[Show abstract][Hide abstract] ABSTRACT: In this paper, a new method for automatic vaginal bacteria cell segmentation and classification is proposed. Segmentation algorithm based on superpixel is first investigated to segment region of interest of the input image into cells. Feature extraction based on the segmented regions is trained by supervised deep learning method. Four types of different bacteria are studied for classification. Our experimental results show the classification result yields an accuracy of 99%, sensitivity of 100% and specificity of 98.04%. Compared to the state-of-the-arts method, better segmentation results have been achieved. Furthermore, our comparative analysis also shows that deep learning method outperforms traditional methods such as neural network and support vector machine.
Journal of Medical Imaging and Health Informatics 10/2014; 4(5). · 0.64 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Acquisition of the standard plane is crucial for medical ultrasound diagnosis. However, this process requires substantial experience and a thorough knowledge of human anatomy. Therefore it is very challenging for novices and even time consuming for experienced examiners. We proposed a hierarchical, supervised learning framework for automatically detecting the standard plane from consecutive 2-D ultrasound images. We tested this technique by developing a system that localizes the fetal abdominal standard plane from ultrasound video by detecting three key anatomical structures: the stomach bubble, umbilical vein and spine. We first proposed a novel radial component-based model to describe the geometric constraints of these key anatomical structures. We then introduced a novel selective search method which exploits the vessel probability algorithm to produce probable locations for the spine and umbilical vein. Next, using component classifiers trained by random forests, we detected the key anatomical structures at their probable locations within the regions constrained by the radial component-based model. Finally, a second-level classifier combined the results from the component detection to identify an ultrasound image as either a "fetal abdominal standard plane" or a "non- fetal abdominal standard plane." Experimental results on 223 fetal abdomen videos showed that the detection accuracy of our method was as high as 85.6% and significantly outperformed both the full abdomen and the separate anatomy detection methods without geometric constraints. The experimental results demonstrated that our system shows great promise for application to clinical practice.
[Show abstract][Hide abstract] ABSTRACT: Compared to 1D phased array probes with a fixed focus in elevation, multi-row arrays can significantly improved the slice thickness throughout image by expanding aperture and dynamic focusing in elevation. This paper describes the design and measurement of a 64 × 5 element 1.5D ultrasonic transducer that enables dynamic focusing and apodization in the elevation direction. We manufactured a transducer with an aperture size of 24 mm × 16 mm using a widely-used piezoelectric ceramic (PZT-5H) as the piezoelectric vibrator. The measured center frequency and -6 dB fractional bandwidth of the 1.5D transducer were 3 MHz and 79%, respectively. A two-way insertion loss of -58 dB was obtained at the average center frequency. We carried out a sound field simulation and measured the actual transmitting (one-way) sound field data by using a hydrophone. In this way the sound beam profile in elevation direction was obtained, showing the -6 dB slice thickness measured was about 2 mm throughout the whole range of interest (from near to far field). This provided a much greater focal depth and much better imaging resolution in the elevation direction than can be achieved by using 1D probe.
