In the photoacoustic microscopy coupled with the optical fiber, the photoacoustic intensity of the irradiated tissue is one of most important factors of Furthermore, the coupling coefficient of the fiber also impacts the final irradiated laser energy. Furthermore, the coupling coefficient of the fiber also impacts the final irradiated laser energy. At the characteristic wavelength of 532nm, the effects of the optical positions of the With the operations of parameters scanning, the optimal values of the optical positions of the focusing lens and optical fiber on the coupling efficiency of the fiber were investigated. values of the optical positions of the focusing lens and optical fiber were obtained under the maximum coupling efficiency of the fiber. After that, the effect of the fiber's mode field diameter on the coupling efficiency of fiber under the optimal positions of the focusing lens and fiber was also investigated. The coupling efficiencies of fiber corresponding to seven different mode field diameters of fiber from 1 to 9μm were computed, the The study results show that with the increase of the mode field Under the optimal positions of the focusing lens and the fiber, as well as the mode field diameter, the optical efficiency of fiber first increase then decrease. Under the optimal positions of the focusing lens and the fiber, as well as the mode field diameter, the optical efficiency of fiber can be improved from 23.174% to 91.638%. Therefore, the reasonable positions of the optical path and the mode field diameter of the fiber are all important to ensure the satisfactory optical Therefore, the reasonable positions of the optical path and the mode field diameter of the fiber are all important to ensure the satisfactory optical coupling efficiency in the photoacoustic microscopy system coupled with the optical fiber.
In this study, 100 groups of apples with different sweetness were measured in transmission mode using visible light spectroscopy (VIS). The absorption spectra of all samples were obtained in the wavelength range of 400-800 nm with a step of 5 nm. To classify and identify the sweetness of apples, a qualitative classification model of apple absorption spectra and sweetness was constructed using BP neural network. The sweetness of all apples was classified into three different classes and labeled with Arabic numbers from one to three. In the experiment, 80 groups of apples were randomly selected as training samples and 20 groups of apples as test samples. Through the test, the sweetness classification accuracy of the test samples based on BP neural network reached 75%. To further improve the classification accuracy of sweetness, a Particle Swarm Optimization (PSO) algorithm was used to optimize the parameters of the BP neural network. With the optimal values of BP-PSO model parameters, the sweetness classification accuracy reached 90% for 20 sets of test samples. Finally, traditional classification models of extreme learning machine (ELM), competitive neural network (CNN) and self-organizing mapping neural network (SOMNN) were established to compare the classification accuracy of different algorithms, and the accuracy of 50%, 35% and 65% was achieved using ELM, CNN and SOMNN models, respectively. The results show that the classification using BP-PSO model has higher classification accuracy. Therefore, the BP-PSO model can be applied to the quality identification and classification of apples based on VIS technique.
In this work, the photoacoustic detection of blood glucose with the interference of multiple factors was studied. A set of photoacoustic detection system of blood glucose was established, in which the interference of multiple factors including the laser energy, concentration, temperature, flow velocity and detection distance were combined into. Under different conditions of multiple factors, the time-resolved photoacoustic signals and peak-to-peak values of blood samples were all obtained. To accurately classify the concentration of blood glucose samples, back propagation (BP) neural network was employed to train the photoacoustic peak-to-peak values and the multiple factors. In BP neural network, five different Arabic numerals from 1 to 5 were labeled to denote five kinds of blood glucose levels ranged from2mmol/Lto14mmol/L. The photoacoustic peak-to-peak values, laser energy, temperature, flow velocity and detection distance were used as the input data, the labels denoted different concentrations were used as the output data. Meanwhile, the effects of neurons number in hidden layer and learning factor on the classification accuracy of blood glucose level were investigated. Under the optimal parameters of BP neural network, the accuracy of classifying concentration of blood glucose level reached 85.6% for the test blood glucose samples. Compared with the classification accuracy (71.2%) of blood glucose level based on support vector machine (SVM) algorithm, it is demonstrated that the photoacoustic spectroscopy combined with BP neural network has a good performance in qualitative classification of blood glucose under the interference of multiple factors.
