2nd IET International Conference on Biomedical Image and Signal Processing (ICBISP 2017)
Buy conference proceeding
- Location: Wuhan, China
- Conference date: 13-14 May 2017
- ISBN: 978-1-78561-318-0
- Conference number: CP718
- The following topics are dealt with: Biomedical signal processing; Biomedical imaging & image processing; Bioinstrumentation; biosensors; bio-micro/nano technologies; wearable & portable medical technologies/devices; Medical informatics & telemedicine; Cardiovascular & respiratory systems engineering; Neural engineering, neuromuscular systems; Rehabilitation engineering; Medical diagnostics and biomarkers; and, Healthcare information systems.
1 - 20 of 26 items found
-
An optical fibre based ex-vivo device for detection of cytokines
- Author(s): G.Z. Liu ; K.Z. Zhang ; A. Nadort ; M.R. Hutchinson ; E.M. Goldys
- + Show details - Hide details
-
p.
1 (4 .)
(1)
Cytokine detection is an essential topic for biomedical research. Herein we demonstrated an ex-vivo device based on a silica optical fibre for monitoring of locally variable cytokine IL-6 concentrations using a sandwich immunoassay scheme. The fibre was designed to be introduced into the perforated catheter with micrometre size holes drilled along its length to enable fluid exchange between the outside and inside of the catheter. An exposed optical fibre (diameter 125 μm) modified with a layer of gold nanoparticles was functionalized with the IL-6 capture antibody to form the sensing interface. The IL-6 detection antibody which was loaded on the fluorescently labelled magnetic nanoparticles, was used for reporting the analyte signal. This device was successfully applied to detect cytokine IL-6 with the detection limit of 1 pg mL-1 in the sample volume of 1 μL. It has the linear detection range of 1-400 pg mL-1 and is capable of detecting localised IL-6 secreted by BV2 cells following their liposaccharide stimulation. This universal biological detection system is suitable as a platform for the in-vivo devices able to monitor multiple health conditions.
-
Distributed functional connectivity impairment in schizophrenia: a multi-site study
- Author(s): Yong Yang ; Yue Cui ; Kaibing Xu ; Bing Liu ; Ming Song ; Jun Chen ; Huiling Wang ; Yunchun Chen ; Hua Guo ; Peng Li ; Lin Lu ; Luxian Lv ; Ping Wan ; Huaning Wang ; Hao Yan ; Jun Yan ; Hongxing Zhang ; Dai Zhang ; Tianzi Jiang
- + Show details - Hide details
-
p.
10 (6 .)
(1)
Schizophrenia has been considered as a dysconneciton syndrome, which means the disintegration, or over interaction between brain regions may underlie the pathophysiology of this disease. Noninvasive techniques like functional magnetic resonance imaging (fMRI) were utilized to test this hypothesis. However, there is no consensus on which brain areas and which functional network is related with it, mostly due to the small sample size of previous studies. Supervised machine learning techniques are able to examine fMRI connectivity data in a multivariate manner and extract features predictive of group membership. This technique requires large sample sizes and results from small sample study may not generalize well. By applying a multi-task classification framework to large size multi-site schizophrenia resting functional MRI (rsMRI) dataset, we were able to find consistent and robust features. We observed that schizophrenia patients had widespread deficits in the brain. The most informative and robustly selected functional connectivity (FC) features were between and within functional networks such as the default mode network (DMN), the fronto-parietal control network (FPN), the subcortical network, and the cingulo-opercular task control network (CON). Our finding validated the dysconnection hypothesis of schizophrenia and shed light on the details of the impaired functional connectivity.
-
Point-spread-function analysis for ultrasound computed tomography with ring array
- Author(s): Xiaoyue Fang ; Mingyue Ding ; Ming Yuchi
- + Show details - Hide details
-
p.
11 (5 .)
