New Publications are available for Patient diagnostic methods and instrumentation
http://dl-live.theiet.org
New Publications are available now online for this publication.
Please follow the links to view the publication.Epidural needle length measurement by video processing
http://dl-live.theiet.org/content/conferences/10.1049/cp.2012.0446
This paper presents a novel image processing algorithm to measure the length and depth of an epidural needle during insertion. A wireless camera is used which transmits video during insertion to a host computer. The computer contains the image processing algorithm to detect the visible needle in the image and measures the length. The measurement is done by HSV background removal, colour comparison and using RGB histograms to locate 10mm markings on the Tuohy needle shaft. The visible length is then subtracted from the known length of the needle to calculate the depth of the needle tip. The camera can be placed in the operating theatre up to one meter away from the needle insertion site. The purpose of measuring needle depth in real time is to precisely place the needle in the epidural space. (6 pages)Self-calibrated wireless sleep sensing system for brain injury diagnostics
http://dl-live.theiet.org/content/conferences/10.1049/cp.2012.0586
A wide range of conditions fall under the category of Mild Traumatic Brain Injury (mTBI) diagnosis. In this article, we focus on a new method during sleep to detect mTBI associated with neuro-cognitive impairment that is not apparent using standard neuro-imaging methods. A wireless pressure sensor system comprising a piezo-resistive flexible substrate paired with a microcontroller and a radio is designed and built to provide information relevant to mTBI detection. The collected information is then processed with a software program to remove noise and interference, and detect both sleep states and cardio-respiratory movements. (5 pages)Mass detection in digital mammograms using gabor filter bank
http://dl-live.theiet.org/content/conferences/10.1049/cp.2012.0465
Digital Mammograms are currently the most effective imaging modality for early detection of breast cancer but the number of false negatives and false positives is high. Mass is one type of breast lesion and the detection of masses is highly challenged problem. Almost all methods that have been proposed so far suffer from high number of false positives and false negatives. In this paper, a method for detecting true masses is presented, especially, for the reduction of false positives and false negatives. The key idea of the proposal is the use of Gabor filter banks for extracting the most representative and discriminative local spatial textural properties of masses that are present in mammograms at different orientations and scales. The system is evaluated on 512 (256 normal+256 true mass) regions of interests (ROIs) extracted from digital mammograms of DDSM database. We performed experiments with Gabor filter banks having different numbers of orientations and scales to find the best parameter setting. Using a powerful feature selection technique and support vector machines (SVM) with 10-fold cross validation, we report to achieve Az = 0.995±0.011, the area under ROC. Comparison with state-of-the-art techniques suggests that the proposed system outperforms similar methods, which are based on texture description, and the difference is statistically significant. (6 pages)A new mutual information based similarity measure for medical image registration
http://dl-live.theiet.org/content/conferences/10.1049/cp.2012.0424
Medical image registration (IR) is the systematic process of aligning spate images, often involving different modalities with common reference framework, so complementary information can be combined and compared. This paper presents a new similarity measure which uses Expectation Maximization for Principal Component Analysis allied with mutual information (EMPCA-MI) for medical IR. The new measure has been analysed on multimodal, three band magnetic resonance images (MRI) T1, T2 and PD weighted, in the presence of both intensity non-uniformities (INU) and noise. Both quantitative and qualitative experimental results clearly demonstrate both improved robustness and lower computational complexity of the new EMPCA-MI paradigm compared with existing MI-based similarity measures, for various MRI test datasets. (6 pages)Validating the neutrosophic approach of MRI denoising based on structural similarity
http://dl-live.theiet.org/content/conferences/10.1049/cp.2012.0419
