Search

Abstract This study proposes a synthetic data generation model to create a classification framework for cerebellar ataxia patients using trajectory data from the visuomotor adaptation task. The classification objectives include patients with cerebellar ataxia, age-matched normal individuals, and young healthy subjects. Synthetic data for the three classes is generated based on class conditions and random noise by leveraging a combination of conditional adversarial generative neural networks and reconstruction networks. This synthetic data, alongside real data, is utilized as training data for the patient classification model to enhance classification accuracy. The fidelity of the synthetic data is assessed visually to measure the validity and diversity of the generated data qualitatively while quantitatively evaluating distribution similarity to real data. Furthermore, the clinical efficacy of the patient classification model employing synthetic data is demonstrated by showcasing improved classification accuracy through a comparative analysis between results obtained using solely real data and those obtained when both real and synthetic data are utilized. This methodological approach holds promise in addressing data insufficiency in the digital healthcare domain, employing deep learning methodologies, and developing early disease diagnosis tools.

Abstract As an alternative to thermionic X‐ray generators, cold‐cathode X‐ray tubes are being developed for portable and multichannel tomography. Field emission propagating from needle structures such as carbon nanotubes or Si tips currently dominates related research and development, but various obstacles prevent the widespread of this technology. An old but simple electron emission design is the planar tunnelling cathode using a metal–oxide–semiconductor (MOS) structure, which achieves narrow beam dispersion and low supplying voltage. Directly grown vertical graphene (VG) is employed as the gate electrode of MOS and tests its potential as a new emission source. The emission efficiency of the device is initially ≈1% because of unavoidable fabrication damage during the patterning processes; it drastically improves to >40% after ozone treatment. The resulting emission current obeys the Fowler–Nordheim tunnelling model, and the enhanced emission is attributed to the effective gate thickness reduction by ozone treatment. As a proof‐of‐concept experiment, a clustered array of 2140 cells is integrated into a system that provides mA‐class emission current for X‐ray generation. With pulsed digital excitations, X‐ray imaging of a chest phantom, demonstrating the feasibility of using a VG MOS electron emission source as a new and innovative X‐ray generator is realized.

Recently, multi-modal research combining visual and textual information has emerged in Pedestrian Attribute Recognition (PAR). In this field, textual information has primarily been addressed through text modeling using tokenizers and textual encoders. However, separately learned visual and text encoders often find correlations between visual and textual features that may be insufficient at the human cognitive level. To address this issue, we drew inspiration from the way people describe pedestrian attributes and developed a method that mimics this cognitive process. This approach enhances visual encoders’ ability to discriminate by generating sentences from images, masking important words, and then reconstructing them. Our method, which improves visual pedestrian attributes using textual information, demonstrates significant performance enhancements on the RAP and PA100k datasets, as well as on zero-shot datasets like RAP2zs and PETAzs, which do not overlap with the training and test sets. These improvements yield more meaningful results.

Purpose This study analyzes the characteristics and prognosis of patients with de novo and recurrent metastatic gastric cancers. Methods This retrospective study included 301 advanced, pathologically confirmed gastric cancer patients who received palliative chemotherapy between 2012 and 2022. The de novo cohort included patients who presented with distant metastasis at diagnosis, and the recurrent metastatic cohort was composed of patients with metastasis after curative gastrectomy. We analyzed prognostic association by Cox regression and compared survival time of both cohorts using the Kaplan-Meier survival analysis. Results The study included 167 de novo and 112 recurrent patients. No differences were noted among the patients with de novo disease and recurrent metastatic disease concerning age, sex, Eastern Cooperative Oncology Group scores, primary cancer location, and pathologic features. Patients in the recurrent group versus the de novo group had a longer duration of chemotherapy (1.16±1.18 years vs. 0.85±0.84 years, P= 0.01) and lower mean body mass index (20.22±2.78 kg/m2 vs. 21.93±3.38 kg/m2, P< 0.001). The median overall survival in the de novo group was 11.6 versus 14.4 months in the recurrent group (P= 0.02). Conclusion De novo and recurrent metastatic gastric cancer are characterized by distinct subpopulations. Advanced gastric cancer patients with recurrence have significantly better survival versus those in the de novo cohort. The differences between de novo and recurrent gastric cancer patient outcomes could facilitate discussions about the selection of the clinical trial of each patient and help with their personalized treatment.

