This PDF file contains the front matter associated with SPIE Proceedings Volume 10245, including the Title Page, Copyright information, Table of Contents, and Conference Committee listing.

Medical imaging has been widely used in clinical practice. It is an important basis for medical experts to diagnose the disease. However, medical images have many unstable factors such as complex imaging mechanism, the target displacement will cause constructed defect and the partial volume effect will lead to error and equipment wear, which increases the complexity of subsequent image processing greatly. The segmentation algorithm which based on SLIC (Simple Linear Iterative Clustering, SLIC) superpixels is used to eliminate the influence of constructed defect and noise by means of the feature similarity in the preprocessing stage. At the same time, excellent clustering effect can reduce the complexity of the algorithm extremely, which provides an effective basis for the rapid diagnosis of experts.

Photoacoustic imaging can provide high-resolution and high-contrast image under unprecedented depth compared with pure optical imaging techniques by making use of laser-induced ultrasound waves. Although a series of absorption-enhanced optical contrast agents for photoacoustic imaging were developed, the probe with fully conversion from absorbed light energy to acoustic energy has not been achieved so far. Here we develop a high-efficiency photoacoustic probes with fluorescence resonance energy transfer (FRET) effect for enhancement of nonradiative energy. Graphene oxide (GO) binding optical dyes (GO-dyes) were achieved to show highly fluorescence quenching and violently increased photoacoustic signal intensity. GO-dyes were constructed and testified for multi-spectral photoacoustic imaging. As a representative probe, GO-Cy7 nanoparticles were used to validate the feasibility of photoacoustic tumor molecular imaging in vivo. Our work demonstrated a new approach to construct high-efficiency FRET-enhanced multi-spectrum probes for photoacoustic molecular imaging.

This paper studies the realization of image processing algorithm of multispectral endoscope. The research contents include: local brightness enhancement and adaptive contrast enhancement. Firstly, this paper transforms the image from the RGB space to the HSV space, and then carries on the image enhancement processing to the V space, finally transforms to the RGB space. Local brightness enhancement algorithm divides V space image into smaller windows, and then calculates the nonlinear transfer function of each window, which enhances the pixels in the window, and finally the contrast of brightness enhanced image is restored. The adaptive contrast enhancement adopts the unsharp mask technique based on the guided filter. First of all, this paper uses guided filter to the RGB channel of the original image and gets the unsharp mask of each channel, then plus a scaled image which is the result of the original image subtracts the unsharp mask. So the enhancement of the image is achieved. This paper uses subjective evaluation criteria and enhance factor α to evaluate the effect of enhancement.

And this paper compares the enhancement effect of the proposed image enhancement algorithm and the traditional algorithm. The results show that the α of histogram equalization is smallest and AINDANE method is better than histogram equalization. The proposed method has the best α. The subjective evaluation also shows that the effect of HE is not satisfactory and the proposed method enhances the detail information tremendously. The subjective and objective criteria shows that the proposed method produces better enhancement effect.

Early detection of knee osteoarthritis (KOA) is meaningful to delay or prevent the onset of osteoarthritis. In consideration of structural complexity of knee joint, position of light incidence and detector appears to be extremely important in optical inspection. In this paper, the propagation of 780-nm near infrared photons in three-dimensional knee joint model is simulated by Monte Carlo (MC) method. Six light incident locations are chosen in total to analyze the influence of incident and detecting location on the number of detected signal photons and signal to noise ratio (SNR). Firstly, a three-dimensional photon propagation model of knee joint is reconstructed based on CT images. Then, MC simulation is performed to study the propagation of photons in three-dimensional knee joint model. Photons which finally migrate out of knee joint surface are numerically analyzed. By analyzing the number of signal photons and SNR from the six given incident locations, the optimal incident and detecting location is defined. Finally, a series of phantom experiments are conducted to verify the simulation results. According to the simulation and phantom experiments results, the best incident location is near the right side of meniscus at the rear end of left knee joint and the detector is supposed to be set near patella, correspondingly.

Total internal reflection fluorescence microscopy (TIRFM) has been widely used in biomedical research to visualize cellular processes near the cell surface. In this study, a novel multi-angle ring-illuminated TIRFM system, equipped with two galvo mirrors that are on conjugate plan of a 4f optical system was developed. Multi-angle TIRFM generates images with different penetration depths through the controlled variation of the incident angle of illuminating laser. We presented a method to perform three-dimensional (3-D) reconstruction of microtubules from multi-angle TIRFM images. The performance of our method was validated in simulated microtubules with variable signal-to-noise ratios (SNR) and the axial resolution and accuracy of reconstruction were evaluated in selecting different numbers of illumination angles or in different SNR conditions. In U373 cells, we reconstructed the 3-D localization of microtubules near the cell surface with high resolution using over a hundred different illumination angles. Theoretically, the presented TIRFM setup and 3-D reconstruction method can achieve ∼40 nm axial resolution in experimental conditions where SNR is as low as 2, with ∼35 different illumination angles. Moreover, our system and reconstruction method have the potential to be used in live cells to track membrane dynamics in 3-D.