This paper focuses on validating the proposed Neutrosophic Set (NS) approach of Magnetic Resonance Image (MRI) denoising based on structural similarity such as Structural Similarity Index (SSIM) and Quality Index based on Local Variance (QILV). The Neutrosophic Set approach of median filter is used to reduce the Rician noise in MR image. The experiments have conducted on real MR image with Rician noise added. The visual and the diagnostic quality of the denoised image is well preserved. The performance of this filter is compared with median filter and non local mean filter (NLM). (6 pages)Segmentation of sputum cell image for early lung cancer detection
http://dl-live.theiet.org/content/conferences/10.1049/cp.2012.0433
Lung cancer has been the largest cause of cancer deaths worldwide with an overall 5-year survival rate of only 15%. Its early detection significantly increases the chances of an effective treatment. To that end, computer-aided diagnosis system using images of sputum stained smears has been an attractive approach due to its practicality, low cost, and invasiveness. In this context, we present a framework for the detection and segmentation of sputum cells in sputum images using respectively, a Bayesian classification and mean shift segmentation. Our methods are validated and compared with an other competitive technique via a series of experimentation conducted with a data set of 88 images. (6 pages)MRI mammogram image classification using ID3 algorithm
http://dl-live.theiet.org/content/conferences/10.1049/cp.2012.0464
Breast cancer is one of the most common forms of cancer in women. In order to reduce the death rate , early detection of cancerous regions in mammogram images is needed. The existing system is not so accurate and it is time consuming. The Proposed system is mainly used for automatic segmentation of the mammogram images and classify them as benign,malignant or normal based on the decision tree ID3 algorithm. A hybrid method of data mining technique is used to predict the texture features which play a vital role in classification. The sensitivity, the specificity, positive prediction value and negative prediction value of the proposed algorithm accounts to 93.45% , 99.95%,94% and 98.5% which rates very high when compared to the existing algorithms. The size and the stages of the tumor is detected using the ellipsoid volume formula which is calculated over the segmented region. (5 pages)A thresholding approach for detection of sputum cell for lung cancer early diagnosis
http://dl-live.theiet.org/content/conferences/10.1049/cp.2012.0442
In this paper, we address the problem of detection and extraction of sputum cells that help in lung cancer early diagnosis. Our approach is based on thresholding classifier looking at the distribution of sputum pixels and non sputum pixels in RGB space for extracting the sputum cell from the raw sputum image. In this method the problem is viewed as a segmentation problem focusing on extraction of such sputum cells from the images whereby we want to partition the image into sputum cell regions including the nuclei, cytoplasm and the background that includes all the rest. These cells will be analyzed to check whether they are cancerous or not. In this study, we used a database of 100 sputum color images to test the thresholding classifier by comparing it with the ground truth data of extracted sputum cells and it has shown a better extraction result than previous work. Moreover, we computed a histogram for different color spaces (RGB, YCbCr, HSV, L*a*b* and XYZ) to find the best color space with low false detection rate. We used some performance criteria such as precision, specificity and accuracy to evaluate the improved thresholding classifier. (6 pages)Towards assisted living via probabilistic vital-sign monitoring in the home
http://dl-live.theiet.org/content/conferences/10.1049/ic.2011.0022
This paper describes the development of a reliable multi-sensor data fusion system for monitoring patient vital-signs in the home. Initial investigatory work has taken place using ambulatory hospital patients, in the Oxford Cancer Hospital. Our monitoring approach is based on a probabilistic model of normality learned from a data-set of vital signs acquired from a representative group of high-risk patients. Alerts are provided to carers whenever patient vital signs are deemed "abnormal" with respect to the model of normality. We show examples of how this approach correctly detects physiological deterioration in the target patient group, and describe future work in further validation of the technology in home monitoring applications. (6 pages)CT-based robust statistical shape modeling for forensic craniofacial reconstruction
http://dl-live.theiet.org/content/conferences/10.1049/ic.2011.0126
