KMITL

Permanent URI for this communityhttps://dspace.kmitl.ac.th/handle/123456789/1

Browse

Search Results

Now showing 1 - 10 of 15

Some of the metrics are blocked by your
consent settings
EEG-BBNet: A Hybrid Framework for Brain Biometric Using Graph Connectivity
(2025-01-01)
Lakhan, Payongkit
;
Banluesombatkul, Nannapas
;
Sricom, Natchaya
;
Sawangjai, Phattarapong
;
Sangnark, Soravitt
Most EEG-based biometrics rely on either convolutional neural networks (CNNs) or graph convolutional neural networks (GCNNs) for personal authentication, potentially overlooking the limitations of each approach. To address this, we propose EEG-BBNet, a hybrid network that combines CNNs and GCNNs. EEG-BBNet leverages CNN’s capability for automatic feature extraction and the GCNN’s ability to learn connectivity patterns between EEG electrodes through graph representation. We evaluate its performance against solely CNN-based and graph-based models across three brain–computer interface tasks, focusing on daily motor and sensory activities. The results show that while EEG-BBNet with Rho index functional connectivity metric outperforms graph-based models, it initially lags behind CNN-based models. However, with additional fine-tuning, EEG-BBNet surpasses CNN-based models, achieving a correct recognition rate of approximately 90%. This improvement enables EEG-BBNet to adapt its learning in new sessions and to acquire different domain knowledge across various BCI tasks (e.g., motor imagery to steady-state visually evoked potentials), demonstrating promise for practical authentication.
Some of the metrics are blocked by your
consent settings
BAMS: Binary Sequence-Augmented Spectrogram with Self-Attention Deep Learning for Human Activity Recognition
(2024-01-01)
Sricom, Natchaya
;
Charakorn, Rujikorn
;
Manoonpong, Poramate
;
Limpiti, Tulaya
Human Activity Recognition (HAR) has rapidly gained interest over the years due to its wide range of applications in AI-based systems, particularly healthcare monitoring. HAR methods typically involve extracting relevant features from data provided by wearable sensors, smartphone sensors, cameras, or their combinations to classify different activities. Nevertheless, a major challenge lies in achieving high classification accuracy with limited data samples, particularly when distinguishing between activities with similar signal attributes. To address this challenge, we propose a novel HAR method called BinAry sequence-augmented spectrograM with Self-attention deep learning (BAMS). Our proposed method leverages only basic wearable sensor data. It utilizes short-time Fourier transform spectrograms to extract spatio-temporal sensor information. The spectrogram is integrated with a binary sequence that captures movement direction. We integrate a scaled dot-product self-attention mechanism into the model to prioritize data from wearable sensors, thereby enhancing the model's performance. The proposed method is evaluated on a public dataset using leave-one-subject-out cross-validation for efficacy and robustness. The method is found to achieve significant improvement over other state-of-the-art methods with the classification accuracy percentage and weighted F-1 scores of 88.06±5.11 and 87.36±5.96, respectively, for a twelve-activity classification.
Some of the metrics are blocked by your
consent settings
Non-invasive Optical Blood Glucose Measuring System using Regression Models
(2023-01-01)
Limpiti, Tulaya
;
Seemavijai, Mayravee
;
Pongmee, Narutchai
;
Kanayat, Pannita
;
Koseeyaporn, Jeerasuda
Diabetes is one of the chronic diseases with an increasing number of patients every year. Therefore, blood glucose level (BGL) is an essential health information for diagnosis and management of diabetes. Commonly, the BGL is usually measured invasively by taking blood samples, which can cause pain or discomfort to the patients. Our paper aims to design and develop a low-cost, low-computational non-invasive blood glucose measuring system using optical technique. The proposed system consists of a hardware for measuring photoplethysmogram (PPG). The PPG signal is subsequently analyzed to estimate the corresponding BGL, which is sent wirelessly to be stored in a database and displayed on an Android application. The application recommends the appropriate dosage of insulin injection based on the type of users and their diets. The report of historical blood glucose levels in the database can also be generated for further medical examination.Three PPG features, namely peak-to-peak amplitude, maximum amplitude, and standard deviation of signal amplitude, have the strongest correlations to the BGL. We investigate three regression models for estimating the BGL from these PPG parameters. The robustness of the system is assessed using cross validations. The average root mean squares error (RMSE) for three-fold cross-validations are 17.38, 16.76, and 16.39 mg/dL for simple linear regression, principal component regression, and partial least squares regression, respectively. The RMSEs from leave-one-out cross validation are approximately 12 mg/dL, with partial least squares regression model having the best accuracy. Furthermore, the results from Clarke Error Grid Analysis indicate that the system can be implemented with any of the three models for practical usage.