(1)
Breast cancer is the most common cancer women suffer worldwide. Ultrasound computed tomography is a promising technology for early breast cancer detection. The image quality of ultrasound system mainly depends on the beam properties of the transmitting array and the receiving array. In this paper, the point-spread-function (PSF) of ring array used in ultrasound computed tomography is derived in the polar coordinate and analyzed using the two-way response of continuous wave sound field to approximate the spatial pulse echo response of the system. Six ring arrays with different element numbers, i.e. spatial sampling, from 256 to 8192 and the fixed radius 10cm and the center frequency 2.5MHz were evaluated by the derived PSF. The corresponding beam properties were calculated, including the -6dB main lobe widths (Width_r_-6, Width_T_-6 defined for ring array firstly), the average side lobe level (ASLL) and the main side lobe energy ratio (MSR). The result shows that the spatial resolution is not influenced by the spatial sampling, while the contrast resolution can be enhanced with the increase of element number. However, when the element number is bigger than 2048, little contrast enhancement can be observed. This work provides an evaluation method for the ring array and can help design the array scheme.
-
Segmentation of dynamic contrast enhanced micro-CT images for fluorescence molecular tomography reconstruction
- Author(s): D.M. Yan ; W.H. Xie ; Z.H. Zhang ; Q.M. Luo ; X.Q. Yang
- + Show details - Hide details
-
p.
12 (5 .)
(1)
The hybrid imaging system combined fluorescence molecular tomography (FMT) and micro-computed tomography (microCT) has been widely used in preclinical researches, which can provide molecular and anatomical information simultaneously. FMT reconstruction is known as a highly illposed inverse problem. Fortunately, the anatomical structural priors from micro-CT could improve FMT reconstruction. In this paper, we propose a mouse segmentation scheme based on dynamic contrast enhanced (DCE) micro-CT images to build up a heterogeneous mouse model for reconstruction of FMT. DCE micro-CT images are collected after administration of non-ionic iodinated contrast agents. Based on the feature vectors consisting of signal intensities of different time points, the heart, liver, spleen, lung, and kidney are classified into different categories and extracted from separate category by morphological postprocessing. The segmentation method and its utilization in FMT reconstruction is demonstrated in an experiment in vivo. The results suggest that the proposed method can realize high accuracy segmentation of mouse anatomical structures, ultimately improve the reconstruction performance of FMT.
-
Dynamic synchronization state discrimination in local field potentials of neuropathic pain
- Author(s): Huichun Luo ; Xueying Du ; Yongzhi Huang ; Shouyan Wang
- + Show details - Hide details
-
p.
13 (6 .)
(1)
Local field potentials (LFPs) contain rich information of deep brain nucleus. And the dynamic multiple oscillations' synchronization activity of deep brain structures are involved in the neurophysiological and neuropathological function of a nucleus. In this study, a state identification approach was generated for identify the synchronization state of theta (6-9 Hz) and alpha (9-12 Hz) neural oscillations in neuropathic pain. The state identification approach consists oscillation extraction model and state discrimination model based on wavelet packet transform and adaptive thresholding strategy. The value of wavelet packet coefficients represented the synchronization state of theta and alpha oscillation. Then the coefficients were compared to the threshold to determine the state of oscillation according to the determination strategy. The parameters of this approach were optimized with 20 sides the sensory thalamus and periventricular gray/ periaqueductal gray (PVAG) LFPs of neuropathic pain and simulation signals. The best performance of theta oscillation state identification in -9 dB simulation signal achieved 85% sensitivity and 85% specificity. The best performance of alpha oscillation state identification in -11 dB simulation signal achieved 77% sensitivity and 85% specificity. Finally, the state identification approach was apply to identify theta and alpha oscillation state for neuropathic pain and got total 6 combination states. This study provides an approach to reliably discriminate the synchronization state of oscillations in neural signals. Based on this approach, a real-time monitoring of the pain neural state and an adaptive treatment regimen can be achieved.
-
A New Statistical-Based Algorithm for Medical Image Feature Extraction
- Author(s): Kuo-Kun Tseng ; Jiaqian Li ; Lantian Wang
- + Show details - Hide details
-
p.
14 (7 .)
(1)
In this paper, a new statistical-based algorithm, which applies the ideas of reduced binary pattern and multiple angle scheme, is proposed to extract image feature for identification or classification. The proposed statistical-based algorithm, multi-angle Reduced Binary Pattern (multi-angle RBP), is tested on the antinuclear antibody image database and sperm cell image database. The experimental results demonstrate that this algorithm is feasible for high accuracy, low complexity and fast processing for image classification in these two databases.
-
An acceleration based heuristics algorithm for gait phase detection
- Author(s): Yingying Wang ; Hui Zhou ; Yuanyuan Wang ; Hongli Guan ; Zhen Huang ; Guanglin Li
- + Show details - Hide details
-
p.