Estimating the facial outlook from an unidentified skull is a challenging task in forensic investigations. This paper presents the definition and implementation of a craniofacial model for computerized craniofacial reconstruction (CFR). The craniofacial model consists of a craniofacial template that is warped towards an unidentified target skull. The allowed transformations for this warping are statistically defined using a PCA-based transformation model, resulting in a linear combination of major modes of deformations. This work builds on previous work [1] in which a statistical model was constructed based on facial shape (represented as a dense set of points) variations and sparse soft tissue depths at 52 craniofacial landmarks. The main contribution of this work is the extension of the soft tissue depth measurements to a dense set of points derived from a database of head CT-images of 156 patients. Despite the limited amount of training data compared to the number of degrees of freedom, the reconstruction tests show good results for a larger part of the test data. Root mean squared error (RMSE) values between reconstruction results and ground truth data smaller than 4 mm over the total head and neck region are observed. (6 pages)Clustering performance analysis of FCM algorithm on iterative relaxed median filtered medical images
http://dl-live.theiet.org/content/conferences/10.1049/ic.2011.0069
Noise removal is a major concern in image processing particularly in medical imaging. In this paper, a novel noise removal technique called Iterative relaxed median filter (IRMF) has been proposed and the effect of noise removal, by means of median filtering, on Fuzzy C-Means Clustering (FCM) has been analysed. Noise removal is carried out by various median filtering methods such as standard median filter (SMF), adaptive median filter (AMF), hybrid median filter (HMF) & relaxed median filter (RMF) and the performance of these methods is compared with the proposed method.A robust segmentation algorithm for branch structure and its implementation
http://dl-live.theiet.org/content/conferences/10.1049/cp.2011.0298
Medical image segmentation is a serious challenge in medical image processing. The medical images have low contrast, the variability of the organizational characteristics, the ambiguity of the border of the tissues and the complexity of microstructures (e.g. blood vessels, nerves). These features restrict the segmentation of the branch structure in the medical images. This paper presents a robust medical segmentation algorithm that combines the active contour model and region growth segmentation method. General locations, given by the region growth segmentation method, act as the initial position of snake model for segmentation. In this way, we can get the branch structure in the abdominal CT, such as the abdominal aorta, the celiac trunk and mesenteric artery. The experimental result shows that the revised algorithm achieves a better practical effect through surface rendering from VTK. It can help the doctor diagnose the illness wisely and objectively. (5 pages)Local transparency technique for heart model from VTK
http://dl-live.theiet.org/content/conferences/10.1049/cp.2011.0863
In the surgical operation, the catheter can be used to collect the point clouds data and do the ablation operation. Then we can get the three-dimensional heart reconstruction model from point clouds using the surface reconstruction algorithm in visualization system. Usually, the heart model is displayed opaquely in the visualization window and light can not go through the object. So doctors could hardly see the location of the catheter inside the model in real-time system. In addition, sometimes doctors have to observe the internal structure and the interested region in the model for clinical diagnosis and analysis. Therefore, it is necessary for us to do the transparency to meet the doctor requirement. Transparency can be generally divided into overall transparency and local transparency. Sometimes it is difficult for doctor to recognize the relative position between the catheter and the model surface from the overall transparency. Allowing for this reason, in this paper, we not only introduce the overall transparency, but also propose two ways to deal with the interest region transparency. All the methods are implemented on VTK platform.Research on model correction based on scattered point cloud data surface reconstruction
http://dl-live.theiet.org/content/conferences/10.1049/cp.2011.0855