Fluorescence molecular tomography (FMT) is an imaging modality that exploits the specificity of fluorescent biomarkers to enable 3D visualization of molecular targets and pathways in small animals. FMT has been used in surgical navigation for tumor resection and has many potential applications at the physiological, metabolic, and molecular levels in tissues. The hybrid system combined FMT and X-ray computed tomography (XCT) was pursued for accurate detection. However, the result is usually over-smoothed and over-shrunk. In this paper, we propose a region reconstruction method for FMT in which the elastic net (E-net) regularization is used to combine L1-norm and L2-norm. The E-net penalty corresponds to adding the L1-norm penalty and a L2-norm penalty. Elastic net combines the advantages of L1-norm regularization and L2-norm regularization. It could achieve the balance between the sparsity and smooth by simultaneously employing the L1-norm and the L2-norm. To solve the problem effectively, the proximal gradient algorithms was used to accelerate the computation. To evaluate the performance of the proposed E-net method, numerical phantom experiments are conducted. The simulation study shows that the proposed method achieves accurate and is able to reconstruct image effectively.

Diffraction Phase Microscopy (DPM) is a widely used quantitative phase imaging method, whose common-path nature endows it with low noise and high sensitivity. Current applications of DPM include biological topography as well as biological dynamics for its nondestructive feature. Many different forms of DPM based on the original idea have appeared according to the different demands. In this paper, both the principle and the DPM classification will be given a comprehensive description. Furthermore, the future trend of DPM development is also discussed.

In this paper, we report the coupling between the localized surface plasmon resonance (LSPR) of Au-nanoparticles and surface plasmon resonance (SPR) of the Au-film. According to the conditions for SPR excitation of the classical Kretschmann-Raether structure with 50nm Au thin film, the commonly used classes of spherical Au-nanoparticle is studied and optimized. We used the finite element analysis (COMSOL Multiphysics 5.0), to simulate the coupling. The results from calculation and simulation indicate that the resonant plasmonic coupling between Au-nanoparticles and Au-film could lead to a large field enhancement and thus improve SPR. We demonstrate that the resonant plasmonic coupling could be regulated by the size of nanoparticles, the distance between nanoparticles .

One method for three-dimensional (3D) measurement based on structured light is proposed for the medical electric endoscope in the present study. The structured light of black and white strips is generated by that the point sources illuminate the grating mask plate. Four point sources are aligned linearly by a certain space and they are lighted sequentially. Then four images of modulated fringes by the height of object with different phase shifts can be obtained. The algorithm proposed by Wang Z is employed to extract the accurate phase shift from the fringe images since the phase shift cannot be exactly set as π/2 by hardware. An experimental prototype endoscope was built according to the proposed method. One high-definition CMOS camera module developed by ourselves was used to acquire the endoscopic images and the structured light was generated by four fiber LEDs and a transmission grating with a pitch of 0.1 mm. one C# program was designed to light up LEDs in turn, acquire the phase shifted images and calculate the 3D information. The experimental results indicate that its precise of depth measurement at the working distance of 40 mm is better than 0.5 mm and its consuming time of 3D depth calculation is less than 0.5 s.

Back propagation neural network(BP neural network) is a type of multi-layer feed forward network which spread positively, while the error spread backwardly. Since BP network has advantages in learning and storing the mapping between a large number of input and output layers without complex mathematical equations to describe the mapping relationship, it is most widely used. BP can iteratively compute the weight coefficients and thresholds of the network based on the training and back propagation of samples, which can minimize the error sum of squares of the network.

Since the boundary of the computed tomography (CT) heart images is usually discontinuous, and it exist large changes in the volume and boundary of heart images, The conventional segmentation such as region growing and watershed algorithm can’t achieve satisfactory results. Meanwhile, there are large differences between the diastolic and systolic images. The conventional methods can’t accurately classify the two cases.

In this paper, we introduced BP to handle the segmentation of heart images. We segmented a large amount of CT images artificially to obtain the samples, and the BP network was trained based on these samples. To acquire the appropriate BP network for the segmentation of heart images, we normalized the heart images, and extract the gray-level information of the heart. Then the boundary of the images was input into the network to compare the differences between the theoretical output and the actual output, and we reinput the errors into the BP network to modify the weight coefficients of layers. Through a large amount of training, the BP network tend to be stable, and the weight coefficients of layers can be determined, which means the relationship between the CT images and the boundary of heart.

