. The synergistic enhancement for catalytic aerobic oxidation of benzyl alcohol exhibited a volcano-shaped trend, with a maximum at 20–65 at. % Pd. From X-ray photoelectron spectroscopic measurements, we confirmed that the enhanced activity originates from the enhanced electron density at the Au sites, donated by Pd sites.
Background:When the maximal diameter of an abdominal aortic aneurysm (AAA) exceeds a threshold, the likelihood of catastrophic rupture increases markedly. Therefore, surveillance at optimal intervals ...should be offered to patients with AAA. However, other than AAA diameter, there is no useful marker or index for predicting the expansion rate of an AAA or determining the optimal intervals for surveillance. The aim of this study was to evaluate the usefulness of calcium accumulation in the AAA for predicting its expansion rate.Methods and Results:We performed a retrospective cohort study in 414 patients with infrarenal AAA who visited The University of Tokyo Hospital. The maximal diameter and extent of calcification of each AAA were evaluated by multidetector-row computed tomography imaging. There was an inverse correlation between the extent of calcification and the subsequent AAA expansion. A lower extent of calcification in the AAA as well as the AAA diameter and absence of coronary artery disease correlated with an accelerated expansion of the AAA.Conclusions:In AAA, a lower extent of calcification correlated with accelerated expansion. The calcification index of an AAA can be a useful predictor of its expansion rate. The study findings also support the theory that the mechanisms for progression in atherosclerosis with calcification and external expansion of an aneurysm are distinct. (Circ J 2016; 80: 332–339)