Heterogeneous data have been used to enhance behavior prediction performance; however, it involves issues such as missing data, which need to be addressed. This paper proposes enhanced pet behavior prediction via Sensor to Skeleton Generative Adversarial Networks (S2GAN)-based heterogeneous data synthesis. The S2GAN model synthesizes the key features of video skeletons based on collected nine-axis sensor data and replaces missing data, thereby enhancing the accuracy of behavior prediction. In this study, data collected from 10 pets in a real-life-like environment were used to conduct recognition experiments on 9 commonly occurring types of indoor behavior. Experimental results confirmed that the proposed S2GAN-based synthesis method effectively resolves possible missing data issues in real environments and significantly improves the performance of the pet behavior prediction model. Additionally, by utilizing data collected under conditions similar to the real environment, the method enables more accurate and reliable behavior prediction. This research demonstrates the importance and utility of synthesizing heterogeneous data in behavior prediction, laying the groundwork for applications in various fields such as abnormal behavior detection and monitoring.

Ensuring superior video quality is essential in various fields such as VFX film production, digital signage, media facades, product advertising, and interactive media, as it directly elevates the viewer’s engagement and experience. The ability to accurately quantify a video’s visual quality not only influences its valuation but is pivotal in maintaining high standards. Among the attributes influencing video quality, subjective quality stands out, however, several other elements also contribute significantly. Although automated video evaluations offer efficiency, there are situations necessitating expert editorial insight to measure nuanced subjective attributes. Our research primarily focuses on two prevalent issues undermining video quality: erratic camera motions and suboptimal focus. We employed a deep learning-driven optical flow technique to quantify inconsistent camera movements and adopted a Fast Fourier Transform (FFT)-based algorithm for blur detection. Moreover, our proposed adaptive threshold, grounded in statistical analysis, effectively delineates scenes as either desirable or substandard. Testing this framework on a diverse set of videos, we found it proficiently assessed video quality within a practical threshold range.

In this study, we propose a deep learning framework for multi-source deep data fusion and super-resolution for generative adversarial network-based spatiotemporal dependency learning to produce accurately downscaled sea surface temperature (SST) through simultaneously achieving error correction and improvement of spatial resolution. The proposed method is applied to the global ocean and the Korean waters, which is a regional sea, and experiments are conducted to downscale the SST by 2.5 and 5 times, respectively. The multi-source SST data used are numerical reanalysis, multiple satellite composites, and in-situ measurements, and two loss functions of super-resolution and mean square error are applied for adversarial learning. For more reliable performance evaluation, spatially, the global ocean and Korean waters are divided into a number of regional seas classified by characteristics of ocean physics, and temporally, the overall test period is divided into seasons and when extreme events occur. The overall results showed good performance for most experiments when both error correction through data fusion and spatiotemporal dependency learning from consecutive multiple input sequences using low-resolution reanalysis data, high-resolution satellite composite, and in-situ measurements were performed. However, for summer, winter, or extreme event periods, high performance was shown when using low-resolution satellite composite data with the same modality as the target data was used as an input. Furthermore, as a result of a blind test on the trained model with high-resolution target data used as target for the test period as input, the model that learned spatiotemporal dependency learning with error correction through data fusion showed the best and most consistent generalized downscaling performance compared to the test performance.

Abstract To learn spatiotemporal representations and anomaly predictions from geophysical data, we propose STANet, a spatiotemporal neural network with a trainable attention mechanism, and apply it to El Niño predictions for long-lead forecasts. The STANet makes two critical architectural improvements: it learns spatial features globally by expanding the network’s receptive field and encodes long-term sequential features with visual attention using a stateful long-short term memory network. The STANet conducts multitask learning of Nino3.4 index prediction and calendar month classification for predicted indices. In a comparison of the proposed STANet performance with the state-of-the-art model, the accuracy of the 12-month forecast lead correlation coefficient was improved by 5.8% and 13% for Nino3.4 index prediction and corresponding temporal classification, respectively. Furthermore, the spatially attentive regions for the strong El Niño events displayed spatial relationships consistent with the revealed precursor for El Niño occurrence, indicating that the proposed STANet provides good understanding of the spatiotemporal behavior of global sea surface temperature and oceanic heat content for El Niño evolution.

