Photoacoustic Blood Pressure Recognition Based on Deep LearningXiaoman Zhang, Huaqin Wu, Biying Yu, Sulian Wu, Weijie Wu,Jianyong Cai*and Hui Li* Key Laboratory of OptoElectronicScience and Technology for Medicine of Ministry of Education,Fujian Provincial Key Laboratory of Photonics Technology,College of Photonic and Electronic Engineering, Fujian Normal University Ministry of Education, Fuzhou 350007, P.R. ChinaABSTRACTContinuous and non-invasive real-time measurement of human blood pressure is of great importance for health care and clinical diagnosis.Photoacoustic imaging allows absorption-based high-resolution spectroscopyin vivo imaging with a depth beyond that of optical microscopy. In this study,a novel photoacoustic imaging systemis usedfor monitoring and imaging of vesselpulsation,whichcan realize simple, non-invasive and continuous measurement and recognition of blood pressure. Combined with deep learning method, a model is established to effectively evaluate the dependence of blood vessel elasticity on theblood pressure.These results can quickly and accurately identify the photoacoustic signals of blood vessels under different pressures.
In the current clinical care, Gleason grading system based on the architectural pattern of cancerous epithelium in histological images is the most powerful prognostic predictor for prostate cancers (PCa). However, the standard procedure of histological examination often includes complicated tissue fixation and staining, which are time-consuming and may delay the diagnosis and surgery. In this study, the unstained prostate tissues were investigated with multiphoton microscopy (MPM) to produce subcellular-resolution images. And then, a deep learning network (AlexNet) was introduced for automated Gleason grading. We achieved an average accuracy of Gleason grading of 78.1%±3.4% for classification. And the area under the curve (AUC) in test set achieves 0.943 which indicates that the proposed model is effective in Gleason grading. At the end, the heat map was performed to visualize the Gleason score of tumour. Our results suggested that MPM, combined with deep learning method, holds the potential to be used as a real-time clinical diagnostic tool for PCa diagnosis.
Stage IA endometrial cancer is the only candidate for conservative management. Therefore, early diagnosis of endometrial cancer is very important. Co-registered photoacoustic (PA) and ultrasonic (US) imaging system is available to detect early endometrial cancer (EEC) based on a cylindrical diffuser. To correctly detect and diagnose EEC from FIGO stage IA and stage IB by co-registered PA and US imaging system, a convolutional neural network (CNN) classifier of EEC for co-registered PA and US images was proposed. Activation function ReLU and the dropout technique were used in the CNN classifier. The experiment results show the area under the receiver operating characteristic curve of the proposed algorithm is 0.9998 with a sensitivity of 98.75% and specificity of 98.75%. The CNN classifier could be used in the computer-aided diagnosis for early endometrial cancer of the co-registered PA and US imaging system.
This paper describes a novel method for correcting a color cast in an image which contains RGB and near-infrared crosstalk. We name this type of images as four-band images. The algorithm is based on the back propagation neural network. When a four-band image is put into the algorithm, it will use a trained weight matrix to transform this image into the normal one. This matrix is obtained by a BP algorithm which is used to find the implicit relationship between the input image and the corresponding target image. The size of the matrix is determined by the number of hidden layers of the neural network and the number of rows of the pixel matrix of the input and output images. In this paper, we acquire the weight matrix of the input layer to the hidden layer with a size of 506*486 and the weight matrix of the hidden layer to the output layer of 486*506. The camera used herein is a slightly modified CMOS sensor that replaces the IR-Cut filter with an 850 nm bimodal filter. The dataset used in this paper has five pairs of one-to-one correspondence images. One of them is a three-band image with normal color using the original IR-Cut filter, and the other is a four-band image taken by the modified CMOS sensor. The proposed method has been tuned and tested with positive results in this dataset.
Photoacoustic Imaging (PAI) has potential for clinical applications in real-time after a tiny modification of a current US scanner. The shared detector platform facilitates a natural integration of PA and US imaging creating a hybrid imaging technique that combines functional and structural information. In this work, two blood vessels phantom experiment was conducted by coregistered photoacoustic and ultrasonic imaging using clinical ultrasonic system. The vessels were placed about 6 cm away from the transducer. With conventional irradiation, real-time PA and US images could be obtained during the experiment. 450 of 2D PA and US images and reconstructed 3D imaging were taken by transducer scanning. The result indicates the system has the ability to get the PA signal in a deep tissue depth. 3D PA image clearly describes the tissue structure and benefits the detecting in clinical application.
