In vivo confocal laser scanning microscopy (CLMS) using 830 nm diode laser has been reported inDermatology. CLMS is a device that can non-invasively and in real time obtain horizontal ...cross-sectionalimages up to around upper dermis (up to 300 μm). CLMS image has the characteristic that placeswith high refractive index and scattering coefficient are captured as high-brightness reflection images.Therefore, melanin is the best contrast source for CLMS image in human skin.In this time, not only facial benign pigmentation, such as solar lentigine, melasma, and acquired dermalmelanocytosis, but also extramammary Paget’s disease are reported.
Objectives
To investigate the clinical utility of the Vesical Imaging-Reporting and Data System (VI-RADS) by comparing its diagnostic performance for muscle-invasive bladder cancer (MIBC) between ...radiologists and urologists based on multiparametric MRI, including three-dimensional (3D) fast spin-echo (FSE) T2-weighted acquisitions.
Methods
This study included 66 treatment-naïve patients (60 men, 6 women; mean age 74.0 years) with pathologically proven bladder cancer who underwent multiparametric MRI, including 3D FSE T2-weighted imaging, before transurethral bladder tumour resection between January 2010 and November 2018. The MRI scans were categorised according to the five-point VI-RADS score by four independent readers (two board-certified radiologists and board-certified urologists each), blinded to the histopathological findings. The VI-RADS scores were compared with the postoperative histopathological diagnosis. Interobserver agreement was assessed using weighted kappa coefficients. ROC analysis and generalised estimating equations were used to evaluate the diagnostic performance.
Results
Forty-nine (74.2%) and 17 (25.8%) tumours were confirmed to be non-MIBC and MIBC, respectively, based on pathological examination. The interobserver agreement was good-to-excellent between all pairs of readers (range, 0.73–0.91). The urologists’ sensitivity/specificity values for DCE-MRI VI-RADS scores were significantly lower than those of radiologists. No significant differences were observed for the overall VI-RADS score. The AUC for the overall VI-RADS score was 0.94, 0.92, 0.89, and 0.87 for radiologists 1 and 2 and urologists 1 and 2, respectively.
Conclusions
The VI-RADS score, based on multiparametric MRI including 3D FSE T2-weighted acquisitions, can be useful for radiologists and urologists to determine the bladder cancer muscle invasion status preoperatively.
Key Points
•
VI-RADS (using multiparametric MRI including 3D FSE T2-weighted acquisitions) achieves good to excellent interobserver agreement and has similar diagnostic performance for detecting muscle invasion by both radiologists and urologists.
•
The diagnostic performance of the overall VI-RADS score is high for both radiologists and urologists, particularly due to the dominant effect of diffusion-weighted imaging on the overall VI-RADS score.
•
The sensitivity and specificity values of the T2WI VI-RADS scores for four readers in our study (using 3D FSE T2-weighted acquisitions) were similar (with slightly higher specificity values) to previously published results (using 2D FSE T2-weighted acquisitions).
Predicting of chemical compounds is one of the fundamental tasks in bioinformatics and chemoinformatics, because it contributes to various applications in metabolic engineering and drug discovery. ...The recent rapid growth of the amount of available data has enabled applications of computational approaches such as statistical modeling and machine learning method. Both a set of chemical interactions and chemical compound structures are represented as graphs, and various graph-based approaches including graph convolutional neural networks have been successfully applied to chemical network prediction. However, there was no efficient method that can consider the two different types of graphs in an end-to-end manner.
We give a new formulation of the chemical network prediction problem as a link prediction problem in a graph of graphs (GoG) which can represent the hierarchical structure consisting of compound graphs and an inter-compound graph. We propose a new graph convolutional neural network architecture called dual graph convolutional network that learns compound representations from both the compound graphs and the inter-compound network in an end-to-end manner.
Experiments using four chemical networks with different sparsity levels and degree distributions shows that our dual graph convolution approach achieves high prediction performance in relatively dense networks, while the performance becomes inferior on extremely-sparse networks.
Angiomyolipoma is the most common benign solid renal neoplasm observed in clinical practice. Once thought to be a hamartoma and almost always diagnosed by the imaged-based detection of fat, ...angiomyolipomas are now known to consist of a heterogeneous group of neoplasms. Although all are considered perivascular epithelioid cell tumors, many display different pathology, imaging features, and clinical behavior. The importance of understanding this group of neoplasms is emphasized by the fact that many types of angiomyolipoma contain little to no fat, and despite being benign, sometimes escape a pre-operative diagnosis. These types of angiomyolipomas can all be considered when encountering a renal mass that is both hyperattenuating relative to renal parenchyma on unenhanced CT and T2-hypointense, features that reflect their predominant smooth muscle component. We review recent developments and provide a radiological classification of angiomyolipomas that helps physicians understand the various types and learn how to both diagnose and manage them.