Some of the metrics are blocked by your
consent settings
Cardiac Arrhythmia Teletriage using Electrocardiogram
(2021-01-01)
Limpiti, Tulaya
;
Chokchaichumnankit, Thunchanok
;
Sanguanchom, Jirayu
;
Soyphan, Natchanon
;
Puttarak, Nattakan
Telehealth has become a favorable method for receiving medical care during the COVID-19 pandemic. It reduces physical contact and also benefits those who live a distance from hospitals. In this paper we present a cardiac arrhythmia teletriage using electrocardiogram (ECG). The system consists of a diagnostic algorithm for arrhythmias and an Android application. The diagnostic algorithm can detect five types of cardiac problems-arrhythmia, bradycardia, tachycardia, bradyarrhythmia, and tachyarrhythmia. The Android application is the main communication channel between patients and healthcare providers. The user uploads their ECG and receives a preliminary diagnosis of their heart health via the application. The system notifies the user if it detects any abnormalities. The user can then make an appointment online for further examination at the hospital. The capability of the proposed system is evaluated using four databases from PhysioNet - MITBIH Normal Sinus Rhythm Database, MIT-BIH Arrhythmia Database, MIT-BIH Atrial Fibrillation Database, and CU Ventricular Tachyarrhythmia Database. It is found that the algorithm is able to detect abnormal ECG signals with an average accuracy of 82.1%.
Some of the metrics are blocked by your
consent settings
Transfer Learning for Classifying Motor Imagery EEG: A Comparative Study
(2021-01-01)
Limpiti, Tulaya
;
Seetanathum, Kornthum
;
Sricom, Natchaya
;
Puttarak, Nattakan
Motor imagery (MI) is the cognitive process when a person imagines performing a specific movement of their body. The corresponding electroencephalographic (EEG) signals can be measured externally to the head using scalp electrodes. Such signals have been applied in healthcare brain-computer-interface (BCI) systems, for example, motor rehabilitation and prosthetics control. These systems convert different MI EEG input signals to directives, so their performances depend on the efficiency of the embedded signal classification algorithm. In this paper we investigate the effectiveness of transfer learning in classifying the MI EEG data. The Continuous Wavelet Transform (CWT) is used to construct the scalograms, which serve as the inputs to the deep learning structure. The efficacies of five pre-trained networksAlexNet, ResNet18, ResNet50, InceptionV3 and ShuffleNet-are evaluated on the BCI competition IV data set 2a. Binary (left hand vs. right hand) and four-class (left hand, right hand, both feet, and tongue) classifications are trained and tested using fivefold cross validation. The result indicates that using the CWT with transfer learning models provides very high classification accuracies. The ResNet18 network achieves the best accuracies in both cases at 95.03±2.95% and 91.86±2.90%, respectively. In addition, we examine the effect of different time-frequency features on the classification performance by comparing the scalogram of the CWT and the spectogram of the Short-Time Fourier Transform (STFT) as the inputs. It is found that the CWT is the preferred choice as it is superior to the STFT.
Some of the metrics are blocked by your
consent settings
A smart dialogue box for the hearing impaired
(2019-07-01)
Limpiti, Tulaya
;
Khamyan, Sudarat
;
Chumsaeng, Setthawut
;
Puttarak, Nattakan
Conversations between a person with severe hearing impairment and a hearing person is difficult because people who know sign language are rare. Therefore, to facilitate the conversation, we design and develop a portable dialogue box that can be used without an interpreter. The system detects sign language of a deaf person by capturing hand images with a camera. The images are pre-processed and the histograms of oriented gradients (HOG) are extracted as image features. A multi-class support vector machine uses the HOGs as inputs to classify the alphabets. The vocal reply from a hearing person is received through a microphone and interpreted via Google cloud speech API. Both parts are processed by a microcomputer to produce messages of the conversation. The dialogue is displayed on the graphical user interface. The system can also record the conversation logs into the database.
Some of the metrics are blocked by your
consent settings
PCA-based informative SNP selection for analyzing population structure
(2016-12-19)
Limpiti, Tulaya
;
Intarapanich, Apichart
;
Tongsima, Sissades