Weakly supervised temporal action localization (WTAL) aims to localize temporal intervals of actions in an untrimmed video using only video-level action labels. Although the learning of the background is an important issue in WTAL, most previous studies have not utilized an effective background. In this study, we propose a novel method for robustly separating contexts, e.g., action-like background, from the foreground to more accurately localize the action intervals. First, we detect background segments based on their probabilities to minimize the impact of background estimation errors. Second, we define the entropy boundary of the foreground and the positive distance between the boundary and background entropy. The background probability and entropy boundary allow the segment-level classifier to robustly learn the background. Third, we improve the performance of the overall actionness model based on a consensus of the RGB and flow features. The results of extensive experiments demonstrate that the proposed method learns the context separately from the action, consequently achieving new state-of-the-art results on the THUMOS-14 and ActivityNet-1.2 benchmarks. We also confirm that using feature adaptation helps overcome the limitation of a pretrained feature extractor on datasets that contain many backgrounds, such as THUMOS-14.

Abstract Accurate water surface elevation estimation is essential for understanding nearshore processes, but it is still challenging due to limitations in measuring water level using in-situ acoustic sensors. This paper presents a vision-based water surface elevation estimation approach using multi-view datasets. Specifically, we propose a visual domain adaptation method to build a water level estimator in spite of a situation in which ocean wave height cannot be measured directly. We also implemented a semi-supervised approach to extract wave height information from long-term sequences of wave height observations with minimal supervision. We performed wave flume experiments in a hydraulic laboratory with two cameras with side and top viewpoints to validate the effectiveness of our approach. The performance of the proposed models were evaluated by comparing the estimated time series of water elevation with the ground-truth wave gauge data at three locations along the wave flume. The estimated time series were in good agreement within the averaged correlation coefficient of 0.98 and 0.90 on the measurement and 0.95 and 0.85 on the estimation for regular and irregular waves, respectively.

Abstract Chronic cutaneous graft‐vs‐host disease (GVHD) has several atypical variants. We describe two cases of GVHD with clinical and histopathologic features of pityriasis rubra pilaris (PRP), which responded to additional immunosuppression. Recognition of this newly described PRP‐like clinical presentation of GVHD may prompt early consideration of additional steroid‐sparing therapies.

Introduction: Opioid overdose emergencies are increasing every year, naloxone is the antidote for the treatment of opioid overdose. Naloxone is being dispensed to even lay persons through some programs to prevent opioid overdose deaths. Case: 23 year old patient presented with naloxone treated opioid overdose complained of chest pain, pink frothy sputum production and shortness of breath. Physical exam showed tachycardia, tachypnea and coarse breath sounds. Imaging of the lungs showed diffuse pulmonary edema. Within an hour after the administration of naloxone patient developed pulmonary edema and lung injury. Patient was managed with non-invasive positive pressure ventilation which improved patient's symptoms in less than 6 hours confirmed by radiological improvement in 24–36 hrs. Discussion: There are no specific observation guidelines post naloxone treatment in opiate overdose patients. We recommend early treatment of naloxone induced pulmonary complications during the observation period with non-invasive positive pressure ventilation to reduce the morbidity.

Although various studies on monitoring dog behavior have been conducted, methods that can minimize or compensate data noise are required. This paper proposes multimodal data-based dog behavior recognition that fuses video and sensor data using a camera and a wearable device. The video data represent the moving area of dogs to detect the dogs. The sensor data represent the movement of the dogs and extract features that affect dog behavior recognition. Seven types of behavior recognition were conducted, and the results of the two data types were used to recognize the dog’s behavior through a fusion model based on deep learning. Experimentation determined that, among FasterRCNN, YOLOv3, and YOLOv4, the object detection rate and behavior recognition accuracy were the highest when YOLOv4 was used. In addition, the sensor data showed the best performance when all statistical features were selected. Finally, it was confirmed that the performance of multimodal data-based fusion models was improved over that of single data-based models and that the CNN-LSTM-based model had the best performance. The method presented in this study can be applied for dog treatment or health monitoring, and it is expected to provide a simple way to estimate the amount of activity.

