Taweechai Ouypornkochagorn

Dr. Taweechai Ouypornkochagorn


Research Focus & Metrics

Academic Career Length

23Years (2004-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
6 February 2020 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
Shengheng LiuShengheng Liu
Shengheng Liu
RCRuisong Cao
Ruisong Cao
YHYongming Huang
Yongming Huang
TOTaweechai Ouypornkochagorn
Taweechai Ouypornkochagorn
JJJiabin Jia
Jiabin Jia
As an emerging technology for continuous monitoring of a bounded domain, electrical impedance tomography (EIT) gains increasing popularity in various applications. Despite unprecedented progress, the EIT inverse solvers at the present stage are incompetent to guarantee sufficient fidelity as well as efficient investigation of the internal impedance dynamics. In this context, this paper introduces a spatiotemporal structure-aware sparse Bayesian learning (SA-SBL) framework for solving the time-continuous EIT inverse problems. Specifically, in the process of reconstructing EIT time sequence, both intra-frame spatial clustering and inter-frame temporal continuity are explored and exploited in an unsupervised manner by using hierarchical Bayesian model and structure-aware priors. A multiple measurement vector model is established to capture the spatiotemporal correlations and describe the underlying multi-dimensional reconstruction problem. The resultant large-scale inversion is efficiently solved by applying the approximate message passing to the expectation updating. A speedup ratio of O(N2M) is achieved compared with original SA-SBL. Simulation results indicate that, the proposed algorithm exhibits superior reconstruction performance to the existing methods, where the scores evaluated by the quantitative metrics are improved by at least 17%. The presented algorithm is envisioned to offer broader applicability, since it yields improved image quality and recovery efficiency.

21 January 2022 in IEEE Sensors Journal
Verified
Journal Article
Open access
TOTaweechai Ouypornkochagorn
Taweechai Ouypornkochagorn
NTNatasa Terzija
Natasa Terzija
PWPaul Wright
Paul Wright
JDJohn L. Davidson
John L. Davidson
NPNick Polydorides
Nick Polydorides
HMHugh McCann
Hugh McCann
An Electrical Impedance Tomography (EIT) system has been developed for dynamic three-dimensional imaging of changes in conductivity distribution in the human head, using scalp-mounted electrodes. We attribute these images to changes in cerebral perfusion. At 100 frames per second (fps), voltage measurement is achieved with full-scale signal-to-noise ratio of 105 dB and common-mode rejection ratio > 90 dB. A novel nonlinear method is presented for 3-D imaging of the difference in conductivity distribution in the head, relative to a reference time. The method achieves much reduced modelling error. It successfully localizes conductivity inclusions in experimental and simulation tests, where previous methods fail. For > 50 human volunteers, the rheoencephalography (REG) waveform is observed in EIT voltage measurements for every volunteer, with peak-to-peak amplitudes up to approx. $50 ~\mu \text{V}_{{\mathrm {rms}}}$ . Images are presented of the change in conductivity distribution during the REG/cardiac cycle, at 50 fps, showing maximum local conductivity change of approx. 1% in grey/white matter. A total of 17 tests were performed during short (typically 5s) carotid artery occlusions on 5 volunteers, monitored by Transcranial Doppler ultrasound. From EIT measurements averaged over complete REG/cardiac cycles, 13 occlusion tests showed consistently decreased conductivity of cerebral regions on the occluded side, and increased conductivity on the opposite side. The maximum local conductivity change during occlusion was approx. 20%. The simplicity of the carotid artery intervention provides a striking validation of the scalp-mounted measurement system in imaging cerebral hemodynamics, and the REG images indicate its unique combination of sensitivity and temporal resolution.

1 July 2018
Verified
Conference Paper
Closed access
PLPhattaratorn Lasomsri
Phattaratorn Lasomsri
PYPhattharaporn Yanbuaban
Phattharaporn Yanbuaban
OKOrawipa Kerdpoca
Orawipa Kerdpoca
TOTaweechai Ouypornkochagorn
Taweechai Ouypornkochagorn
Poor Indoor Air Quality (IAQ) is an issue of environmental health problems that cause headaches, fatigue, irritation of the eyes and skin. IAQ in hospital areas is a serious concern since there are many occupants i.e. patients, medical practitioners, and staffs. Furthermore, medical material, medicine, medical gases can cause the quality of air deteriorate. IAQ then requires to regularly check the service areas in order to maintain a good level of air quality. The instrument to measure IAQ is usually expensive. In this study, a low-cost device to measure IAQ was developed, it comprises of two main sensors: BME680 and CCS811. The device was capable of measuring and recording continuously which allows it to monitor air quality including Total Volatile Organic Compounds (TVOC), temperature and humidity. The device was employed to measure three service areas of a large-scale hospital in Nakhonnayok, Thailand: the waiting area, the emergency room, and the outpatient pharmacy. IAQs of these areas obtained from the developed device were all in the very bad level. It is possibly due to the influence of structural renovation, medical material, air conditioning, air ventilation, and the people occupied in the areas.