Sensors and Actuators A Physical 08/2014; · 1.84 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Ultrasonic elastography, a non-invasive technique for assessing the elasticity properties of tissues, has shown promising results for disease diagnosis. However, biological soft tissues are viscoelastic in nature. Shearwave dispersion ultrasound vibrometry (SDUV) can simultaneously measure the elasticity and viscosity of tissue using shear wave propagation speeds at different frequencies. In this paper, the viscoelasticity of rat livers was measured quantitatively by SDUV for normal (stage F0) and fibrotic livers (stage F2). Meanwhile, an independent validation study was presented in which SDUV results were compared with those derived from dynamic mechanical analysis (DMA), which is the only mechanical test that simultaneously assesses the viscoelastic properties of tissue. Shear wave speeds were measured at frequencies of 100, 200, 300 and 400Hz with SDUV and the storage moduli and loss moduli were measured at the frequency range of 1-40Hz with DMA. The Voigt viscoelastic model was used in the two methods. The mean elasticity and viscosity obtained by SDUV ranged from 0.84±0.13kPa (F0) to 1.85±0.30kPa (F2) and from 1.12±0.11Pas (F0) to 1.70±0.31Pas (F2), respectively. The mean elasticity and viscosity derived from DMA ranged from 0.62±0.09kPa (F0) to 1.70±0.84kPa (F2) and from 3.38±0.32Pas (F0) to 4.63±1.30Pas (F2), respectively. Both SDUV and DMA demonstrated that the elasticity of rat livers increased from stage F0 to F2, a finding which was consistent with previous literature. However, the elasticity measurements obtained by SDUV had smaller differences than those obtained by DMA, whereas the viscosities obtained by the two methods were obviously different. We suggest that the difference could be related to factors such as tissue microstructure, the frequency range, sample size and the rheological model employed. For future work we propose some improvements in the comparative tests between SDUV and DMA, such as enlarging the harmonic frequency range of the shear wave to highlight the role of viscosity, finding an appropriate rheological model to improve the accuracy of tissue viscoelasticity estimations.
[Show abstract][Hide abstract] ABSTRACT: Shear wave elstography based on acoustic radiation force is used for quantitative assessment of liver fibrosis in a rat model. The results show that the mean shear elasticity increases with the stage of liver fibrosis. The range of mean shear elasticity for all liver fibrosis stages is 1.25-3.17 kPa. The 95% confidence intervals of mean shear elasticity are overlapping for F1, F2, F3 and F4 stage. The results of ANOVA suggest that shear elasticity has significance difference between F0, F1 stage and F2, F3, F4 stage (P<;0.02), while shear elasticity has no significance between F0 and F1 stage (P=0.128), between F2, F3 and F4 stage (P>0.23). The AUC values of ROC curve of shear elasticity at METAVIR score threshold are 0.98 (≥F1), 0.95 (≥F2), 0.83(≥F3) and 0.83 (≥F4) respectively. The results suggest that shear wave elastography base on acoustic radiation force can be used potentially for early diagnosis and study of liver fibrosis.
2014 International Conference on Medical Biometrics (ICMB); 05/2014
[Show abstract][Hide abstract] ABSTRACT: Detection and recognition of standard plane automatically during the course of US examination is an effective method for diagnosis of fetal development. In this paper, an automatic algorithm is developed to address the issue of recognition of standard planes (i.e. axial, coronal and sagittal planes) in the fetal ultrasound (US) image. The dense sampling feature transform descriptor (DSIFT) with aggregating vector method (i.e. fish vector (FV)) is explored for feature extraction. The learning and recognition of the planes have been implemented by support vector machine (SVM) classifier. Experimental results on the collected data demonstrate that high recognition accuracy is obtained.
2014 IEEE 11th International Symposium on Biomedical Imaging (ISBI 2014); 04/2014
[Show abstract][Hide abstract] ABSTRACT: The acquisition of standard planes is crucial for medical ultrasound (US) diagnosis. In this paper, we present a hierarchical supervised learning framework for automatically detecting standard plane in consecutive 2D US images. The technique is demonstrated by developing a system that localizes fetal abdominal standard plane (FASP) from US videos. We first propose a novel radial component-based model (RCM) to describe the geometric constrains of key anatomical structures (KAS). In order to enhance the detection accuracy, we further adopt random forests classifier for detection of KAS within the regions constrained by RCM. Finally, a second-level classifier combines the results of component detectors to identify a US image as a “FASP” or a “nonFASP”. Experimental results show that our method significantly outperforms both the full abdomen and the separate anatomy detection methods without geometric constrains.