15 (6 .)
(1)
Detection of gait phase has multiple of applications such as lower limb prosthetics, drop foot stimulator, and rehabilitation assessment. Different types of wearable sensors such as force sensitive sensors (FSR), gyroscopes, and the combination of gyroscopes and accelerometers, have been used for the classification of gait phase. Among these sensors, since accelerometers are low in cost, easy to use, reliable, and lower power consumption, they have been widely used in some wearable devices. The accelerometer-based methods for gait phase classification have been also proposed in few previous studies. In this paper, we developed an acceleration-signal-based heuristics algorithm to detect the gait phase, which could divide the gait phase into loading response (LR), mid-stance (MS), terminal stance (TS), and swing phase (SW). In the proposed algorithm, the corresponding gait events of the four phases were detected using local inflection points and curve turning points from the filtered composite acceleration signal. The performance of the proposed algorithm was evaluated with ten healthy subjects walking on a level ground at their comfortable speed for 60 s. The preliminary results showed that the proposed algorithm is reliable and accurate in classifying the four gait phases (LR, MS, TS, and SW).
-
Ultrasound parallel delay multiply and sum beamforming algorithm based on GPU
- Author(s): Ting Su ; Ding Jie Yao ; Da Yu Li ; Shi Zhang
- + Show details - Hide details
-
p.
16 (4 .)
(1)
In 2015, Giulia Matrone proposed a nonlinear beamforming algorithm, Delay Multiply and Sum Beamforming (DMAS), which is superior to the Capon algorithm and Delay and Sum Beamforming (DAS) algorithm in resolution and contrast. The complexity of the DMAS algorithm is O(n2). In this paper, a parallel beamforming algorithm named PDMAS is proposed. The complexity of the PDMAS algorithm is O(n). In PMDAS algorithm, the channel operation is independent with each other and with good parallelism. At the same time, the software architecture of PDMAS algorithm is designed on GPU by using OpenCL programming language to test the performance of the proposed algorithm. The experiments are performed on the Precision T7810 workstation which contains two CPUs (Intel (R) Xeon (R) CPU E5-2620 v4) and AMD FirePro W7100 GPU respectively. The experimental result shows that both the PDMAS and the DMAS algorithm can't meet the real-time imaging requirements on the CPU. On the GPU, the frame rate of the DMAS algorithm is greatly improved, but still can't meet the real-time imaging requirements. On the other hand, the frame rate of PDMAS algorithm on the GPU can reach to 83fps, which means can fully meet the needs of real-time imaging. It can be concluded that the PDMAS algorithm has a better performance on GPU and fully meet the requirements of ultrasound real-time imaging systems.
-
Robust Medical Image Authentication using 2-D Stationary Wavelet Transform and Edge Detection
- Author(s): R. Singh ; P. Rawat ; P. Shukla
- + Show details - Hide details
-
p.
17 (8 .)
(1)
Watermarking is a data authentication technique which embeds the watermark data within the original image in order to claim for copyright protection. Watermark techniques must be design to be robust against the various attacks. In this paper an invisible robust watermarking method is designed using edge detection in stationary wavelet transform domain. Proposed method initially decomposes the input image in to wavelet sub-band coefficients using L level SWT. The edge coefficients are calculated using the standard Sobel edge detection mask. The watermark is embedded around the edges of the high frequency sub-band SWT coefficients which lead to better invisibility. The advantage of using the SWT is its robustness against shift variations. In addition scaled morphological dilation of the low frequency sub-band is used as key along with the SWT coefficients. This in turns improves the robustness and invisibility of the watermark embedding. Performance of proposed method is tested on the four distinct categories of medical image data sets including CT scan, MRI, X ray and Fingerprints. The parametric performance is evaluated by calculating the MSE and PSNR. It is observed that proposed method not only improves the invisibility of the watermark but is robust to various attacks like cropping, compression, and interpolation. Therefore method is useful to authenticate the patients imaging data using robust watermarking.
-
An automated wide-view imaging system of pathological tissue under optical microscopy
- Author(s): Shangbin Han ; Jimin Yang ; Honglin Wan ; Juan Yang
- + Show details - Hide details
-
p.
18 (6 .)