Photoacoustic imaging (PAI) is a promising technique to image tumor angiogenesis development and detect endometrial carcinoma in earlier stages. The light absorption distribution of uterine tissue determines the imaging depth and range of PAI. In this work, a 3D triangular meshes tumor-embedded uterine optical model was established by the histological structure of uterus. The model is filled with strong scattering media (undiluted raw and homogenized milk, URHM) and air, respectively. Monte Carlo simulation is implemented based on the molecular optical simulation environment (MOSE) to find the absorption profiles of photons by transcervical laser illumination with cylindrically diffused light source (CDLS) or spherically diffused light source (SDLS) at wavelength 800nm. The results show the media with an extremely high scattering coefficient and an extremely low absorption coefficient like URHM helps the light propagations in a relatively small cavity. CDLS performs better when the tumor happens far from the light source center than the SDLS. On the same time, embedded tumors of the model filled with URHM are easier to detect by the transcervical laser illumination of CDLS than SDLS in the fundus of the uterus. The conclusions are helpful to optimize the laser source and to improve the imaging depth in a photoacoustic imaging system.
Resting heart rate (RHR) is considered an important biomedical indicator to evaluate cardiovascular function. High RHR is an important prognostic factor for sudden cardiac death and heart failure in the general population, and especially among patients with known cardiac disease. The imaging photoplethysmography (IPPG) technology is used to achieve the accurate detection of RHR signal, which has the advantages of low cost, simple operation, fast acquisition speed etc. In this paper, we propose a new simple, inexpensive and easy-to-use method to measure the RHR in vivo. The result shows that Fast Fourier Transform with Hamming window filters, band-pass filter gives more accurate results. The color change of the fingertip is enlarged by using the mobile phone camera. From the distribution of color change of the fingertip, the RHR is estimated with the primary calibration result of the relationship between color variation and the blood volume change.
The Jones matrix and the Mueller matrix are main tools to study polarization devices. The Mueller matrix can also be used for biological tissue research to get complete tissue properties, while the commercial optical coherence tomography system does not give relevant analysis function. Based on the LabVIEW, a near real time display method of Mueller matrix image of biological tissue is developed and it gives the corresponding phase retardant image simultaneously. A quarter-wave plate was placed at 45 in the sample arm. Experimental results of the two orthogonal channels show that the phase retardance based on incident light vector fixed mode and the Mueller matrix based on incident light vector dynamic mode can provide an effective analysis method of the existing system.
Quantitative methods for noninvasive diagnosis of scars are a challenging issue in medicine. This work aims to implement a texture analysis method for quantitatively discriminating abnormal scars from normal scars based on second-harmonic generation (SHG) images. A local difference local binary pattern (LD-LBP) operator combined with a wavelet transform was explored to extract diagnosis features from scar SHG images that were related to the alteration in collagen morphology. Based on the quantitative parameters including the homogeneity, directional and coarse features in SHG images, the scar collagen SHG images were classified into normal or abnormal scars by a support vector machine classifier in a leave-one-out cross-validation procedure. Our experiments and data analyses demonstrated apparent differences between normal and abnormal scars in terms of their morphological structure of collagen. By comparing with gray level co-occurrence matrix, wavelet transform, and combined basic local binary pattern and wavelet transform with respect to the accuracy and receiver operating characteristic analysis, the method proposed herein was demonstrated to achieve higher accuracy and more reliable classification of SHG images. This result indicated that the extracted texture features with the proposed method were effective in the classification of scars. It could provide assistance for physicians in the diagnostic process.
A new virtual instrument of spectroscopy based on LabVIEW was developed for the diagnosis system of nasopharyngeal carcinoma. Some methods of the signal acquisition of wavelength calibration were devised and then discussed in detail including the LabVIEW programming. The mechanical part of the diagnosis system is also presented in this paper.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.