Explore the vast range of titles available.

There is a growing need for algorithms to automatically detect objects in satellite images. Object detection algorithms using deep learning have demonstrated a significant improvement in object detection performance. However, deep-learning models have difficulty in interpreting the features for inference. This difficulty is practically problematic when analyzing earth-observation images, which are often used as evidence for public decision-making. In addition, for the same reason, it is difficult to set an explicit policy or criteria to improve the models. To deal with these challenges, we introduce a feature attribution method that defines an approximate model and calculates the attribution of input features to the output of a deep-learning model. For the object detection models of satellite images with complex textures, we propose a method to visualize the basis of inference using pixel-wise feature attribution. Furthermore, we propose new methods for model evaluation, regularization, and data selection, based on feature attribution. Experimental results demonstrate the feasibility of the proposed methods for basis visualization and model evaluation. Moreover, the results illustrate that the model using the proposed regularization method can avoid over-fitting and achieve higher performance, and the proposed data selection method allows for the efficient selection of new training data.

With the spread of aerial laser bathymetry (ALB), seafloor topographies are being measured more frequently. Nevertheless, data deficiencies occur owing to seawater conditions and other factors. Conventional interpolation methods generally need to produce digital elevation models (DEMs) with sufficient accuracy. If the topographic features are considered as a basis, the DEM should be reproducible based on a combination of such features. The purpose of this study is to develop new DEM generation methods based on sparse modeling. Based on a review of the definitions of sparsity, we developed DEM generation methods based on a discrete cosine transform (DCT), DCT with elastic net, K-singular value decomposition (K-SVD), Fourier regularization, wavelet regularization, and total variation (TV) minimization, and conducted a comparative analysis. The developed methods were applied to artificially deficient DEM and ALB data, and their accuracy was evaluated. Thus, as a conclusion, we can confirm that the K-SVD method is appropriate when the percentage of deficiencies is low, and that the TV minimization method is appropriate when the percentage of deficiencies is high. Based on these results, we also developed a method integrating both methods and achieved an RMSE of 0.128 m.

The extensive monitoring of shorelines is becoming important for investigating the impact of coastal erosion. Satellite synthetic aperture radar (SAR) images can cover wide areas independently of weather or time. The recent development of high-resolution satellite SAR images has made observations more detailed. Shoreline extraction using high-resolution images, however, is challenging because of the influence of speckle, crest lines, patterns in sandy beaches, etc. We develop a shoreline extraction method based on the spatial pattern analysis of satellite SAR images. The proposed method consists of image decomposition, smoothing, sea and land area segmentation, and shoreline refinement. The image decomposition step, in which the image is decomposed into its texture and outline components, is based on morphological component analysis. In the image decomposition step, a learning process involving spatial patterns is introduced. The outline images are smoothed using a non-local means filter, and then the images are segmented into sea and land areas using the graph cuts’ technique. The boundary between these two areas can be regarded as the shoreline. Finally, the snakes algorithm is applied to refine the position accuracy. The proposed method is applied to the satellite SAR images of coasts in Japan. The method can successfully extract the shorelines. Through experiments, the performance of the proposed method is confirmed.

Three-dimensional (3D) road maps have garnered significant attention recently because of applications such as autonomous driving. For 3D road maps to remain accurate and up-to-date, an appropriate updating method is crucial. However, there are currently no updating methods with both satisfactorily high frequency and accuracy. An effective strategy would be to frequently acquire point clouds from regular vehicles, and then take detailed measurements only where necessary. However, there are three challenges when using data from regular vehicles. First, the accuracy and density of the points are comparatively low. Second, the measurement ranges vary for different measurements. Third, tentative changes such as pedestrians must be discriminated from real changes. The method proposed in this paper consists of registration and change detection methods. We first prepare the synthetic data obtained from regular vehicles using mobile mapping system data as a base reference. We then apply our proposed change detection method, in which the occupancy grid method is integrated with Dempster–Shafer theory to deal with occlusions and tentative changes. The results show that the proposed method can detect road environment changes, and it is easy to find changed parts through visualization. The work contributes towards sustainable updates and applications of 3D road maps.

Physiological parameters are crucial for the caring of trauma patients. There is a significant loss of prehospital vital signs data of patients during handover between prehospital and in-hospital teams. Effective strategies for reducing the loss remain a challenging research area. We tested whether the newly developed electronic automated prehospital vital signs chart sharing system would increase the amount of prehospital vital signs data shared with a remote trauma center prior to hospital arrival. Fifty trauma patients, transferred to a level I trauma center in Japan, were studied. The primary outcome variable was the number of prehospital vital signs shared with the trauma center prior to hospital arrival. The prehospital vital signs chart sharing system significantly increased the number of prehospital vital signs, including blood pressure, heart rate, and oxygen saturation, shared with the in-hospital team at a remote trauma center prior to patient arrival at the hospital (P < .0001). There were significant differences in prehospital vital signs during ambulance transfer between patients who had severe bleeding and non–severe bleeding within 24 hours after injury onset. Vital signs data collected during ambulance transfer via patient monitors could be automatically converted to easily visible patient charts and effectively shared with the remote trauma center prior to hospital arrival. The prehospital vital signs chart sharing system increased the number of precise vital signs shared prior to patient arrival at the hospital, which can potentially contribute to better trauma care without increasing labor and reduce information loss during clinical handover.

Hereditary angioedema (HAE) is a rare genetic disease caused by a deficiency of functional C1 esterase inhibitor that causes swelling attacks in various body tissues. We hereby report a case of out-of-hospital cardiac arrest due to airway obstruction in HAE. Cutaneous swelling and abdominal pain attacks caused by gastrointestinal wall swelling are common symptoms in HAE, whereas laryngeal swelling is rare. Emergency physicians may have few chances to experience cases of life-threatening laryngeal edema resulting in a delay from symptom onset to the diagnosis of HAE. Hereditary angioedema is diagnosed by performing complement blood tests. Because safe and effective treatment options are available for the life-threatening swellings in HAE, the diagnosis potentially reduces the risk of asphyxiation in patients and their blood relatives.

Liver abscess remains a life-threatening disease, particularly when it results in systemic organ failure necessitating intensive care. Only few cases of respiratory failure caused by liver abscess and treated with veno-venous extracorporeal membrane oxygenation (ECMO) have been reported. Here we present a case of liver abscess with rapid progression of multiple organ dysfunction, including severe acute respiratory failure on admission to the intensive care unit (ICU). Upon admission, we immediately initiated artificial organ support systems, including ventilator, continuous renal replacement therapy, and cardiovascular drug infusion for septic multiple organ failure and source control. Despite this initial management, respiratory failure deteriorated and V-V ECMO was introduced. The case developed abdominal compartment syndrome, for which we performed a bedside decompressive laparotomy in the ICU. The case gradually recovered from multiple organ failure and was discharged from the ICU on day 22 and from the hospital on day 53. Since liver abscess is potentially lethal and respiratory failure on admission is an additional risk factor of mortality, V-V ECMO may serve as an adjunctive choice of artificial organ support for cases of severe acute respiratory failure caused by liver abscess.