Phenotypic differences among individuals of the same species are the result of a set of genetic variations which can be observed in the DNA sequence. To conduct a population genetic study, a high throughput genotyping platform such as Single Nucleotide Polymorphism (SNP) array is popularly used to obtain a large set of SNPs for each individual. However, analyzing today's genotypic data can be computationally expensive due to its large size and complexity. Faulty substructure may also be detected if the data is noisy from redundant or non-informative SNPs. Considerable efforts have been done to extract a smaller informative SNP subset that still represents the same intrinsic structure of populations within a data set as the full panel of SNPs. This work describes a foundation of a PCA-based informative marker selection technique. The proposed technique is simple and efficient. It improves upon another spectral analysis technique called PCA-correlated SNPs. A new informativeness score based on a basis function expansion of the SNP variation patterns across individuals is introduced. Such score is computed for each SNP to select a subset of SNPs with the best scores. Using a bovine data set, we demonstrate that our technique is superior to the PCAcorrelated SNPs method, which requires accurate rank estimation to perform well. In contrast, our method is robust to the assumed rank of the data. High data representation accuracy is also achieved after a significant reduction of the number of SNPs while retaining information about the underlying population structure from the original data.
Some of the metrics are blocked by your
consent settings
Automated classification of malaria parasite species on thick blood film using support vector machine
(2016-02-04)
Pinkaew, Aeggarut
;
Limpiti, Tulaya
;
Trirat, Akraphon
Malaria is a serious global health problem. It requires fast and effective diagnosis for detecting and classifying the type of infection. Proper treatment should be administered in a timely fashion to prevent an outbreak. Microscopic examination of thick blood films is one of the current standards for malaria diagnosis. However, inspecting a thick blood film is time-consuming and requires experienced technicians. Hence, for developing countries where most cases of malaria occur but microscopy expertise may not be available, a computerized system to aid such diagnosis is desirable. In this paper, an automated classification system operating on digitized images of thick blood film has been developed to classify between Plasmodium falciparum and Plasmodium vivax malaria parasite species. The system is fully automated. It is fast and can be handled by non-experts. We calculate five statistical features - mean, standard deviation, kurtosis, skewness and entropy - from four color channels (green, intensity, saturation, and value) of these images. The features are then projected onto a subspace representing image characteristics from both species. The projected features are used by the support vector machine for classification. It is found that the algorithm has acceptable training error and can classify test images with good accuracy.
Some of the metrics are blocked by your
consent settings
A study on Ka-band site switching using advanced prediction for rain-induced attenuation
(2015-01-28)
Poolsawut, Jirutchaya
;
Limpiti, Tulaya
;
Puttarak, Nattakan
High frequency satellite communications suffers from rain-induced attenuations, which may cause service discontinuity during high-intensity rain events. To combat this, we propose a novel site switching technique that anticipates signal attenuations using a linear dynamical system response model and performs site switching in advance. Using only the rainfall data measured at a ground station for attenuation prediction, we demonstrate that our method is able to lessen signal attenuation while maintaining link availability.
Some of the metrics are blocked by your
consent settings
INJclust: Iterative neighbor-joining tree clustering framework for inferring population structure
(2014-09-01)
Limpiti, Tulaya
;
Amornbunchornvej, Chainarong
;
Intarapanich, Apichart
;
Assawamakin, Anunchai
;
Tongsima, Sissades
Understanding genetic differences among populations is one of the most important issues in population genetics. Genetic variations, e.g., single nucleotide polymorphisms, are used to characterize commonality and difference of individuals from various populations. This paper presents an efficient graph-based clustering framework which operates iteratively on the Neighbor-Joining (NJ) tree called the iNJclust algorithm. The framework uses well-known genetic measurements, namely the allele-sharing distance, the neighbor-joining tree, and the fixation index. The behavior of the fixation index is utilized in the algorithm's stopping criterion. The algorithm provides an estimated number of populations, individual assignments, and relationships between populations as outputs. The clustering result is reported in the form of a binary tree, whose terminal nodes represent the final inferred populations and the tree structure preserves the genetic relationships among them. The clustering performance and the robustness of the proposed algorithm are tested extensively using simulated and real data sets from bovine, sheep, and human populations. The result indicates that the number of populations within each data set is reasonably estimated, the individual assignment is robust, and the structure of the inferred population tree corresponds to the intrinsic relationships among populations within the data.

KMITL

Browse

Filters

Settings

Sort By

Results per page

Search Results