(1)
Optical microscopy imaging techniques have been found widespread applications in medicine because of its capability of identifying anatomic features of pathological tissue. For the purpose of diagnostic, the image should be obtained in a form of wide-view with ultra-high resolution which provide quantitative information about the tissue. However, due to the limitation of acquired pixel size and field of view in imaging camera, an overview of biological specimens may not be acquired directly using current devices. In this contribution, an automated imaging system of wide-view of optical microscopy of pathological tissue was presented. Our research also involves seeking the fast stitching algorithm to form a panorama of the whole pathological tissue. Digital pathological images were obtained using our new microscope system and these images were scanned in blocks. During image acquisition, overlapping regions were appeared between neighboring block images to avoid missing border details. The whole images acquired were then reconstructed as a panorama of the whole sample with our improved fast stitching methods. The feasibility and performance of our proposed method was compared with classical methods and modern keypoint detectors to validated in processing clinical pathological images and proved to be high efficient and accurate.
-
Normalization method of suppressing excitation noise in fluorescence molecular tomography
- Author(s): L.C. Lian ; Y. Deng ; W.H. Xie ; X.Q. Yang ; Q.M. Luo
- + Show details - Hide details
-
p.
19 (4 .)
(1)
Fluorescence molecular tomography (FMT) is a promising imaging technique to noninvasively resolve the three-dimensional spatial distributions of fluorescent targets associated with molecular and cellular functions. The reconstruction quality, including the localization and quantitative accuracy, are greatly improved by the use of normalized Born ratio method. However, the reconstruction quality is still far from reaching their potential of accurately visualizing biological information with a high localization and quantitative accuracy. It is because the inverse problem of FMT is severely ill-posed and is easily affected by the noise in the measurement data. Here we provided a scaling normalization method for FMT to restrain the influence of excitation noise on fluorescence molecular tomography reconstruction. In this method, we use the total intensity of the excitation data detected by CCD camera to normalize the fluorescence measurements. It's feasible to suppress excitation noise when compared with the normalized Born ration method. Furthermore, by applying singular-value decomposition to the forward matrices, we find that the method we proposed has more useful singular values than the normalized Born ratio method which means that it reduces the ill-posedness of FMT inverse problem. To examine its performance, we perform phantom experiments with a hybrid FMT-XCT system. To check the robustness of our method to excitation noise, we add 10% Gaussian white noise to the measured excitation data. The reconstruction results are compared with those of the normalized Born ratio method. The results indicate that our method has a better reconstruction results and a better noise robustness.
-
Understanding and assessing low-light cameras for super-resolution localization microscopy
- Author(s): Yujie Wang ; Zhaoning Zhang ; Mengting Li ; Luchang Li ; Tingwei Quan ; Zhen-Li Huang
- + Show details - Hide details
-
p.
2 (6 .)
(1)
Low-light camera is a special kind of detector which is capable of detecting weak signal down to a few photons or even single photon. Low-light camera is an essential element in super-resolution localization microscopy, because the latter requires detection of the weak fluorescence from single molecules. Currently, there are several types of widely-used low-light cameras which are commercially available in the market. For each type of low-light cameras, tens of products from different manufacturers are available with different performance and prices. Therefore, choosing an appropriate low-light camera for localization microscopy is always time-consuming and sometimes confusing. In this work, we will firstly discuss the principle of localization microscopy, and the performance requirements of low-light cameras in this new technique. Then, we will present a brief explanation on the key parameters for determining the imaging performance of low-light cameras, and explain how to assess the imaging performance of low-light cameras both experimentally and theoretically. Finally, we will show the imaging performance of a new type of low-light cameras called sCMOS in localization microscopy. We hope to convince the readers that EMCCD is not always the best and/or only camera choice in localization microscopy, and argue that the most appropriate choice of camera depends on the application scenario and necessitates a deeper look at cameras beyond typical specifications.
-
Fusion estimation of respiration rate from ECG and PPG signal based on Android platform and wearable watch
- Author(s): Zhengling He ; Xianxiang Chen ; Zhen Fang ; Tingyu Sheng ; Shanhong Xia
- + Show details - Hide details
-
p.
20 (6 .)