Surface reconstruction based on point cloud data has been extensively studied. In this paper, the original data has a lot of noise and without normal vector information, so we use Poisson surface reconstruction algorithm to create 3D heart model, which has good robustness for noisy and irregular point cloud data. Then correct the shape by removing or adding points from the model surface to perfect the heart model, and reconstruct the new model by Power Crust algorithm based on the correctional surface point cloud data with normal vector, which is quick, accurate and efficient and very suitable for fast model correction. The results not only accurately display the spatial relationship of the heart, but also can be discretionary scaling and rotation in 3D space. The visual 3D graphical structures of the reconstructed heart model can display diseases (cardiopathy & arrhythmia) and enable catheter navigation in real time, which can help doctors to detect and diagnose diseases effectively, and improve the accuracy and security of medical diagnosis.A de-noising method for heart sound signal using Otsu's threshold selection
http://dl-live.theiet.org/content/conferences/10.1049/cp.2011.0848
In this research, Otsu's method methods is quoted and applied to de-noise heart sounds so that unwanted noises can be separated from a combined set of noises extracted through an electronic stethoscope. The noise was analyzed by applying a random but suitable threshold section method, the unwanted noise was reduced and the useful sound was reconstructed. The initial result shows that the method applied is effective and the reconstructed signal appears to be providing a better quality of heart sound.Research of medical image reconstruction system based-on MAC OS
http://dl-live.theiet.org/content/conferences/10.1049/cp.2011.0315
Unlike other 3D image visualization system based on Windows, this paper takes full use of the image processing advantages on MAC OS, with the library of development of the platform VTK, to achieve the medical reconstruction system. In this paper, we will not only discuss the environment building of VTK on MAC OS, but also achieve some functions of medical image system. The stability and feasibility of the system are being verified by the simulation results. (4 pages)Unsupervised textural segmentation of SonoElastographic breast images
http://dl-live.theiet.org/content/conferences/10.1049/cp.2011.0462
Breast Cancer, the most common malignancy in women and the second leading cause of death for women all over the world. By earlier detection of cancer, the better treatment can be provided. SonoElastography, a new medical imaging technique reduces un necessary biopsies compared to mammography and conventional ultrasonography. The diagnosis and treatment of the cancer rely on segmentation of SonoElastographic images. Texture features are widely used in classification problems, i. e. mainly for diagnostic purposes where the Region Of Interest (ROI) is delineated manual ly. It has not yet been considered for SonoElastographic segmenta tion. SonoElastographic images of 15 patients taken using Siemens Acuson Antares are considered for experimentation. The images contain both benign and malignant tumors. From the experimental procedure it is proposed that the combination of texture features, Local Binary Pattern (LBP), Contrast and Variance are best suited for segmentation of SonoElastographic breast images. The images are first enhanced using sticks filter to remove noise, to improve contrast, and emphasize tumor boundary. Then extract the features to segment the breast image. The resultant images undergo some post-processing steps to remove the spurious spots. The segmented image is thinned to mark the tumor boundary. The results are then quantified with the help of an expert radiologist. The proposed work can be used for further diagnostic process, to decide if the segmented tumor is benign or malignant.Intelligent fussy system based dermoscopic image segmentation for melanoma detection
http://dl-live.theiet.org/content/conferences/10.1049/cp.2011.0461
Medical image segmentation is the most essential and crucial process in order to facilitate the characterization and visualization of the structure of interest in medical images. This paper explains the task of segmenting skin lesions in Dermoscopy images based on intelligent Fuzzy clustering techniques for the early diagnosis of Malignant Melanoma. The various Fuzzy clustering techniques used are Fuzzy C Means Algorithm (FCM), Possibilistic C Means Algorithm and Hierarchical C Means Algorithm. The segmented images are compared with the ground truth image using various parameters such as False Positive Error (FPE), False Negative Error (FNE) Coefficient of similarity, spatial overlap and their performance is evaluated.Epicardial contour segmentation by using level set with prior shape and region
http://dl-live.theiet.org/content/conferences/10.1049/cp.2011.0993