2014 IEEE 11th International Symposium on Biomedical Imaging (ISBI 2014); 04/2014
[Show abstract][Hide abstract] ABSTRACT: Currently, placental maturity staging is mainly based on subjective observation of the physician. To address this issue, a new method is proposed for automatic staging of placental maturity based on B-mode ultrasound images. Due to small variations in the placental images, dense descriptor is utilized in place of the sparse descriptor to boost performance. Dense sampled DAISY descriptor is investigated for the demonstrated scale and translation invariant properties. Moreover, the extracted dense features are encoded by vector locally aggregated descriptor (VLAD) for performance boosting. The experimental results demonstrate an accuracy of 0.874, a sensitivity of 0.996 and a specificity of 0.874 for placental maturity staging. The experimental results also show that the dense features outperform the sparse features.
Bio-medical materials and engineering 01/2014; 24(6):2821-9. · 1.09 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: The shear stress exerted on the cell membrane is an important factor in sonoporation. However, almost all previous calculations of shear stress were based on the Rooney's assumption, which is not applicable for the sonoporation experiments. In the article, to construct the microstreaming-shear stress model in sonoporation, it theoretically analyzed the microstreaming-shear stress exerted on the cell membrane generated by oscillating microbubble based on Nyborg's acoustic streaming theory. And the response of the model was compared with that of the sonoporation experiment. Cells were exposed by 1MHz 150kPa ultrasound in the presence of SonoVue® microbubbles. The sonoporated cells were labeled by fluorescent markers and detected by fluorescence microscopy and flow cytometry. The theoretically analyzed microstreaming-shear stress was in accordance with the cell experimental result. Although some minor factors are ignored when building the model to calculate the microstreaming-shear stress, the model was still reasonable.
Bio-medical materials and engineering 01/2014; 24(1):861-8. · 1.09 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Blood-Brain Barrier (BBB) can be opened locally, noninvasively and reversibly by low frequency focused ultra-sound (FUS) in the presence of microbubbles. In this study, Evans blue (EB) dye extravasation across BBB was enhanced by 1 MHz FUS at acoustic pressure of 0.35MPa in the presence of microbubbles at clinically comparable dosage. The spatial distribution of EB extravasation was visualized using fluorescence imaging method. The center region of BBB disruption area showed more enhanced fluorescence signal than the surrounding region in general. However, EB dye deposition was heterogeneous in the center region. The findings in this study indicated potential use of fluorescence imaging to evaluate large molecules delivery across BBB.
Bio-medical materials and engineering 01/2014; 24(6):2831-8. · 1.09 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: This paper describes a new method for shear wave velocity estimation that is capable of extruding outliers automatically without preset threshold. The proposed method is an adaptive random sample consensus (ARANDSAC) and the metric used here is finding the certain percentage of inliers according to the closest distance criterion. To evaluate the method, the simulation and phantom experiment results were compared using linear regression with all points (LRWAP) and radon sum transform (RS) method. The assessment reveals that the relative biases of mean estimation are 20.00%, 4.67% and 5.33% for LRWAP, ARANDSAC and RS respectively for simulation, 23.53%, 4.08% and 1.08% for phantom experiment. The results suggested that the proposed ARANDSAC algorithm is accurate in shear wave speed estimation.
Bio-medical materials and engineering 01/2014; 24(1):467-74. · 1.09 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Automation-assisted reading (AAR) techniques have the potential to reduce errors and increase productivity in cervical cancer screening. The sensitivity of AAR relies heavily on automated segmentation of abnormal cervical cells, which is handled poorly by current segmentation algorithms. In this paper, a global and local scheme based on graph cut approach is proposed to segment cervical cells in images with a mix of healthy and abnormal cells. For cytoplasm segmentation, the multi-way graph cut is performed globally on the a* channel enhanced image, which can be effective when the image histogram presents a non-bimodal distribution. For segmentation of nuclei, especially when they are abnormal, we propose to use graph cut adaptively and locally, which allows the combination of intensity, texture, boundary and region information. Two concave points-based approaches are integrated to split the touching-nuclei. As part of an ongoing clinical trial, preliminary validation results obtained from 21 cervical cell images with non-ideal imaging condition and pathology show that our segmentation method achieved 93% accuracy for cytoplasm, and 88.4% F-measure for abnormal nuclei, outperforming state of the art methods in terms of accuracy. Our method has the potential to improve the sensitivity of AAR in screening for cervical cancer.