The paper proposes a registration method on 3D point cloud of the bone tissue surface extracted by B-mode ultrasound image and the CT model . The B-mode ultrasound is used to get two-dimensional images of the femur tissue . The binocular stereo vision tracker is used to obtain spatial position and orientation of the optical positioning device fixed on the ultrasound probe. The combining of the two kind of data generates 3D point cloud of the bone tissue surface. The pixel coordinates of the bone surface are automatically obtained from ultrasound image using an improved local phase symmetry (phase symmetry, PS) . The mapping of the pixel coordinates on the ultrasound image and 3D space is obtained through a series of calibration methods. In order to detect the effect of registration, six markers are implanted on a complete fresh pig femoral .The actual coordinates of the marks are measured with two methods. The first method is to get the coordinates with measuring tools under a coordinate system. The second is to measure the coordinates of the markers in the CT model registered with 3D point cloud using the ICP registration algorithm under the same coordinate system. Ten registration experiments are carried out in the same way. Error results are obtained by comparing the two sets of mark point coordinates obtained by two different methods. The results is that a minimum error is 1.34mm, the maximum error is 3.22mm,and the average error of 2.52mm; ICP registration algorithm calculates the average error of 0.89mm and a standard deviation of 0.62mm.This evaluation standards of registration accuracy is different from the average error obtained by the ICP registration algorithm. It can be intuitive to show the error caused by the operation of clinical doctors. Reference to the accuracy requirements of different operation in the Department of orthopedics, the method can be apply to the bone reduction and the anterior cruciate ligament surgery.

Aiming the high failure rate, the high radiation quantity and the poor positioning accuracy of femoral neck traditional surgery, this article develops a set of new positioning robot system of femoral neck hollow screw implants based on X-rays error correction, which bases on the study of x-rays perspective principle and the Motion Principle of 6 DOF(degree of freedom) series robot UR(Universal Robots). Compared with Computer Assisted Navigation System, this system owns better positioning accuracy and more simple operation. In addition, without extra Equipment of Visual Tracking, this system can reduce a lot of cost. During the surgery, Doctor can plan the operation path and the pose of mark needle according to the positive and lateral X-rays images of patients. Then they can calculate the pixel ratio according to the ratio of the actual length of mark line and the length on image. After that, they can calculate the amount of exercise of UR Robot according to the relative position between operation path and guide pin and the fixed relationship between guide pin and UR robot. Then, they can control UR to drive the positioning guide pin to the operation path. At this point, check the positioning guide pin and the planning path is coincident, if not, repeat the previous steps, until the positioning guide pin and the planning path coincide which will eventually complete the positioning operation. Moreover, to verify the positioning accuracy, this paper make an errors analysis aiming to thirty cases of the experimental model of bone. The result shows that the motion accuracy of the UR Robot is 0.15mm and the Integral error precision is within 0.8mm. To verify the clinical feasibility of this system, this article analysis on three cases of the clinical experiment. In the whole process of positioning, the X-rays irradiation time is 2-3s, the number of perspective is 3-5 and the whole positioning time is 7-10min. The result shows that this system can complete accurately femoral neck positioning surgery. Meanwhile, it can greatly reduce the X-rays radiation of medical staff and patients. To summarize, it has a significant value in clinical application.

Skin cholesterol is a novel biomarker to assess the risk of atherosclerotic diseases. To detect skin cholesterol noninvasively and rapidly, a system was designed based on the diffuse reflectance spectroscopy. The feasibility of this system was validated through detecting cholesterol of pig skin samples, and skin cholesterol in vivo of subjects. The experimental results showed that, diffuse reflectance absorbance integrated intensity S measured the concentration of cholesterol in the pig skin samples quantitatively. After adjusting for age, gender and other factors, it showed a significant positive correlation between S of subjects and the total cholesterol (TC), low density lipoprotein cholesterol (LDL-C) which were the main risk factors for atherosclerotic diseases with the correlation coefficients 0.860(P <0.01) and 0.787(P <0.01). The study has shown that the diffuse reflectance spectroscopy provides a noninvasive and convenient method for the detection of skin cholesterol, and the noninvasive detection of skin cholesterol in vivo will contribute to the early detection of atherosclerotic diseases.

