Bi-Directional Long Short-Term Memory-Based Gait Phase Recognition Method Robust to Directional Variations in Subject's Gait Progression Using Wearable Inertial Sensor

Inertial Measurement Unit (IMU) sensor-based gait phase recognition is widely used in medical and biomechanics fields requiring gait data analysis. However, there are several limitations due to the low reproducibility of IMU sensor attachment and the sensor outputs relative to a fixed reference frame. The prediction algorithm may malfunction when the user changes their walking direction. In this paper, we propose a gait phase recognition method robust to user body movements based on a floating body-fixed frame (FBF) and bi-directional long short-term memory (bi-LSTM). Data from four IMU sensors attached to the shanks and feet on both legs of three subjects, collected via the FBF method, are processed through preprocessing and the sliding window label overlapping method before inputting into the bi-LSTM for training. To improve the model's recognition accuracy, we selected parameters that influence both training and test accuracy. We conducted a sensitivity analysis using a level average analysis of the Taguchi method to identify the optimal combination of parameters. The model, trained with optimal parameters, was validated on a new subject, achieving a high test accuracy of 86.43%.

Comparative Study on Distributed Lightweight Deep Learning Models for Road Pothole Detection

This paper delves into image detection based on distributed deep-learning techniques for intelligent traffic systems or self-driving cars. The accuracy and precision of neural networks deployed on edge devices (e.g., CCTV (closed-circuit television) for road surveillance) with small datasets may be compromised, leading to the misjudgment of targets. To address this challenge, TensorFlow and PyTorch were used to initialize various distributed model parallel and data parallel techniques. Despite the success of these techniques, communication constraints were observed along with certain speed issues. As a result, a hybrid pipeline was proposed, combining both dataset and model distribution through an all-reduced algorithm and NVlinks to prevent miscommunication among gradients. The proposed approach was tested on both an edge cluster and Google cluster environment, demonstrating superior performance compared to other test settings, with the quality of the bounding box detection system meeting expectations with increased reliability. Performance metrics, including total training time, images/second, cross-entropy loss, and total loss against the number of the epoch, were evaluated, revealing a robust competition between TensorFlow and PyTorch. The PyTorch environment's hybrid pipeline outperformed other test settings.

Topographic divergence of atypical cortical asymmetry and atrophy patterns in temporal lobe epilepsy

Temporal lobe epilepsy, a common drug-resistant epilepsy in adults, is primarily a limbic network disorder associated with predominant unilateral hippocampal pathology. Structural MRI has provided an in vivo window into whole-brain grey matter structural alterations in temporal lobe epilepsy relative to controls, by either mapping (i) atypical inter-hemispheric asymmetry; or (ii) regional atrophy. However, similarities and differences of both atypical asymmetry and regional atrophy measures have not been systematically investigated. Here, we addressed this gap using the multisite ENIGMA-Epilepsy dataset comprising MRI brain morphological measures in 732 temporal lobe epilepsy patients and 1418 healthy controls. We compared spatial distributions of grey matter asymmetry and atrophy in temporal lobe epilepsy, contextualized their topographies relative to spatial gradients in cortical microstructure and functional connectivity calculated using 207 healthy controls obtained from Human Connectome Project and an independent dataset containing 23 temporal lobe epilepsy patients and 53 healthy controls and examined clinical associations using machine learning. We identified a marked divergence in the spatial distribution of atypical inter-hemispheric asymmetry and regional atrophy mapping. The former revealed a temporo-limbic disease signature while the latter showed diffuse and bilateral patterns. Our findings were robust across individual sites and patients. Cortical atrophy was significantly correlated with disease duration and age at seizure onset, while degrees of asymmetry did not show a significant relationship to these clinical variables. Our findings highlight that the mapping of atypical inter-hemispheric asymmetry and regional atrophy tap into two complementary aspects of temporal lobe epilepsy-related pathology, with the former revealing primary substrates in ipsilateral limbic circuits and the latter capturing bilateral disease effects. These findings refine our notion of the neuropathology of temporal lobe epilepsy and may inform future discovery and validation of complementary MRI biomarkers in temporal lobe epilepsy.

