Brief look through the biomedical optical imaging: Acoustic Resolution Photoacoustic Microscopy (AR-PAM)
Shahriar Zeynali, Shahid Beheshti University (SBU), Tehran, Iran
Group meeting via video conference (Zoom)
Tuesday November 30, 9:00 (MEZ)
Due to the Covid-19 pandemic, we are still holding our group seminars and journal clubs via video conference.This procedure enables us to continue our research, enhance discussions and exchange important information.
The photoacoustic imaging (PAI) technique has been gaining a lot of attention in the past two decades due to its high contrast, scalable resolution, and relatively high imaging depth. PAI can be used for structural and functional imaging as well as for investigating tissue hemodynamics. There are multiple categories of PAI, one of which is acoustic resolution photoacoustic microscopy (AR-PAM). This type of imaging system uses focused transducers to capture generated photoacoustic signals in depth from a focused line. Multimode optical fibers (MMF) are extensively used to transfer light to AR-PAM imaging scan-head due to the feasibility and the capability of carrying high-energy light pulses. Typically, peak-power-compensation (PPC) is used to reduce the effect of pulse-to-pulse peak-power variation which exists in most of the light sources. In MMF, the output intensity profile fluctuates due to the mode exchange caused by variations in the bending of the fibers. Also, the coherent nature of light produces a speckle pattern at the output of the fiber. Therefore, using a photodiode to capture the total power of pulses as a measure of illuminated light in acoustic focus is not appropriate for PPC. In this study, we have investigated the accuracy of PPC in fiber-guided systems under different conditions including the degree of focusing light in the clear and scattering medium. Based on obtained results, in order to apply the PPC method to the pulse variation problem, tightly focused light must be used. PPC method has shown to be more accurate in a clear medium than in a highly scattering medium. Meanwhile, machine/deep learning methods have found many applications in biomedical systems such as photoacoustic microscopy. The implementation of deep learning in PAI has several applications such as solving the optical and acoustic inverse problems and segmentation of tumors. The majority of the efforts are directed towards enhancing the quality of images produced by the PAI systems. The training of a deep learning model (architecture) will require a dataset of images. Unlike commercially available imaging systems, such as magnetic resonance imaging (MRI) and computed tomography (CT), PAI systems suffer from lack of known and standard datasets that can be used for deep learning and/or machine learning applications. In this study, we have simulated (generated) vascular networks dataset of photoacoustic microscopy (PAM) images for use in deep learning models. Also, we have tested our dataset with a simple known deep learning model known as the U-Net architecture and deblurred images successfully. The generated dataset can be used in deep learning models for other biomedical vascular imaging systems including photoacoustic tomography (PAT), photoacoustic endoscopy (PAE), and so forth.
If you would like to join our group seminars via Zoom, please contact us for more information.