A segmentation model is proposed for the epicardial contour of the cardiac MRI image based on level set with prior shape and region. Considering the shape correlation of the left ventricle (LV) in the cardiac image sequences, the segmentation result of the current frame is referred as the prior shape of the next frame and the contraction and diastolic change area of LV is exploited as the prior region. The initial evolving curve is automatically located by calculating the posterior probability of the pixel belonging to the chamber of LV. The prior shape and region prevent the evolving curve from leaking off the outer contour and pull it to the accurate edges. Promising experimental results are obtained on the real cardiac sequences.Dispersion flattened nonlinear square photonic crystal fiber for dental OCT
http://dl-live.theiet.org/content/conferences/10.1049/cp.2011.0783
The objective of this paper is to generate supercontinuum (SC) spectrums for dental optical coherence tomography (OCT) using proposed highly nonlinear dispersion flattened photonic crystal fiber (HNDF-PCF) as a propagating media of picosecond pulse easily produced by less expensive commercially available laser sources. This proposed HNDF-PCF has a very flat dispersion at 1.31 μm wavelength and by doping germanium (Ge) inside the core, nonlinear coefficient of the proposed HNDF-PCF is increased as large as 58.7 W<sup xmlns="http://pub2web.metastore.ingenta.com/ns/">-1</sup>km<sup xmlns="http://pub2web.metastore.ingenta.com/ns/">-1</sup>. Simulation results show that the proposed HNDF-PCF offers efficient SC light for dental OCT at 1.31 μm wavelength. Coherent length of the generated SC light is found 9.1 μm and spatial resolutions in the depth direction for dental OCT are found 5.5 μm for enamel and 5.9 μm for dentin.Research on image filtering method to combine mathematics morphology with adaptive median filter
http://dl-live.theiet.org/content/conferences/10.1049/cp.2010.1152
As a imaging technology of no-traditional lamp-house irradiation ,ultrasound medical imaging with its many advantages, such as no harm to the human body, real-time, cheap and easy to use, is widely used in clinic. But the ultrasonic imaging speckle noise make it difficulties to distinguish between normal tissue and pathological tissue. According to the character of noise in the medical ultrasonic image , an new method of the medical ultrasonic imaging filter based on mathematics morphology and adaptive filtering is proposed after analysis of speckle noise and general filter, and an experiment is made to validate. The experimental method is as follows : Firstly the Rayleigh noise is imposed on the original image , and then the median filter and the adaptive median filter are used on the contaminated image. Secondly the morphological filter is used to improve image quality and enhance the contrast , after the adaptive median filter is used on the image, to retain more necessary details. Finally the three noise filtering methods are compared from the images denoise and evaluation . And the results indicate that the new method is superior to other ones.Efficient EEG compression using JPEG2000 with coefficient thresholding
http://dl-live.theiet.org/content/conferences/10.1049/cp.2010.0488
This paper outlines a scheme for compressing EEG signals based on the JPEG2000 image compression algorithm. Such a scheme could be used to compress signals in an ambulatory system, where low-power operation is important to conserve battery life; therefore, a high compression ratio is desirable to reduce the amount of data that needs to be transmitted. The JPEG2000 specification makes use of the wavelet transform, which can be efficiently implemented in embedded systems. In this research, the JPEG2000 standard was broken down to its core components and adapted for use on EEG signals with additional compression steps added. Variations on the compression architecture were tested to maximize compression ratio (CR) while minimizing reconstructed percentage root-mean-squared difference (PRD) and power requirements. Tests indicate that the algorithm performs well in efficiently compressing EEG data, without significant loss in signal fidelity.A low-power wireless ECG processing node and remote monitoring system
http://dl-live.theiet.org/content/conferences/10.1049/cp.2010.0486
The Electrocardiogram (ECG) is an important physiological measurement that is critical in the diagnosis of many cardiac disorders. In this paper, the design of a wireless system for ambulatory ECG acquisition and analysis is discussed. The wireless ECG node can operate in one of two configurations, where (i) raw ECG data is continuously transmitted to a base station, or (ii) the Heart Rate (HR) is calculated on the node, and only HR data is transmitted. The power requirements of the wireless node are also analysed. Furthermore, the system design includes a complementary base station node, which can be attached via USB to the patient's PC. The system back-end is also presented, whereby patient data is uploaded to a remote server, and can be accessed by the patient's healthcare team via a secure website.The method of CAD model creation of dental shape based on 3D-CT images in orthodontics
