Session Index

Optical techniques have been increasingly applied to the screening and diagnosis of diseases, as well as non-invasive monitoring and surveillance of physiological states of tissue in vivo. For such sensing purposes, photons entering the tissue undergo multiple scattering events, and some of the photons not absorbed by the tissue return to the tissue surface and get detected. To accurately quantify pathological or physiological parameters from multi-dimensional data, the Monte Carlo (MC) method for solving the radiative transport problem has been developed and treated as the gold standard. However, a significant drawback of MC simulations is the high computational cost which hinders real-time processing. To address this issue, our group has adopted artificial neural networks to speed up MC simulations in several applications to quantify tissue optical properties in vivo. Examples include (1) absorption of melanin in the skin, (2) the strength of auto-fluorescence from the mucosa of the cervical uterine for detecting precancerous lesions, (3) scattering and absorption coefficients of major tissues in the human head to improve functional mapping of the brain activity and dosage quantification in transcranial brain stimulation, and (4) changes in oxygen saturation of venous blood.

  Preview abstract

Nature photonic crystals have been attracting researchers for a long time due to their bright reflected lights. Some of them are integrated into laser devices as scattering materials to form random lasing or mirrors to reduce the threshold. Here we discovered the band-edge lasing from photonic crystals in the scales of butterflies with new chromophores of donor-acceptor-donor motifs for the first time. This breakthrough may inspire future designs of photonic circuits or laser applications.

  Preview abstract

Neurite growth is essential for constructing the nervous system in the development process and the repair after trauma or disease. Numerous studies have shown that photobiomodulation by red or near-infrared light can stimulate neurite growth, but the detailed mechanisms are still unclear. In the present work we verified that 620 and 760 nm laser spots illuminating the soma of neuroblastoma cells (N2a) could cause neurite growth. This red light-induced neurite regrowth process is accompanied with an increase in intracellular reactive oxygen species (ROS).

  Preview abstract

This study used functional near-infrared spectroscopy (fNIRS) to measure the changes in hemoglobin concentration in the prefrontal cortex lobe of 33 police officers during the mental arithmetic and verbal fluency tasks. We process the signals to extract physiologically meaningful features and build a model using the Copenhagen overwork scale score as a classification criterion. Finally, the results show a training accuracy rate of 88.9 % and a testing accuracy rate of 80.0 %, indicating significant potential in assessing overwork by combining fNIRS with machine learning.

  Preview abstract

The polarimetric imaging system has been developed recently for many biological applications, such as tissue morphological characterization or cancer detection. This study applied the non-invasive nature of polarized light to evaluate at the polarization characteristics of two groups including healthy and malignant colorectal tissues. The experimental results showed that the normal tissues have more isotropic media features than malignancy. Moreover, the average intensity of almost all Mueller matrix components is much greater in malignant samples than in ones of normal samples. Overall, the proposed approach give a potential identification of colorectal malignant tissues on quantitative or semiquantitative criteria.

  Preview abstract