(1)
We have developed a health system based on Android platform and wearable watch to serve as a portable monitor and analyser for several physiological parameters. The raw photoplethysmogram (PPG) and electrocardiogram (ECG) waveforms were continuously collected on the wearable watch and transmitted to the Android phone/tablet by Bluetooth low energy (BLE) protocol. A robust transfer protocol was applied to reduce the impact of packet loss. The feature points of filtered ECG were extracted by Pan Tompkins QRS detection algorithm and that of PPG was extracted by a slope-based algorithm. A kurtosis-based algorithm was implemented to indirectly estimate the respiration rate (PR) through feature fusion without additional respiratory sensor node. All physiological parameters can be stored in the Android device and transmitted to remote server for further analysis. The preliminary experiment results demonstrated that the proposed portable system was suitable for daily monitoring of ECG, PPG and PR, and achieved the expected accuracy.
-
Design and implementation of a novel human-machine interactive healthcare system for visual reproduction test
- Author(s): Anran Qi ; Wanjin Li ; Anna Zheng ; Linmi Tao
- + Show details - Hide details
-
p.
21 (6 .)
(1)
Cognitive impairment is a noteworthy problem in the healthcare domain due to an aging population. The existing methods to assess mild cognitive impairment such as Visual Reproduction experiment is via paper and pen, and the results are read and diagnosed by doctors. This classic method only collects static characteristics of visual memory and loses details of human behavior in testing. In this paper, we design a novel interactive system coupled with a digital pen and paper. The system records the dynamics information of drawing behavior during the experiments without changing the consistence of tester experience. Thus these dynamic data can be used to show a correlation with the quality of test completion based on the previous Visual Reproduction research. Meanwhile, functions such as real-time recording and offline management highly improve the efficiency and accuracy of medical judgment. Additionally, new matrixes obtained by calculating dynamic data may be important consideration for future phenotype and human behavior association studies. This human-interactive system reveals the dynamic information which can assist speculation of quality of drawing scripts and also can benefit medical researchers on effective evaluation of the Visual Reproduction experiment results.
-
An improved method of calculating MTF from PSF based on CT phantom images
- Author(s): Libin Liang ; Pu Zhang ; Hui Ding ; Guangzhi Wang
- + Show details - Hide details
-
p.
22 (6 .)
(1)
High contrast resolution(HCR) is an important performance parameter of the computed tomography(CT) scanner. The assessment of HCR is regularly performed by measuring the modulation transfer function (MTF). Generally, MTF is calculated from point spread function (PSF) which can be approximated by the point source response in CT phantom images. In the traditional method of calculating MTF, a rectangle region of interest (ROI) is usually placed on the scanned image to crop the point source response. However, the sampling interval of phantom images obtained in practice is often coarse (about 0.40 to 0.50 mm), which means only a few pixels lie in the vicinity of the point source. Therefore, the size of ROI affects the accuracy of MTF when image noise exists. In this study, based on the coarsely sampled CT images of Catphan500 phantom, we analyze how the ROI size influenced the accuracy of MTF. Then the traditional method was improved by optimizing the ROI size. We found that when the ROI size was close to the size of point source response, the obtained MTF was more accurate. We verified the validity of improved method with Catphan500 images from five different medical CT scanners. The results showed that the cut-off frequency of obtained MTF had a good consistency with cut-off frequency evaluated by experts. The improved method is particularly useful for routine HCR assessment of the CT scanner.
-
MRI motion artifact mitigation methodology using spin echo pulse sequences on a 4.7 T scanner
- Author(s): A.R. Farias ; M.F.D. Monies ; H.A. Magalhaes ; E.M.A.M. Mendes
- + Show details - Hide details
-
p.
23 (5 .)
(1)
The movement caused by respiration is one of the most common sources of artifacts in Magnetic Resonance imaging (MRI). In this work, a methodology to mitigate artifacts related to respiratory movement is proposed using sequences of SE (Spin Echo) pulses in a 4.7 Tesla scanner for small animals without a commercial electronic module for detection of respiratory signals. In this methodology, the RF (Radio Frequency) pulses produced by the scanner in each sequence are detected synchronously with the respiration signal using the respiratory gating technique. To validate the methodology, images of a mechanism built to simulate the respiratory movement are collected and compared to the still images using objective metrics: PSNR (Peak Signal-to-noise-ratio), RMSE (Root Mean Square) and SSIM (Structural Similarity Index Measure). The results showed that the mitigation methodology of the simulated respiratory motion artifacts was efficient and can be used, in future experiments, to mitigate respiratory movement artifacts from abdominal organs of small animals, such as rats and mice.