13 October 2023 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
TOTaweechai Ouypornkochagorn
Taweechai Ouypornkochagorn
NNNapatsawan Ngamdi
Napatsawan Ngamdi
In this paper, we present a high-precision multi-frequency Electrical Impedance Tomography system for wide areas of medical applications. The system has 32 channels with full parallel data acquisition. Instead of using superposition excitation for producing excitation current, we propose to use package excitation of five different frequencies between 2 kHz and 125 kHz. The excitation period is divided into five sub-periods, where only one frequency of excitation is performed at a sub-period. This excitation method can increase the readout SNR by 30 dB. The data of each frequency can be collected simultaneously. The amplitudes are programable with a maximum of 2.83 mArms. The recording speed is 20 frames per second. Eighteen-bit analog-to-digital converters are used to obtain a fine resolution with a wide measurement range. The maximum load of the current source is 2000 Ω. The SNR of the current source and the voltage measurement unit is 78.5 dB and 91.7 dB, respectively. However, the SNRs reduce to 67.9 dB and 81 dB, respectively, at the 2 kHz and 0.2 mArms of excitation. With high SNR and a wide measurement range, this system is then applicable for wide medical applications where high precision is required. The system validation was performed on a phantom tank, a human forearm, and a human chest with satisfactory results.

4 July 2023 in IEEE Access
Verified
Journal Article
Open access
TOTaweechai Ouypornkochagorn
Taweechai Ouypornkochagorn
NNNapatsawan Ngamdi
Napatsawan Ngamdi
SOSairoong Ouypornkochagorn
Sairoong Ouypornkochagorn
TTTherdkiat Trongwongsa
Therdkiat Trongwongsa
Pathological investigation for addressing abnormal or precancerous cells in cervix specimens is necessary to diagnose or treat. This process is exhaustive in histologic examination and requires experienced pathologists. Detecting the presence of precancerous cells with bioimpedance is another method to help pathologists by using a small-diameter probe to measure the impedance. However, it requires several measurements and is possibly another exhaustive task. This study proposed the localization of abnormal cells in a specimen using reconstruction images with Frequency-difference Electrical Impedance Tomography (fdEIT). Due to the various shape of specimens in practice, a new approach to cope with modeling error was also proposed to enable using a generic inversion model in the reconstruction process. Simulation of cervix specimens with different orifice shapes and thicknesses was performed. The proposed method was compared with the traditional methods i.e. the simple-fdEIT and the weighted-fdEIT. Amplitude response, radius, position error and ringing ratio of the reconstruction images were determined. The image reconstruction of the proposed method showed superior performance in localization. Although the image amplitude response was smaller, the artifacts that occurred due to modeling error were significantly reduced. Even though the shape of cervix specimens is varied, the proposed method allows to use a generic model for precancerous cells localization. The influence of modeling error can be reduced. Consequentially, the ringing artifact can be significantly reduced and the position error is also slightly reduced.

27 October 2023 in IEEE Sensors Journal
Verified
Journal Article
Closed access
TOTaweechai Ouypornkochagorn
Taweechai Ouypornkochagorn
WDWatcharee Dankul
Watcharee Dankul
LRLawan Ratanasathien
Lawan Ratanasathien
Fetal movement counting is a simple method for monitoring fetal health. The counting can be self-performed; however, pregnant women require full attention during the process. It becomes difficult to perform other routine activities while counting. Therefore, wearable counting devices that can detect fetal movement and distinguish it from non-fetal movement are needed. In this work, an acoustic device was proposed for detecting fetal movement. The device’s dish was large i.e. 31 mm, and therefore a single device was sufficient for detection. Power spectral density (PSD) was used for creating the spectrogram of the moving signal. Four classifiers were used to identify fetal and non-fetal movements. The experiment was conducted on 12 pregnant women at their homes or workplaces where body movement was allowed. The device was attached to the abdomen and recorded using a smartphone connected via Bluetooth. The total recorded time was 50.5 hours. The signal amplitude of the fetal movement was 8.5 dB higher than that of the non-fetal movement. The detectivity of the device was 67.2%-75.2%. Lower detectivity was observed in subjects with small and very large abdominal circumferences. Gestation showed a negative correlation with maternal perception but no correlation with the device’s detection rate The convolutional neural network classifier exhibited the highest performance, with a F1-score of 95.2% and an accuracy of 94.9%.