Computerized medical imaging and graphics: the official journal of the Computerized Medical Imaging Society 01/2014; · 1.04 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Experimental MS(n) mass spectral libraries currently do not adequately cover chemical space. This limits the robust annotation of metabolites in metabolomics studies of complex biological samples. In-silico fragmentation libraries would improve the identification of compounds from experimental multi-stage fragmentation data when experimental reference data is unavailable. Here we present a freely-available software package to automatically control Mass Frontier software to construct in-silico mass spectral libraries, and to perform spectral matching. Based on two case studies we have demonstrated that HAMMER allows researchers to generate in-silico mass spectral libraries in an automated and high-throughput fashion with little or no human intervention required.
Documentation, examples, results and source code are available at http://www.biosciences-labs.bham.ac.uk/viant/hammer/ CONTACT: email@example.com.
[Show abstract][Hide abstract] ABSTRACT: Muscle thickness is one of the most widely used parameters for quantifying muscle function in both diagnosis and rehabilitation assessment. Ultrasound imaging has been frequently used to non-invasively study the thickness of human muscles as a reliable method. However, the measurement is traditionally conducted by manual digitization of reference points at the superior and inferior muscle fascias, thus it is subjective and time-consuming. In this paper, a novel method is proposed to detect the muscle thickness automatically. The superficial and deep fascias of a muscle are detected by line detection algorithm at the first ultrasound frame, and the image regions of interest (ROI) for the fascias are subsequently located and tracked by optical flow technique. The muscle thickness is geometrically obtained based on the location of the fascias for each frame. Six ultrasound sequences (250 frames in each sequence) are used to evaluate this method. The correlation coefficient of the detection results between the proposed method and manual method is 0.95 ± 0.01, and the difference is −0.05 ± 0.22 mm. The linear regression of the total 1500 detections show that a good linear correlation between the results of the two methods is obtained (R2 = 0.981). The automated method proposed here provides an accurate, high repeatable and efficient approach for estimating fascicle thickness during human motion, thus justifying its application in biological sciences.
Biomedical Signal Processing and Control 11/2013; 8(6):792–798. · 1.07 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Currently, most medical images are stored and exchanged with little or no security; hence it is important to provide protection for the intellectual property of these images in a secured environment. In this paper, a new and reversible watermarking method is proposed to address this security issue. Specifically, signature information and textual data are inserted into the original medical images based on recursive dither modulation (RDM) algorithm after wavelet transform and singular value decomposition (SVD). In addition, differential evolution (DE) is applied to design the quantization steps (QSs) optimally for controlling the strength of the watermark. Using these specially designed hybrid techniques, the proposed watermarking technique obtains good imperceptibility and high robustness. Experimental results indicate that the proposed method is not only highly competitive, but also outperforms the existing methods.
Expert Systems with Applications 11/2013; · 1.85 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: New vibration pulses are developed for shear wave generation in a tissue region with preferred spectral distributions for ultrasound vibrometry applications. The primary objective of this work is to increase the frequency range of detectable harmonics of the shear wave. The secondary objective is to reduce the required peak intensity of transmitted pulses that induce the vibrations and shear waves. Unlike the periodic binary vibration pulses, the new vibration pulses have multiple pulses in one fundamental period of the vibration. The pulses are generated from an orthogonal-frequency wave composed of several sinusoidal signals, the amplitudes of which increase with frequency to compensate for higher loss at higher frequency in tissues. The new method has been evaluated by studying the shear wave propagation in in vitro chicken and swine liver. The experimental results show that the new vibration pulses significantly increase tissue vibration with a reduced peak ultrasound intensity, compared with the binary vibration pulses.
IEEE Transactions on Ultrasonics Ferroelectrics and Frequency Control 11/2013; 60(11):2359-2370. · 1.82 Impact Factor