Session Index

S6. Biophotonics and Biomedical Imaging

Biophotonics and Biomedical Imaging III
Saturday, Dec. 2, 2023  10:45-12:00
Presider: Prof. Fu-Jen Kao (National Yang Ming Chiao Tung University, Taiwan) Prof. Chia-Yuan Chang (National Cheng Kung University, Taiwan)
Room: 92271 (2F)
10:45 - 11:00
Manuscript ID.  1019
Paper No.  2023-SAT-S0603-O001
Krishna K Mahato Patient-derived-breast tumor xenograft model development and detection by photoacoustic spectroscopy enabled machine learning: ex vivo
Jackson Rodrigues, Darshan C Mukunda, Subhash Chandra, Manipal School of Life Sciences, Manipal Academy of Higher Education (India); Bhavya K P, Stanley Mathew, Kasturba Medical College, Manipal Academy of Higher Education (India); Krishna K Mahato, Manipal School of Life Sciences, Manipal Academy of Higher Education (India)

The current study aimed to establish a patient-derived xenografts (PDXs) model in athymic nude mice. The breast tumor progression was monitored at two progressive tumor volumes of 300 mm3 & 600 mm3, respectively. Ex vivo tumor specimens were also collected, reflecting the diversity of breast cancer types, tissue changes, and tumor behavior. These samples were subjected to photoacoustic spectral measurements compared to control samples. The photoacoustic spectral data analysis involving preprocessing, wavelet transform, feature selection, and classification using support vector machine learning demonstrated an overall accuracy of 98.3%, revealing the tool's potentiality in breast cancer detection.

  Preview abstract
11:00 - 11:15
Manuscript ID.  0003
Paper No.  2023-SAT-S0603-O002
Po-Sheng Lee Using Transformer with mm-Wave FMCW Radar for Skeleton Joint Point Prediction from Sparse Point Cloud
Po-Sheng Lee, National Yang Ming Chiao Tung University (Taiwan)

We propose using a Transformer model to predict 17 joint locations from sparse point clouds, enabling more comprehensive body information extraction. Our current results achieve an mae of 1.678cm.

  Preview abstract
11:15 - 11:30
Manuscript ID.  1069
Paper No.  2023-SAT-S0603-O003
Chi-Wen Chen Video-rate three-photon imaging in deep drosophila brain based on a single Cr:forsterite laser oscillator
Chi-Wen Chen, Je-Chi Jang, Shao-Hsuan Wu, Lu-Ting Chou, National Yang Ming Chiao Tung University (Taiwan); Hen Chang, Department of Biomedical Engineering and Environmental Sciences (Taiwan); Ting-Chen Chang, Chung-Ming Chen, Department of Physics (Taiwan); Li-An Chu, Department of Biomedical Engineering and Environmental Sciences (Taiwan); Shi-Wei Chu, Department of Physics (Taiwan); Shih-Hsuan Chia, National Yang Ming Chiao Tung University (Taiwan)

We demonstrated three-photon fluorescence microscopy for drosophila brain imaging based on a 24-MHz Cr:forsterite oscillator. We studied the soliton mode-locking dynamics to optimize the output pulse width and peak power by managing the intracavity dispersion. We have realized three-photon fluorescence imaging, and the imaging contrast and penetration depth are much better than the results obtained from the two-photon excitation. Moreover, we found that the signal-to-background ratio at depth is greatly improved when using shorter pulses from the laser oscillator. For functional imaging applications, we also demonstrated three-photon calcium imaging stimulated by an external electric shock.

  Preview abstract
11:30 - 11:45
Manuscript ID.  0053
Paper No.  2023-SAT-S0603-O004
Chia-Chen Li Using functional near-infrared spectroscopy in young migraine detection
Chia-Chen Li, National Yang Ming Chiao Tung University (Taiwan); Wei-Ta Chen, Keelung Hospital (Taiwan), Taipei Veterans General Hospital (Taiwan); Yao-Hong Liu, Chia-Wei Sun, National Yang Ming Chiao Tung University (Taiwan)

This study combines functional near-infrared spectroscopy (fNIRS) with machine learning to find the difference between healthy individuals and migraine patients in young populations during a concentration task (CT). Statistical methods are applied to find the features which have significant differences. In this research, we employ the support vector machine (SVM) for young migraine detection, achieving accuracies of 96.4% and 83.3% on training and testing data, respectively.

  Preview abstract
11:45 - 12:00
Manuscript ID.  0328
Paper No.  2023-SAT-S0603-O005
Takhellambam Gautam Meitei Evaluating Tourette Syndrome Severity using Neurofeedback
Takhellambam Gautam Meitei, Yu-Jiun Chen, National Yang Ming Chiao Tung University (Taiwan); Pou-Leng Cheong, National Yang Ming Chiao Tung University (Taiwan), National Taiwan University Hospital (Taiwan); Chia-Wei Sun, National Yang Ming Chiao Tung University (Taiwan)

This study explores the increasing prevalence of Tourette syndrome (TS) and its impact on patient’s quality of life. TS symptoms, known as tics, can be motor or vocal. While drugs are commonly used for treatment, they often entail significant side effects. Therefore, behavior therapy, particularly neurofeedback training using functional near-infrared spectroscopy, is considered a preferable approach. The research demonstrates the effectiveness of neurofeedback training for TS patients, with an 83.3% training and 75.0% testing accuracy in distinguishing high and low tic severity groups. The study confirms the model's generalization ability, offering the potential for advanced classification using machine learning.

  Preview abstract