Intravenous urography has been widely used for the evaluation of upper tract urothelial carcinoma. However, computed tomography urography presently has a higher diagnostic accuracy for upper tract ...urothelial carcinoma (94.2–99.6%) than intravenous urography (80.8–84.9%), and has replaced intravenous urography as the first‐line imaging test for investigating patients with a high risk of upper tract urothelial carcinoma. Although the detection rate for bladder tumors using standard computed tomography urography is not yet high enough to replace cystoscopy, the addition of a 60‐ to 80‐s delayed scan after the administration of contrast material for the whole pelvis improves the detection rate. A drawback to computed tomography urography is the higher radiation dose of 15–35 mSv, compared with a mean effective dose of 5–10 mSv for intravenous urography. Among several approaches to reducing the radiation dose, the use of an iterative reconstruction algorithm is most likely to become an effective solution because of its simplicity. One advantage of computed tomography urography over intravenous urography is its ability to reliably differentiate between upper tract urothelial carcinoma and calculi or blood clots. Computed tomography urography also shows characteristic findings of other benign conditions. These findings, in combination with negative cytology, are very important diagnostic clues for avoiding an unnecessary nephroureterectomy. For the clinical staging, a recent study has reported the high diagnostic accuracy of computed tomography urography with respect to ≥pT3 tumors. The present review shows the current status of computed tomography urography for the evaluation of upper tract urothelial carcinoma.
Tumor contact length (TCL) is defined as the extent of contact between prostate cancer and the prostatic capsule, and its predictive value for microscopic extraprostatic extension (EPE) has been ...reported. However, the impact of the zonal origin (anterior or posterior tumor) of the tumor on the diagnosis of EPE is controversial.
We retrospectively analyzed the records of 233 consecutive patients who underwent preoperative MRI and radical prostatectomy. We designated their tumors as anterior or posterior, and evaluated the correlation between the TCL measured by MRI and microscopic EPE in the radical prostatectomy specimen. Then, we created the predicted probability curves for EPE versus TCL for anterior and posterior prostate cancer.
There were 109 patients (47%) with an anterior tumor and 124 patients (53%) with a posterior tumor. Postoperative pathological analysis confirmed pT3 in 18 patients (17%) with an anterior tumor and in 53 patients (43%) with a posterior tumor. Multivariate analysis demonstrated that the zonal origin of the tumor was an independent predictive factor for EPE. We developed separate probability curves of EPE versus TCL for anterior and posterior prostate cancer, which revealed that anterior tumors were less likely to invade the extraprostatic tissues. Among patients whose TCL was 10-20 mm, 9/32 patients (28%) with an anterior tumor had EPE compared with 24/45 patients (53%) with a posterior tumor (p = 0.036). The decision curve of this EPE predictive model had high clinical efficacy.
Our results indicate that anterior tumors have more favorable pathological characteristics than posterior tumors with the same TCL measured by MRI. We constructed two separate predicted probability curves for EPE after discriminating anterior and posterior tumors, which will be useful for decision making in clinical practice.
Renal cell carcinoma (RCC) is often found incidentally in asymptomatic individuals undergoing abdominal computed tomography (CT) examinations. The purpose of our study is to develop a deep ...learning-based algorithm for fully automated detection of small (≤4 cm) RCCs in contrast-enhanced CT images using a multicenter database and to evaluate its performance.
For the algorithmic detection of RCC, we retrospectively selected contrast-enhanced CT images of patients with histologically confirmed single RCC with a tumor diameter of 4 cm or less between January 2005 and May 2020 from 7 centers in the Japan Medical Image Database. A total of 453 patients from 6 centers were selected as dataset A, and 132 patients from 1 center were selected as dataset B. Dataset A was used for training and internal validation. Dataset B was used only for external validation. Nephrogenic phase images of multiphase CT or single-phase postcontrast CT images were used. Our algorithm consisted of 2-step segmentation models, kidney segmentation and tumor segmentation. For internal validation with dataset A, 10-fold cross-validation was applied. For external validation, the models trained with dataset A were tested on dataset B. The detection performance of the models was evaluated using accuracy, sensitivity, specificity, and the area under the curve (AUC).
The mean ± SD diameters of RCCs in dataset A and dataset B were 2.67 ± 0.77 cm and 2.64 ± 0.78 cm, respectively. Our algorithm yielded an accuracy, sensitivity, and specificity of 88.3%, 84.3%, and 92.3%, respectively, with dataset A and 87.5%, 84.8%, and 90.2%, respectively, with dataset B. The AUC of the algorithm with dataset A and dataset B was 0.930 and 0.933, respectively.
The proposed deep learning-based algorithm achieved high accuracy, sensitivity, specificity, and AUC for the detection of small RCCs with both internal and external validations, suggesting that this algorithm could contribute to the early detection of small RCCs.