arXiv subjects

Theerasarn Pianpanit

Publications and source records attributed to Theerasarn Pianpanit.

4 recordsLinked to original sources

arXiv2025

Flow Crossover and Parallel Outflow during Collisionless Magnetic Reconnection

Using particle-in-cell simulations that label ions and electrons according to their initial inflow region, we find that during 2D collisionless magnetic reconnection, the bulk flow of the plasma from each inflow side crosses paths with plasma from the other inflow side and crosses the midplane before being redirected into an outflow jet. This feature, which we term ``flow crossover,'' implies mechanisms to generate bulk motion in a direction parallel to the magnetic field. We find that ions and electrons undergo different parallel driving mechanisms, leading to different flow crossover patterns. The parallel bulk flow for ions is generated more locally within the ion diffusion region, whereas the parallel bulk flow for electrons is mostly generated outside the electron diffusion region. Consequently, the reconnection outflows are more of a parallel flow than a perpendicular flow, especially for the electron outflow. The flow crossover and the parallel outflow patterns occur not only in symmetric reconnection but also in the more complex scenario of a guide-field asymmetric reconnection, suggesting that it is a general feature of collisionless magnetic reconnection. Because the plasma on one side of the outflow mostly originates from the inflow plasma on the other side, we predict that near an asymmetric reconnection site in a collisionless space plasma, in situ observations across the outflow region could reveal locally reversed gradients in plasma properties.

physics.plasm-ph↗

arXiv2021

Parkinson's Disease Recognition Using SPECT Image and Interpretable AI: A Tutorial

In the past few years, there are several researches on Parkinson's disease (PD) recognition using single-photon emission computed tomography (SPECT) images with deep learning (DL) approach. However, the DL model's complexity usually results in difficult model interpretation when used in clinical. Even though there are multiple interpretation methods available for the DL model, there is no evidence of which method is suitable for PD recognition application. This tutorial aims to demonstrate the procedure to choose a suitable interpretation method for the PD recognition model. We exhibit four DCNN architectures as an example and introduce six well-known interpretation methods. Finally, we propose an evaluation method to measure the interpretation performance and a method to use the interpreted feedback for assisting in model selection. The evaluation demonstrates that the guided backpropagation and SHAP interpretation methods are suitable for PD recognition methods in different aspects. Guided backpropagation has the best ability to show fine-grained importance, which is proven by the highest Dice coefficient and lowest mean square error. On the other hand, SHAP can generate a better quality heatmap at the uptake depletion location, which outperforms other methods in discriminating the difference between PD and NC subjects. Shortly, the introduced interpretation methods can contribute to not only the PD recognition application but also to sensor data processing in an AI Era (interpretable-AI) as feedback in constructing well-suited deep learning architectures for specific applications.

eess.IV↗

arXiv2020

SleepPoseNet: Multi-View Learning for Sleep Postural Transition Recognition Using UWB

Recognizing movements during sleep is crucial for the monitoring of patients with sleep disorders, and the utilization of ultra-wideband (UWB) radar for the classification of human sleep postures has not been explored widely. This study investigates the performance of an off-the-shelf single antenna UWB in a novel application of sleep postural transition (SPT) recognition. The proposed Multi-View Learning, entitled SleepPoseNet or SPN, with time series data augmentation aims to classify four standard SPTs. SPN exhibits an ability to capture both time and frequency features, including the movement and direction of sleeping positions. The data recorded from 38 volunteers displayed that SPN with a mean accuracy of $73.7 \pm 0.8 \%$ significantly outperformed the mean accuracy of $59.9 \pm 0.7 \%$ obtained from deep convolution neural network (DCNN) in recent state-of-the-art work on human activity recognition using UWB. Apart from UWB system, SPN with the data augmentation can ultimately be adopted to learn and classify time series data in various applications.

eess.SP↗

arXiv2020

Benchmark of a new multi-ion-species collision operator for $δf$ Monte Carlo neoclassical simulation

A numerical method to implement a linearized Coulomb collision operator in the two-weight $δf$ Monte Carlo method for multi-ion-species neoclassical transport simulation is developed. The conservation properties and the adjointness property of the operator in the collisions between two particle species with different temperatures are verified. The linearized operator in a $δf$ Monte Carlo code is benchmarked with other two kinetic simulations, a $δf$ continuum gyrokinetic code with the same linearized collision operator and a full-f PIC code with Nanbu collision operator. The benchmark simulations of the equilibration process of plasma flow and temperature fluctuation among several particle species show very good agreement between $δf$ Monte Carlo code and the other two codes. An error in the H-theorem in the two-weight $δf$ Monte Carlo method is found, which is caused by the weight spreading phenomenon inherent in the two-weight $δf$ method. It is demonstrated that the weight averaging method serves to restoring the H-theorem without causing side effect.

physics.plasm-ph↗