In this study, finite-difference time-domain (FDTD) algorithm has been used to work out the cell light scattering problem. Before beginning to do the simulation contrast, finding out the changes or the differences between normal cells and abnormal cells which may be cancerous or maldevelopment is necessary. The preparation of simulation are building up the simple cell model of cell which consists of organelles, nucleus and cytoplasm and setting up the suitable precision of mesh. Meanwhile, setting up the total field scattering field source as the excitation source and far field projection analysis group is also important. Every step need to be explained by the principles of mathematic such as the numerical dispersion, perfect matched layer boundary condition and near-far field extrapolation. The consequences of simulation indicated that the position of nucleus changed will increase the back scattering intensity and the significant difference on the peak value of scattering intensity may result from the changes of the size of cytoplasm. The study may help us find out the regulations based on the simulation consequences and the regulations can be meaningful for early diagnosis of cancers.

Sustained attention is the ability to keep focused and vigilance for long time in external stimulation, which was crucial in safe-critical human-machine system. While the ability of sustained attention will decline because of mental fatigue, even lead to serious accidents in fatigue state. Therefore, it is of great significance to explore the impact of fatigue on sustained attention. Functional Near-Infrared Spectroscopy (fNIRS) can measure cerebral hemoglobin in order to reflect cognitive function indirectly. In previous related fatigue studies, monotonous and long-time CPT (continuous performance test task) was often used to explore the performance change and brain activity, but the effect of time on task (TOT) was always involved. In this study, in order to avoid the TOT effect, the sustained attention task and fatigue task were separated. It was adopted in the study that the modified continuous performance test (CPT) was chosen as the sustained attention task and verbal 2-back task as the fatigue induced task. The fNIRS signals were extracted from 10 channels in the prefrontal cortex (PFC) from 20 healthy subjects. Studies found that cerebral lateralization increased significantly from alert to fatigue state in sustained attention task. Besides, Average oxyhemoglobin (HBO) of PFC increased significantly from alert to fatigue task, and the spatial pattern of activity of oxyhemoglobin also changed, which c be sensitive features to fatigue detection.

In the domain of biomedical signals measurements, monitoring human physiological parameters is an important issue. With the rapid development of wireless body area network, it makes monitor, transmit and record physiological parameters faster and more convenient. Infants and the elderly completely bedridden are two special groups of the society who need more medical care. According to researches investigating current frontier domains and the market products, the detection of physiological parameters from the excrement is rare. However, urine and faeces contain a large number of physiological information, which are high relative to health. The mainly distributed odour from urine is NH₄ and the distributed odour from feces is mainly H₂S, which are both could be detected by the sensors. In this paper, we introduce the design and implementation of a portable wireless device based on body area network for real time monitoring to the odour of excrement for health of infants and the elderly completely bedridden. The device not only could monitor in real time the emitted odour of faeces and urine for health analysis, but also measures the body temperature and environment humidity, and send data to the mobile phone of paramedics to alarm or the server for storage and process, which has prospect to monitoring infants and the paralysis elderly.

Functional near-infrared spectroscopy (fNIRS), which can measure cortex hemoglobin activity, has been widely adopted in brain-computer interface (BCI). To explore the feasibility of recognizing motor imagery (MI) and motor execution (ME) in the same motion. We measured changes of oxygenated hemoglobin (HBO) and deoxygenated hemoglobin (HBR) on PFC and Motor Cortex (MC) when 15 subjects performing hand extension and finger tapping tasks. The mean, slope, quadratic coefficient and approximate entropy features were extracted from HBO as the input of support vector machine (SVM). For the four-class fNIRS-BCI classifiers, we realized 87.65% and 87.58% classification accuracy corresponding to hand extension and finger tapping tasks. In conclusion, it is effective for fNIRS-BCI to recognize MI and ME in the same motion.

We present a spatial frequency domain imaging (SFDI) study of local hemodynamics in the forearm of healthy volunteers performing paced breathing. Real time Single Snapshot Multiple Frequency Demodulation - Spatial Frequency Domain Imaging (SSMD-SFDI) was used to map the optical properties of the subsurface of the forearm continuously. The oscillations of the concentrations of deoxy- and oxyhemoglobin at the subsurface of the forearm induced by paced breathing are found to be close to out-of-phase, attributed to the dominance of the blood flow modulation by paced breathing. The properties of local microcirculation including the blood transit times through capillaries and venules are extracted by fitting to Simplified Hemodynamics Model. Our preliminary results suggest that the real time SSMD-SFDI platform may serve as one effective imaging modality for microcirculation monitoring.

We report the study of lung squamous cell carcinoma diagnosis using the TI-DIC microscopy and the scattering-phase theorem. The spatially resolved optical properties of tissue are computed from the 2D phase map via the scattering-phase theorem. The scattering coefficient, the reduced scattering coefficient, and the anisotropy factor are all found to increase with the grade of lung cancer. The retrieved optical parameters are shown to distinguish cancer cases from the normal cases with high accuracy. This label-free microscopic approach applicable to fresh tissues may be promising for in situ rapid cancer diagnosis.