http://dl-live.theiet.org/content/conferences/10.1049/cp.2010.1292
The key difficulty of orthodontics based on CAD/CAM is how to obtain the dense, complete and ordered point cloud data of teeth. A method of obtaining the CAD model of dental shape rapidly based on 3D-CT(Three Dimensional Computed Tomography) is introduced. Firstly, the dental plaster model is scanned in 3D-CT systems and the precise gray images are obtained. The method about extracting the internal and external contour of the CT images is proposed, and some relative techniques are studied in detail during thinning and tracing the contour. In order to improve the accuracy of the contour coordinates, the moment-based contour tracing algorithm is introduced, and the sub-pixel accuracy is reached. Finally, the point cloud data are input to CAD platform to design the CAD model of the physical prototype. Compare with traditional RE(Reverse Engineering) method, this method has the virtue of fast speed, same resolution in each dimension and high accuracy.Live cell index sensing based on the reflection mode of tilted fiber tip with gold nanoparticles
http://dl-live.theiet.org/content/conferences/10.1049/cp.2010.1166
We presented a micrometer size tapered fiber tip for real-time refractive index measurement of live cell. The tip is cut with a specific tilted angle. Only the cutting surface of tip apex is coated with gold nanoparticles (GNPs) by selective bonding method. The evanescent wave at the apex will excite the localized surface plasmon resonance (LSPR) of GNPs. With enhanced scattering at the resonance wavelength, the tip reflection will experience a loss, which is analogous to attenuated total reflection (ATR). By measure the reflection beam, a wavelength dependent loss dip in the reflection spectrum or an intensity loss of a specific wavelength could be used to detect the refractive index of live cell.Remote detection of stress using hyperspectral imaging technique
http://dl-live.theiet.org/content/conferences/10.1049/ic.2009.0249
Emotional or physical stresses induce a surge of adrenaline in the blood stream under the command of the sympathetic nerve system, which, cannot be suppressed by training. The onset of this alleviated level of adrenaline triggers a number of physiological chain reactions in the body, such as dilation of pupil and an increased feed of blood to muscles etc. This paper reports for the first time how Electro-Optics (EO) technologies such as hyperspectral and thermal imaging methods can be used for the detection of stress remotely. Preliminary result using hyperspectral imaging technique has shown a positive identification of stress through an elevation of haemoglobin oxygenation saturation level in the facial region, and the effect is seen more prominently for the physical stressor than the emotional one. However, all results presented so far in this work have been interpreted together with the base line information as the reference point, and that really has limited the overall usefulness of the developing technology. The present result has highlighted this drawback and it prompts for the need of a quantitative assessment of the oxygenation saturation and to correlate it directly with the stress level as the top priority of the next stage of research. (6 pages)Plastic electronics for point-of-care diagnostics
http://dl-live.theiet.org/content/conferences/10.1049/ic.2009.0064
This presentation offers a solution for a low-cost, point-of-care tests that provide lab quality results on the spot. The merging of lab-on-a-chip technology and organic semiconductors can provide better, faster, cheaper diagnosis and treatment, hence reducing burden on clinician time. (19 pages)Design of smart refrigerator for ubiquitous healthcare
http://dl-live.theiet.org/content/conferences/10.1049/ic.2009.0046
This paper addresses the design and system performance of smart refrigerator for health care. Some desired body data can be measured by the handle sensor and foot mat sensor in front of the refrigerator without extra health sensor on the body. Especially, the refrigerator can provide healthcare functions: 1) basic body diagnosis and monitoring, 2) healthcare and report 3) provision of well-being menu or healthy menu and 4) search and recommendation of restaurant information, etc. The refrigerator furnished in all home could play important roles including both healthcare and personal nutritionist. Health or well-being meals after checking health state can be recommended to help improvement of dietary or health life. In this paper, we like to propose the design of the smart refrigerator structure and function, which can be implemented by body data measuring sensors and Internet access. Also, we will show the function realization in the above proposed structure. (4 pages)Estimation of tissue elasticity by image processing of simulated B-mode ultrasound images