In this study, the visible light spectroscopy was used to achieve the sweetness quantitative measurement of apple. In the experiments, the absorption spectra of apple samples in total of 100 groups were obtained in the waveband from 400-800nm with interval of 5nm by using the visible light spectroscopy. At the same time, the real sweetness values of all apples were measured by using a commercial fruit sugar meter. To achieve the sweetness quantitative spectral measurement, the back propagation (BP) neural network was used to supervised train the absorption spectral for 80 groups of training samples, and 20 groups of apples were utilized as the test samples. The effects of neuron numbers in the hidden layer, learning rate factor and the training times on the root-mean-square error (RMSE) of sweetness were investigated. Under the optimal parameters of BP neural network, the RMSE of sweetness for the test apple samples can reach 0.12218%, which is superior to that of the commercial fruit sugar meter (0.2%). Compared with the correlation coefficients for the training samples and test samples based on the partial least square (PLS) algorithm, it can be demonstrated that the visible light spectroscopy combined with BP neural network has the potential superiority and application value in the sweetness quantitative spectral measurement of fruit.
In the system of photoacoustic microscopy coupled with fiber, the coupling efficiency of fiber is one of most important factors to ensure the photoacoustic intensity under the adequate absorbed energy of pulsed laser for the irradiated tissue. In this work, the effects of aspheric lens parameters on the coupling efficiency of fiber with Gaussian pulsed laser were studied under their reasonable ranges of the parameters. The parameters of the aspheric lens includes the curvature radius, cone constant, thickness, tilt angle, radial deviation of vertical shaft. At the same distances between the aspheric lens, fiber, and the parameters of fiber and pulsed laser, the coupling efficiencies of optical fiber under different values of all parameters of aspheric lens were obtained and compared. The influence laws of all parameters of aspheric lens on the coupling efficiency of optical fiber were all obtained. Studies results show that under the optimal distances of aspheric lens and fiber, with the increase of curvature radius, cone constant, thickness, tilt angle, radial deviation of vertical shaft, the coupling efficiency of optical fiber all first increases then decreases with the Gaussian-liked function. Moreover, the parameters scan optimization method was used to obtain the optimal values of parameters of aspheric lens. Studies show that the coupling efficiency of optical fiber can reach 85.45% when curvature radius is 1.8903mm, cone constant is -2.0627 , thickness is 250μm, tilt angle is 0°, radial deviation is 0mm.
In this work, we used photoacoustic spectroscopy to distinguish the different types of blood including four kinds of true blood and two kinds of fake blood. The peak-to-peak spectra of blood were obtained in the wavelength from 700nm to 1064nm based on the established photoacoustic detection system of blood. To accurately discriminate the different types of blood, back propagation (BP) neural network was used to train the photoacoustic peak-to-peak spectra of training blood with 120 groups, the correct rate of distinguishing blood is 76.7% for 30 groups of test samples. Particle swarm optimization (PSO) algorithm was used to optimize the parameters of BP network. The effects of neurons number in the hidden layer, learning rate factor, inertia weight, two acceleration factors, iteration times and training times on the corret rate and mean square error were all investigated. Under the optimal parameters, the correct rate of BP-PSO algorithm was increased to 93.3%. To further improve the correct rate, the dynamic inertia weight strategy was used. Moreover, a kind of improved dynamic inertia weight strategy function was proposed. The correct rate of the improved dynamic inertia weight strategy function was compared with that of the static inertia weight and two other dynamic inertia weight strategy functions. Under the optimal value of the improved dynamic inertia weight, the correct rate reached 96.7%. Therefore, the BP-PSO algorithm combined with the improved dynamic inertia weight strategy function has a potential value in the photoacoustic discrimination of blood.