-
Extraction of visual evoked potential using improved Wiener filter
- Author(s): Dacheng Liu ; Bing Sun ; Chunqi Chang ; Jianfeng Yang ; Jiajun Wang ; Nan Hu
- + Show details - Hide details
-
p.
24 (4 .)
(1)
Visual evoked potentials (VEPs) are electrical responses to visual stimuli aroused in the occipital region of the cerebral cortex. The P100 is an important component of VEP, and its latency is generally used for diagnosis of many clinical diseases. The existing methods to extract VEP have several drawbacks such as involving too many stimuli, costing too much time or requiring a large training database. However, the realtime applications such as brain-computer interfacing (BCI) require that the VEP should be quickly extracted from fewer stimuli, which is seldom met by those existing methods. In this paper, a VEP extraction method involving only a few visual stimuli is proposed based on improved Wiener filter. Through experiments on multiple subjects, it is verified that the proposed method can extract VEP quickly and give a reliable estimation of latency of P100.
-
Patents Analysis on Magnetic Resonance Imaging and Data Processing Technology
- Author(s): Qiuping Ding ; Ruyi Luo ; Qiqi Tong ; Hongjian He ; Jianhui Zhong
- + Show details - Hide details
-
p.
25 (5 .)
(1)
This paper analyzed the patents related to Magnetic Resonance Imaging and Data Processing Technology using the Innography patents analysis software. The clustering analysis method was used and the patents of the latest 5 years and those with high strength value (between 9 to 10) were mainly analyzed. We identified 10 hotpots of the latest MRI technologies. We also found that GPS (GE, Philips, Siemens) remain dominant in the MRI field. More companies and universities of China however began to apply for patents related to MRI. Through the above analysis, we identified the trend of development of magnetic resonance imaging and data processing technology, and provided clues for further MRI innovation.
-
Part-Wise Pedestrian Gender Recognition Via Deep Convolutional Neural Networks
- Author(s): M. Raza ; Chen Zonghai ; S.U. Rehman ; Ge Zhenhua ; Wang Jikai ; Bao Peng
- + Show details - Hide details
-
p.
26 (6 .)
(1)
Gender prediction is a challenging field covering automated pedestrian attributes analysis. Machine learning approaches are in use to closely predict correct gender. In order to improve the accuracy of prediction, a deep convolutional neural network is proposed to analyze the pedestrian gender. In the methodology, the pedestrian's images are parsed with the help of existing deep de-compositional neural network. The parsed images with removed background are then divided into full body and upper body images. Later, these two types of images are given as an input to the proposed fine-tuned convolutional neural network model. The full body input images are classified into frontal-view, back-view, and mixed view gender categories. Using upper body clothing, images are sub-classified into eight categories. The proposed approach proved to have better prediction performance with respect to different sub-classifications.
-
Thyroid nodule detection using attenuation value based on non-enhancement CT images
- Author(s): Yihong Chen ; Chenbin Liu ; Wenxian Peng ; Shunren Xia
- + Show details - Hide details
-
p.
3 (4 .)
(1)
Purpose: To validate the feasibility of thyroid nodule detection using attenuation value based on non-enhancement computed tomography (CT) images. Materials and Methods: One hundred and thirty-four transverse CT images with nodules from 58 inpatients and 128 normal images from 55 outpatients (healthy controls) were enrolled in this study. The inpatients with thyroid nodules (50 malignant, 84 benign) underwent nodule excision operation and final diagnoses were confirmed by histopathology. Thyroid regions of interest (ROIs) from axial CT images were delineated manually by a radiologist. The CT values of every thyroid pixels were extracted from the DICOM images. Median and average filter were applied to reduce image noise. Attenuation values of every 2*2 matrix were compared to the high and low density thresholding to identify the presence of the low density area such as cyst and necrosis or the high density area like calcification. The parameters, thresholding and filter type, were optimized according to accuracy and sensitivity. To evaluate the performance of the proposed method, accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were considered. Result: The experimental results demonstrate that our proposed method offers exceptional accuracy (ACC=0.8511), sensitivity (SEN=0.8060), specification (SPC=0.8984), positive predictive value (PPV=0.8926) and negative predictive value (NPV=0.8156) respectively. Conclusion: Our study provides a practical strategy for addressing thyroid nodule detection. The proposed and deployed thresholding optimization approach could serve as computer-aided diagnosis method in clinical application.