Estimating wave-breaking indexes such as wave height and water depth is essential to understanding the location and scale of the breaking wave. Therefore, numerous wave-flume laboratory experiments have been conducted to develop empirical wave-breaking formulas. However, the nonlinearity between the parameters has not been fully incorporated into the empirical equations. Thus, this study proposes a multilayer neural network utilizing the nonlinear activation function and backpropagation to extract nonlinear relationships. Existing laboratory experiment data for the monochromatic regular wave are used to train the proposed network. Specifically, the bottom slope, deep-water wave height and wave period are plugged in as the input values that simultaneously estimate the breaking-wave height and wave-breaking location. Typical empirical equations employ deep-water wave height and length as input variables to predict the breaking-wave height and water depth. A newly proposed model directly utilizes breaking-wave height and water depth without nondimensionalization. Thus, the applicability can be significantly improved. The estimated wave-breaking index is statistically verified using the bias, root-mean-square errors, and Pearson correlation coefficient. The performance of the proposed model is better than existing breaking-wave-index formulas as well as having robust applicability to laboratory experiment conditions, such as wave condition, bottom slope, and experimental scale.

Satellite-based observations of sea wind are useful for forecasting marine weather and performing marine disaster management. Meteorological Operational Satellite-B (MetOp-B) is one of the satellites that provide wind products through a scatterometer named the Advanced Scatterometer (ASCAT). Since the linear regression method has been conventionally employed for calibrating remotely-sensed wind data, deviations and biases remain un-resolved to some degree. For coastal applications, these remotely-sensed wind data need to be calibrated again using local in-situ measurements in order to provide more accurate and realistic information. Thus, this study proposed a new method to calibrate ASCAT-based wind speed estimates using artificial neural networks. Herein, a deep neural network (DNN) model was applied. Wind databases collected during a period from 2012 to 2019 by the MetOp-B ASCAT and ten buoy stations in Korean seas were considered for deep learning-based calibration. ASCAT-based wind data and in-situ measurements were collocated in space and time. They were then separated into training and validation sets. A DNN model was designed and trained using multiple input variables such as observation location, sensing date and time, wind speed, and wind direction of the training set. The DNN-based best fit calibration model was evaluated using the validation set. The mean of biases between ASCAT-based and in-situ wind speeds was found to be decreased from 0.41 to 0.05 m/s on average for all buoy locations. The root mean squared error (RMSE) of wind speed was reduced from 1.38 m/s to 0.93 m/s. Moreover, the DNN-based calibration considerably improved the quality of wind speeds of less than 4 m/s, and of high wind speeds of 11–25 m/s. These results suggest that ASCAT-based observations can accurately represent real wind fields if a DNN-based calibration approach is applied.

We propose an unsupervised network with adversarial learning, the Raindrop-aware GAN, which enhances the quality of coastal video images contaminated by raindrops. Raindrop removal from coastal videos faces two main difficulties: converting the degraded image into a clean one by visually removing the raindrops, and restoring the background coastal wave information in the raindrop regions. The components of the proposed network—a generator and a discriminator for adversarial learning—are trained on unpaired images degraded by raindrops and clean images free from raindrops. By creating raindrop masks and background-restored images, the generator restores the background information in the raindrop regions alone, preserving the input as much as possible. The proposed network was trained and tested on an open-access dataset and directly collected dataset from the coastal area. It was then evaluated by three metrics: the peak signal-to-noise ratio, structural similarity, and a naturalness-quality evaluator. The indices of metrics are 8.2% (+2.012), 0.2% (+0.002), and 1.6% (−0.196) better than the state-of-the-art method, respectively. In the visual assessment of the enhanced video image quality, our method better restored the image patterns of steep wave crests and breaking than the other methods. In both quantitative and qualitative experiments, the proposed method more effectively removed the raindrops in coastal video and recovered the damaged background wave information than state-of-the-art methods.

In this paper, we propose a series of procedures for coastal wave-tracking using coastal video imagery with deep neural networks. It consists of three stages: video enhancement, hydrodynamic scene separation and wave-tracking. First, a generative adversarial network, trained using paired raindrop and clean videos, is applied to remove image distortions by raindrops and to restore background information of coastal waves. Next, a hydrodynamic scene of propagated wave information is separated from surrounding environmental information in the enhanced coastal video imagery using a deep autoencoder network. Finally, propagating waves are tracked by registering consecutive images in the quality-enhanced and scene-separated coastal video imagery using a spatial transformer network. The instantaneous wave speed of each individual wave crest and breaker in the video domain is successfully estimated through learning the behavior of transformed and propagated waves in the surf zone using deep neural networks. Since it enables the acquisition of spatio-temporal information of the surf zone though the characterization of wave breakers inclusively wave run-up, we expect that the proposed framework with the deep neural networks leads to improve understanding of nearshore wave dynamics.