In this work, to study the effects of multiple factors on the blood glucose photoacoustic detection, five different factors including the laser energy, concentration, temperature, flow velocity and detection distance were considered, and a set of blood glucose photoacoustic detection system combined multiple influence factors was established. The time-resolved photoacoustic signals and peak-to-peak spectra of 625 groups of blood samples were obtained. To predict the blood glucose concentration with high accuracy under the influence of multiple factors, back propagation (BP) neural network was used to train five different factors and photoacoustic peak-to-peak values of 500 groups of blood samples, and 125 groups of blood samples were used as the test samples. Meanwhile, the effects of neurons number in the hidden layer, learning rate and training times on the root-mean-square error(RMSE) of predicting blood glucose concentration were investigated. Under the optimal parameters, the RMSE of blood glucose concentration for 125 groups of test blood samples is about 0.807679mmol/L. Compared with the results of partial least square (PLS) algorithm with RMSE of 1.78mmol/L, it is demonstrated that the BP algorithm has good performance in the prediction blood glucose concentration under multiple influence factors based on photoacoustic detection technology.
The blood is an important tissue in the human and animal body. It has significant role in the fields of bio-medical diagnosis, animal quarantine, criminal investigation, food safety, etc. However, there are some illegal cases reported about the real blood abused by fake blood recently, which seriously impact the human health and society stability. The rapid and accuracy detection of blood is very important and urgent. To achieve this aim, the photoacoustic spectroscopy was used to detect the real blood and fake one. A set of photoacoustic detection system was established based on OPO pulsed laser and focused ultrasonic detector. In experiments, 150 groups of real and fake blood samples was test, where 120 groups were used as the training samples, 30 groups were used as the test samples. The time-resolved photoacoustic signal and peak-to-peak values of all samples were captured in the wavelengths from 700-1064nm. To classify and distinguish the real and fake blood, the support vector machine (SVM) algorithm was used to train the training blood samples and test the correct rate of classification and distinction of the real and fake blood. The results show that the correct rate is 83.3% by using the SVM algorithm. To further improve the correct rate, the principal components analysis (PCA) algorithm was used to extract the characteristic information from the photoacoustic peak-to-peak values of blood samples in full wavelengths. The correct rates of real and fake blood based on PCA-SVM algorithm under the different principal components were obtained and compared. The results show that the correct rate can be improved to 90% for the PCA-SVM algorithm with 21 principal components.
The non-invasive detection of blood glucose based on photoacoustic spectroscopy is a very popular method used to monitor the diabetes mellitus in recent years. The basic mechanism of photoacoustic spectroscopy is the effect of photo-induced ultrasonic. The detection accuracy of blood glucose can be improved due to the captured ultrasonic rather than the photons. The properties of ultrasonic transducer is one of the important influence factors on the detection accuracy of glucose. To study the effect of detection frequency for ultrasonic transducer on the glucose detection based on photoacoustic spectroscopy, a set of photoacoustic detection system was established. The optical parameters oscillator (OPO) pulsed laser pumped with 532nm was used as the excitation light source. Three kinds of ultrasonic transducers with different central echo frequencies (1MHz, 2.5MHz, and 30MHz) were respectively used to capture the photoacoustic signal of test blood phantoms with different concentrations of glucose. The time-resolved photoacoustic signal and peak-to-peak values of test blood phantoms were obtained. The results show that with the increase of glucose concentration, the photoacoustic amplitudes and peak-to-peak values of phantoms increase. Moreover, the time of photoacoustic signal shifts left with the increase of concentration. The prediction models based on linear fitting method were established for three kinds of ultrasonic transducers. Prediction results show that for the ultrasonic transducer with central frequency of 1MHz, the correction coefficient is 0.8681, the root-mean-square error (RMSE) of glucose concentration is about 13.3%. For the ultrasonic transducer with central frequency of 2.5MHz, the correction coefficient is 0.83127, the RMSE of glucose concentration is about 1.8149%. For the ultrasonic transducer with central frequency of 30MHz, the correction coefficient is 0.99598, the RMSE of glucose concentration is about 0.3808%. Therefore, the detection effect of the ultrasonic transducer with central frequency of 30MHz is best compared with the two others.