Coherence in cooperative photon emission from indistinguishable quantum emitters

Photon-mediated interactions between atoms can arise via coupling to a common electromagnetic mode or by quantum interference. Here, we probe the role of coherence in cooperative emission arising from two distant but indistinguishable solid-state emitters because of path erasure. The primary signature of cooperative emission, the emergence of "bunching" at zero delay in an intensity correlation experiment, is used to characterize the indistinguishability of the emitters, their dephasing, and the degree of correlation in the joint system that can be coherently controlled. In a stark departure from a pair of uncorrelated emitters, in Hong-Ou-Mandel-type interference measurements, we observe photon statistics from a pair of indistinguishable emitters resembling that of a weak coherent state from an attenuated laser. Our experiments establish techniques to control and characterize cooperative behavior between matter qubits using the full quantum optics toolbox, a key step toward realizing large-scale quantum photonic networks.

Meta-control of social learning strategies

Social learning, copying other’s behavior without actual experience, offers a cost-effective means of knowledge acquisition. However, it raises the fundamental question of which individuals have reliable information: successful individuals versus the majority. The former and the latter are known respectively as success-based and conformist social learning strategies. We show here that while the success-based strategy fully exploits the benign environment of low uncertainly, it fails in uncertain environments. On the other hand, the conformist strategy can effectively mitigate this adverse effect. Based on these findings, we hypothesized that meta-control of individual and social learning strategies provides effective and sample-efficient learning in volatile and uncertain environments. Simulations on a set of environments with various levels of volatility and uncertainty confirmed our hypothesis. The results imply that meta-control of social learning affords agents the leverage to resolve environmental uncertainty with minimal exploration cost, by exploiting others’ learning as an external knowledge base.

Which individuals have reliable information: successful individuals or the majority? Seeking a suitable compromise between individual and social learning is crucial for optimum learning in a population. Motivated by the recent findings in neuroscience showing that the brain can arbitrate between different learning strategies, termed meta-control, we propose a meta-control approach in social learning context. First, we show that environmental uncertainty is a crucial predictor of the performance of the individual and two social learning strategies: success-based and conformist. Our meta social learning model uses environmental uncertainty to find a compromise between these two strategies. In simulations on a set of environments with various levels of volatility and uncertainty, we demonstrate that our model outperforms other meta-social learning approaches. In the subsequent evolutionary analysis, we show that our model dominated others in survival rate. Our work provides a new account of the trade-offs between individual, success-based, and conformist learning strategies in multi-agent settings. Critically, our work unveils an optimal social learning strategy to resolve environmental uncertainty with minimal exploration cost.

OUP accepted manuscript

Mouse models have been engineered to reveal the biological mechanisms of human diseases based on an assumption. The assumption is that orthologous genes underlie conserved phenotypes across species. However, genetically modified mouse orthologs of human genes do not often recapitulate human disease phenotypes which might be due to the molecular evolution of phenotypic differences across species from the time of the last common ancestor. Here, we systematically investigated the evolutionary divergence of regulatory relationships between transcription factors (TFs) and target genes in functional modules, and found that the rewiring of gene regulatory networks (GRNs) contributes to the phenotypic discrepancies that occur between humans and mice. We confirmed that the rewired regulatory networks of orthologous genes contain a higher proportion of species-specific regulatory elements. Additionally, we verified that the divergence of target gene expression levels, which was triggered by network rewiring, could lead to phenotypic differences. Taken together, a careful consideration of evolutionary divergence in regulatory networks could be a novel strategy to understand the failure or success of mouse models to mimic human diseases. To help interpret mouse phenotypes in human disease studies, we provide quantitative comparisons of gene expression profiles on our website (http://sbi.postech.ac.kr/w/RN).

Functional encryption for set intersection in the multi-client setting

Functional encryption for set intersection (FE-SI) in the multi-client environment is that each client i encrypts a set X i associated with time T by using its own encryption key and uploads it to a cloud server, and then the cloud server which receives a function key of the client indexes i, j from a trusted center can compute the intersection X i ∩ X j of the two client ciphertexts. In this paper, we first newly define the concept of FE-SI suitable for the multi-client setting. Then, we propose an efficient FE-SI scheme in asymmetric bilinear groups and prove the static security of our scheme under newly introduced assumptions. In our FE-SI scheme, a ciphertext consists of O ( ℓ ) group elements, a function key consists of a single group element, and the decryption algorithm has O ( ℓ 2 ) complexity where ℓ is the size of a set in the ciphertext. Next, we propose another FE-SI scheme with time-constrained keys that limits the ability of function keys to be valid only for a specified time period T, and proves the static security of our scheme. Finally, we prove that the two assumptions hold in the general group model to provide confidence in the two newly introduced assumptions.