Multiple hypothalamic neuronal populations that regulate energy balance have been identified. Although hypothalamic glia exist in abundance and form intimate structural connections with neurons, their roles in energy homeostasis are less known. Here we show that selective Ca2+ activation of glia in the mouse arcuate nucleus (ARC) reversibly induces increased food intake while disruption of Ca2+ signaling pathway in ARC glia reduces food intake. The specific activation of ARC glia enhances the activity of agouti-related protein/neuropeptide Y (AgRP/NPY)-expressing neurons but induces no net response in pro-opiomelanocortin (POMC)-expressing neurons. ARC glial activation non-specifically depolarizes both AgRP/NPY and POMC neurons but a strong inhibitory input to POMC neurons balances the excitation. When AgRP/NPY neurons are inactivated, ARC glial activation fails to evoke any significant changes in food intake. Collectively, these results reveal an important role of ARC glia in the regulation of energy homeostasis through its interaction with distinct neuronal subtype-specific pathways.

This paper introduces and explores the basic principles of psychodynamic approaches in music therapy. Music is used as a means to explore both conscious and unconscious issues as well as the internal world of the individuals involved in music therapy. However, the focus of therapy is on therapeutic relationship, especially the dynamics of transference and counter-transference between the client and the music therapist. Musical experiences, such as music listening, songs, and improvisation, can be used to facilitate the therapeutic processes, and to achieve individualized therapeutic goals. When clinically appropriate, verbal processing might play as crucial a role as the musical processing. Practitioners of psychodynamic approaches often strive to gain meaning and in-depth understandings from therapeutic experiences, and the approach is therefore suitable for individuals who are ready to work through their personal issues within a therapeutic relationship. Various approaches and techniques have been developed in psychotherapy as well as in music therapy. Perhaps the only commonality in these approaches is that psychodynamic thinking informs the direction of the therapy and therapeutic processes. Clinical vignettes will be introduced within the article to highlight a triadic dynamic—the client, the music therapist, and the music—in order to illustrate the core aspects of psychodynamic music therapy.

Cutaneous melanoma is a lethal malignancy that arises spontaneously or via in situ precursor neoplasms. While melanoma in situ and locally invasive malignant melanoma can be cured surgically, these lesions can sometimes be difficult to distinguish from melanocytic nevi. Thus, the identification of histolopathologic or molecular features that distinguish these biologically distinct lesions would represent an important advance. To this end, we determined the abundance of melanocytic primary cilia in a series of 62 cases composed of typical cutaneous melanocytic nevi, melanoma in situ, invasive melanoma, and metastatic melanoma. Primary cilia are sensory organelles that modulate developmental and adaptive signaling and notably, are substantially depleted from the neoplastic epithelium of pancreatic carcinoma at a stage equivalent to melanoma in situ. In this series, we find that while nearly all melanocytes in 22 melanocytic nevi possessed a primary cilium, a near-complete loss of this organelle was observed in 16 cases of melanoma in situ, in 16 unequivocal primary invasive melanomas, and in 8 metastatic tumors, each associated with a cutaneous primary lesion. These findings suggest that the primary cilium may be used to segregate cutaneous invasive melanoma and melanoma in situ from melanocytic nevi. Moreover, they place the loss of an organelle known to regulate oncogenic signaling at an early stage of melanoma development.

Therapy often involves the development of an intimate therapeutic relationship between the client and the therapist. By introducing the story of Hyun in music therapy, I invite readers to discover what happens during mutual music making process and what it means to the child and the therapist. I also situate this story within a triadic relationship – the child, the therapist and the family (especially the mother) so that one can see a comprehensive picture of how the happenings of music therapy relate to immediate outside world of music therapy, and vice versa. More importantly, I try to show the inner workings of the therapist and her wonderings in response to the client in order to clarify the core aspects of therapeutic process.