http://dl-live.theiet.org/content/conferences/10.1049/cp.2009.1682
Elastography and strain imaging often use ultrasound to measure mechanical properties of soft tissues. These techniques generally examine radiofrequency signals from an ultrasound scanner. This study investigates the feasibility of strain estimation directly from an ultrasound B-mode image, using segmentation and shape analysis. Several thousand computer generated tissue mimicking phantoms with stiff inclusions were produced and analysed, evaluating the change in shape when the phantom is subjected to 1-D compression in order to estimate strain. The resulting stiffness measurements are accurate to within 8% of the actual values. (6 pages)Virtual colonscopy system based on VTK and ITK
http://dl-live.theiet.org/content/conferences/10.1049/cp.2009.2036
Virtual colonscopy employs many techniques, such as 3D rendering, 3D segmentation, centreline extraction and navigation for computer-assistant diagnosis (CAD). Some CT images that are looked as two-dimensional slice images are computer-generated 3D reconstructed images simulating the optical colonoscopy views. When flying through in the simulative organ, the doctor could inspect the Ployp that attach at the inner wall of the organ. In this paper, we introduce our solutions to implement a typical virtual colonoscopy system based on VTK and ITK tools. Our work mainly focus on 3D rendering and centreline extraction. The result of our experiments reveals that the proposed virtual colonoscopy system is a helpful tool for doctors and researchers to examine the topology of patient's colon in diagnosis.A novel surface rendering algorithm for 3D reconstruction of medical images
http://dl-live.theiet.org/content/conferences/10.1049/cp.2009.1996
This paper implements a novel surface rendering algorithm for three-dimensional reconstruction of medical images. Traditional Marching Cubes (MC) algorithm shows limits for isosurface extraction by using surface configurations of a cube, because it costs much time to detect null cube when abstracting an isosurface. This has an important effect on the reconstruction speed. To overcome the problem, an improved MC algorithm based on image segmentation is proposed. It uses segmented images as input, aiming at reducing the time of detecting null cube and finding out those cubes crossed by the isosurface as soon as possible. Extensive experiments show that the efficiency of three-dimensional reconstruction can be improved a lot by using the novel algorithm.Segmentation of meningioma MR images using SVM and RBF algorithms
http://dl-live.theiet.org/content/conferences/10.1049/ic.2009.0152
The paper explores the field of medical image analysis through the use of ANN algorithms. The processing is carried out on MRI images which would be helpful in detecting brain tumor tissues through the revelation of infected tissues in the image. After the preprocessing of image, features are extracted from spatial domain and are separately fed to the algorithms. This process requires calculation of co-occurrence matrices of the template (size 3×3) from the image. The outputs obtained after these processes are subjected to segmentation technique. The comparison is done by determining the sensitivity and the specificity of the segmented image with reference to the manually segmented image. The algorithms implemented are Radial Basis Function (RBF) and Support Vector Machine (SVM). The important step in RBF is to determine centroids and width of the clusters. For this purpose, k-means clustering is used. The basis function used for RBF is Gaussian and for learning backpropagation is used. In SVM, kernel function used is linear function. (5 pages)Automatic segmentation of brain and tumor images using expectation maximization algorithm
http://dl-live.theiet.org/content/conferences/10.1049/ic.2009.0157