Evaluation of 360° Image Projection Formats; Comparing Format Conversion Distortion Using Objective Quality Metrics

Currently available 360° cameras normally capture several images covering a scene in all directions around a shooting point. The captured images are spherical in nature and are mapped to a two-dimensional plane using various projection methods. Many projection formats have been proposed for 360° videos. However, standards for a quality assessment of 360° images are limited. In this paper, various projection formats are compared to explore the problem of distortion caused by a mapping operation, which has been a considerable challenge in recent approaches. The performances of various projection formats, including equi-rectangular, equal-area, cylindrical, cube-map, and their modified versions, are evaluated based on the conversion causing the least amount of distortion when the format is changed. The evaluation is conducted using sample images selected based on several attributes that determine the perceptual image quality. The evaluation results based on the objective quality metrics have proved that the hybrid equi-angular cube-map format is the most appropriate solution as a common format in 360° image services for where format conversions are frequently demanded. This study presents findings ranking these formats that are useful for identifying the best image format for a future standard.

AUBER: Automated BERT regularization

How can we effectively regularize BERT? Although BERT proves its effectiveness in various NLP tasks, it often overfits when there are only a small number of training instances. A promising direction to regularize BERT is based on pruning its attention heads with a proxy score for head importance. However, these methods are usually suboptimal since they resort to arbitrarily determined numbers of attention heads to be pruned and do not directly aim for the performance enhancement. In order to overcome such a limitation, we propose AUBER, an automated BERT regularization method, that leverages reinforcement learning to automatically prune the proper attention heads from BERT. We also minimize the model complexity and the action search space by proposing a low-dimensional state representation and dually-greedy approach for training. Experimental results show that AUBER outperforms existing pruning methods by achieving up to 9.58% better performance. In addition, the ablation study demonstrates the effectiveness of design choices for AUBER.

5G SLAM Using the Clustering and Assignment Approach with Diffuse Multipath

5G communication systems operating above 24 GHz have promising properties for user localization and environment mapping. Existing studies have either relied on simplified abstract models of the signal propagation and the measurements, or are based on direct positioning approaches, which directly map the received waveform to a position. In this study, we consider an intermediate approach, which consists of four phases-downlink data transmission, multi-dimensional channel estimation, channel parameter clustering, and simultaneous localization and mapping (SLAM) based on a novel likelihood function. This approach can decompose the problem into simpler steps, thus leading to lower complexity. At the same time, by considering an end-to-end processing chain, we are accounting for a wide variety of practical impairments. Simulation results demonstrate the efficacy of the proposed approach.

Effect of self-monitoring on long-term patient engagement with mobile health applications

Despite the growing adoption of the mobile health (mHealth) applications (apps), few studies address concerns with low retention rates. This study aimed to investigate how the usage patterns of mHealth app functions affect user retention. We collected individual usage logs for 1,439 users of single tethered personal health record app, which spanned an 18-months period from August 2011 to January 2013. The user logs contained timestamps whenever an individual uses each function, which enables us to identify the usage patterns based on the intensity of using a particular function in the app. We then estimated how these patterns were related to 1) the app usage over time (using the random effect model) and 2) the probability of stopping the use of the application (using the Cox proportional hazard model). The analyses suggested that the users utilize the app most at the time of the adoption and gradually reduce their usage over time. The average duration of use after starting the app was 25.62 weeks (SD: 18.41). The degree of the usage reduction, however, decreases as the self-monitoring function is more frequently used (coefficient = 0.002, P = 0.013); none of the other functions has this effect. Moreover, engaging with the self-monitoring function frequently (coefficient = −0.18, P = 0.003) and regularly (coefficient = 0.10, P = 0.001) significantly also reduces the probability of abandoning the application. Specifically, the estimated survival rate indicates that, after 40 weeks since the adoption, the probability of the regular users of self-monitoring to stay in use was about 80% while that of non-user was about 60%. This study provides the empirical evidence that sustained use of mHealth app is closely linked to the regular usage on self-monitoring function. The implications can be extended to the education of users and physicians to produce better outcomes as well as application development for effective user interfaces.