To achieve the identification of true and fake blood, the near infrared spectroscopy method was used in this work. The optical absorption spectra of blood samples with 120 groups of training samples and 30 groups of test samples were obtained via a Fourier transform NIR spectroscope. Since the similar spectra profiles and spectra overlap between the blood samples, the accurate identification of true blood and fake blood is difficult from the visual viewpoint. The wavelet neural network was used to train and test the blood samples. The correct rate of identifying true and fake blood is only 23.3%. To improve the correct rate, the particle swarm optimization (PSO) algorithm was used to optimize the weights, two learning rate factors, translation factor and scaling factor of WNN network. At the same time, the effects of the neuron number in the hidden layer, two learning rate factors, two acceleration factors, iteration times and training times on the correct rate and mean square error of identifying blood based on WNN-PSO algorithm were investigated. Under the optimal parameters, the correct rate of WNN-PSO algorithm is improved to 53.3%. Then, the principal component analysis (PCA) method was used to further improve the correct rate. The effect of different principal components on the correct rate of identifying blood based on PCA-WNN-PSO algorithm was also investigated. The results show that the correct rate can reach 96.7% for the identification of blood by using the NIR spectroscopy combined with PCA-WNN-PSO algorithm.
Since the disadvantages of destructive and high cost for the traditional detection methods, a non-invasive identification method based on near infrared spectroscopy was used in this paper. The near infrared spectra of 120 groups of training blood samples and 30 groups of test samples were obtained from 4000cm-1 to 10000cm-1 for four kinds of animal blood and two kinds of fake blood. The classification and identification of blood can’t be easily achieved because of the near infrared spectra overlapping. To accurately identify the different kinds of the blood, back propagation (BP) neural network was used to establish the classification model. The spectra in full wavelengths were used as the input data, and 1, 2, 3, 4, 5, 6 were used to label different blood. Based on the training of 120 groups of training blood samples, the correct rate of blood identification for 30 groups of test samples are 66.7%. To further improve the correct rate, the weights of BP neural network were optimized by the particle swarm optimization (PSO). The effects of neurons number, learn rate factor, iteration times, and training times on the correct rate and mean square error for the identification of blood based on BP-PSO algorithm were investigated. Under the optimized parameters, the correct rate was improved to 96.7%.
In this paper, a kind of photoacoustic detection system was established to measure the photoacoustic signals of glucose with different concentrations. In this system, the lateral model was used based on the optical parameters oscillator (OPO) pulsed laser and ultrasonic transducer. Meanwhile, the optimal absorption of 1064nm for the pulsed laser and the ultrasonic transducer with central frequency of 1MHz was used. The amplified time-resolved photoacoustic signals and photoacoustic peak-to-peak values of glucose aqueous solutions with different concentrations were obtained in the average of 128 times. In addition, the effect of detection distance on the photoacoustic detection of glucoses was experimentally investigated. The detection distance includes the vertical and the transverse distances between the photoacoustic source in the glucose aqueous solutions and the ultrasonic transducer were all considered. Moreover, the effect of detection distance on the prediction of glucose concentrations was also studied by using the linear fitting model. Experimental results show that for each glucose solution with different concentrations, the photoacoustic peak-to-peak value of glucose decreases in the form of exponent function with the increase of detection distance, and the peak point of glucose will shift toward right due to the increase of detection distance although the concentration isn’t changed. When the transverse distance between the photoacoustic source and the ultrasonic transducer increases, the photoacoustic peak-to-peak value of glucose linearly decreases. At the same time, the prediction accuracy of glucose concentration decreases with the increase of vertical and horizontal detection distances.
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.