Deep Learning aplicado en imágenes fotoacústicas para la Identificación del cáncer de seno / Deep Learning Applied in Photoacoustic Images for the Identification of Breast Cancer

La Imagen Fotoacústica (PAI por sus siglas en inglés), es una modalidad de imagen híbrida que fusiona la iluminación óptica y la detección por ultrasonido. Debido a que los métodos de imágenes ópticas puras no pueden mantener una alta resolución, la capacidad de lograr imágenes de contraste óptico de alta resolución en tejidos biológicos hace que la fotoacústica (PA por sus siglas en inglés) sea una técnica prometedora para varias aplicaciones de imágenes clínicas. En la actualidad el Aprendizaje Profundo (Deep Learning) tiene el enfoque más reciente en métodos basados en la PAI, donde existe una gran cantidad de aplicaciones en análisis de imágenes, en especial en el área del campo biomédico, como lo es la adquisición, segmentación y reconstrucciones de imágenes de tomografía computarizada. Esta revisión describe las últimas investigaciones en PAI y un análisis sobre las técnicas y métodos basados en Deep Learning, aplicado en diferentes modalidades para el diagnóstico de cáncer de seno(AU)

Photoacoustic Imaging (PAI) is a hybrid imaging modality that combines optical illumination and ultrasound detection. Because pure optical imaging methods cannot maintain high resolution, the ability to achieve high resolution optical contrast images in biological tissues makes Photoacoustic (PA) a promising technique for various clinical imaging applications. At present, Deep Learning has the most recent approach of methods based on PAI where there are a large number of applications in image analysis especially in the area of ​​the biomedical field, such as acquisition, segmentation and reconstructions of computed tomography imaging. This review describes the latest research in PAI and an analysis of the techniques and methods based on Deep Learning applied in different modalities for the diagnosis of breast cancer(AU)

Progress of motion artifact correction in photoacoustic microscopy and photoacoustic tomography / 生物医学工程学杂志

Photoacoustic imaging (PAI) is a rapidly developing hybrid biomedical imaging technology, which is capable of providing structural and functional information of biological tissues. Due to inevitable motion of the imaging object, such as respiration, heartbeat or eye rotation, motion artifacts are observed in the reconstructed images, which reduce the imaging resolution and increase the difficulty of obtaining high-quality images. This paper summarizes current methods for correcting and compensating motion artifacts in photoacoustic microscopy (PAM) and photoacoustic tomography (PAT), discusses their advantages and limits and forecasts possible future work.

Review on photoacoustic tomographic image reconstruction for acoustically heterogeneous tissues / 生物医学工程学杂志

Acoustic properties of biological tissues usually vary inhomogeneously in space. Tissues with different chemical composition often have different acoustic properties. The assumption of acoustic homogeneity may lead to blurred details, misalignment of targets and artifacts in the reconstructed photoacoustic tomography (PAT) images. This paper summarizes the main solutions to PAT imaging of acoustically heterogeneous tissues, including the variable sound speed and acoustic attenuation. The advantages and limits of the methods are discussed and the possible future development is prospected.

Fast photoacoustic imaging systems using pulsed laser diodes: a review

Photoacoustic imaging (PAI) is a newly emerging imaging modality for preclinical and clinical applications. The conventional PAI systems use Q-switched Nd:YAG/OPO (Optical Parametric Oscillator) nanosecond lasers as excitation sources. Such lasers are expensive, bulky, and imaging speed is limited because of low pulse repetition rate. In recent years, the semiconductor laser technology has advanced to generate high-repetitions rate near-infrared pulsed lasers diodes (PLDs) which are reliable, less-expensive, hand-held, and light-weight, about 200 g. In this article, we review the development and demonstration of PLD based PAI systems for preclinical and clinical applications reported in recent years.

Photoacoustic Tomography for Micro-blood Vessel / 中国医学物理学杂志

Objectieve:In order to get the photoacoustic (PA) image of micro-blood vessel deep inside biological tissue.Materials and Methods:A 532nm pulse laser with the repetition rate of 10Hz was used as the pumping source.A wide band unfocused needle ultrasound transducer was used to collect time-domain PA signal of the tissue-mimicking phantom by circular scan.The 2D optical absorption image was reconstructed by time-domain spherical back projection algorithm.The spatial resolution was measured by line-spread-function(LSF).Results:The experimental results demonstrated that the spatial resolution was 0.1 mm and the PA image of the micro-vascular network agreed well with the tissue-mimicking phantom.Conclusion:The proposed method is valuable for the diagnosis of early cancer inside tissue.

Photoacoustic Tomography and Applications in The Medical Clinic Diagnosis / 生物化学与生物物理进展

Photoacoustic tomography is a developing, promising,non-invasive imaging method in the medical clinic diagnosis. It is an ultrasound-mediated biophotonic imaging method based on the intrinsic optical absorption properties of tissue and ultrasonic detection, and combines the merits of both high contrast advantage of pure optical imaging and high resolution advantage of pure ultrasound imaging. Photoacoustic tomography can be performed by detecting photoacoustic waves instead of detecting photons. In photoacoustic tomography, imaging contrast is based primarily on the optical properties of biological tissues, and imaging resolution is based primarily on the ultrasonic waves. It can avoid the influence of optical scattering on imaging resolution in principle, and can provide tomography of tissues with high contrast and high spatial resolution at medium depths. Photoacoustic tomography can provide an effective approach to studying the structures, physiological properties, metabolisms, pathological properties of biological tissues. It has important potential clinical applications in the early non-invasive detection of cancers, structural and functional in vivo imaging. A brief introduction of photoacoustic imaging mechanisms is gives, and the imaging methods, the image reconstruction algorithm and the potential biomedical applications of photoacoustic tomography are reveiewed.

Time-domain Photoacoustic Tomography of Mouse Brain Vessel in Vivo / 医疗卫生装备

Objective To get photoacoustic (PA) image of vascular structure inside mouse brain in vivo.Methods A 532nm pulsed laser was used as the pumping source to generate PA signal,and a wide-band PVDF unfocused ultrasound transducer was used to collect time-domain PA signal by circular scan.PA signal recorded at each scanning position was then de-noised by wavelets transform using soft-thresholding.The PA image of the mouse brain vessel was reconstructed by time-domain spherical back projection algorithm.Results PA image agreed well with the histological picture of mouse brain vessel.Conclusion The experimental result indicates the potential value of clinical application.