Combining image segmentation based on statistical classification with a geometric prior has been shown to significantly increase robustness and. reproducibility, Using a probabilistic geometric model of sought structures and image registration serves both initialization of probability density functions and definition of spatial constraints. A strong spatial prior, however, prevents segmentation of structures that are not part of the model. In practical applications, we encounter either the presentation of new objects that cannot be modeled with a spatial prior or regional intensity changes of existing structures not explained by the model. Our driving application is the segmentation of brain tissue and tumors from three-dimensional magnetic resonance imaging (MRI). Our goal is a high-quality segmentation of healthy tissue and a precise delineation of tumor boundaries. We present an extension to an existing expectation maximization (EM) segmentation algorithm that modifies a probabilistic brain atlas with an individual subject's information about tumor location obtained from subtraction of post- and pre-contrast MRI. The new method handles various types of pathology, space-occupying mass tumors and infiltrating changes like edema. Preliminary results on five cases presenting tumor types with very different characteristics demonstrate the potential of the new technique for clinical routine use for planning and monitoring in neurosurgery, radiation oncology, and radiology. (5 pages)A basic urine flowmeter and a non-invasive bladder pressure measurement device used in conjunction for the diagnosis of bladder outlet obstruction
http://dl-live.theiet.org/content/conferences/10.1049/ic_20080579
Benign enlargement of the prostate gland (and associated bladder outlet obstruction) is common in ageing men and can cause symptoms such as poor urine flow, urinating frequently and having to pass urine up at night. Many ageing men would benefit from treatment which is increasingly available in developing countries. Benefit from treatment, in particular surgery, is known to be more likely to improve if there is an objective diagnosis of outlet obstruction. Currently, the diagnosis of obstruction is a two-stage process. First of all, symptom assessment and a measurement of urine flow rate using specialised equipment are performed during a visit to a hospital clinic. Then, if a more definitive diagnosis is required, an invasive pressure-flow study, usually performed on a separate visit using specialised equipment and expert staff may be conducted. The test is expensive, uncomfortable and carries a risk of infection. We have been involved in the recent development of simpler alternatives to both of these measurement techniques. The first is a very simple disposable flowmeter which can be supplied to individuals and which allows serial, and by virtue of combining the measurements, more representative results. The second is an alternative to the pressure-flow study, a new device for obtaining bladder pressure measurements using a similar technique to noninvasive blood pressure measurements. We describe the principles underlying these techniques and propose how they may be introduced into a "low-tech" health care programme. (7 pages)Healthcare kiosk
http://dl-live.theiet.org/content/conferences/10.1049/ic_20080576
Taking healthcare services to the doorstep of people in emerging countries requires huge skilled workforce along with equipment for medical testing and monitoring, which requires massive investment. These countries are still at a stage where they cannot afford the involved costs; and yet the number of diseases is growing, especially with the western lifestyle and stress factor in society increasing. We propose a technology and business model for providing diagnostic health-care services in an efficient manner, which could in time become as popular and pervasive as ATMs are today for financial activities. We propose a fusion of technologies - vital signs monitoring (VSM) and Web based prescriptive diagnosis. VSM technology is maturing with many companies integrating health sensors, while web-based self help sites are becoming popular with sites like webmd.com, being used with high degree of confidence by patients and practitioners alike. In our offering, we propose a kiosk like structure, where the vital signs of a patient are measured on the front- end; and with the data collected, a backend database (offline or tele-based) gives on-the-spot diagnosis. The user who already has logged in (like an ATM) is given his summary report, with either a preliminary solution or recommended to nearest speciality hospitals, and his record is updated with the new diagnosis and recommendations. The underlying business model is based on both service and advertising mode. For the sensing devices, the cost is extracted from the customer (minimal), while for the kiosk and display setup, the hospitals or Websites which are further recommended can pay in the form of advertisement revenue. (7 pages)State-space modeling of neural signals
http://dl-live.theiet.org/content/conferences/10.1049/ic.2008.0680
Neural systems encode representations of biological signals in the firing patterns of their spike trains. Spike trains are point process time-series and their codes are both dynamic and stochastic. Although the signal is often continuous, its representation in the nervous systems is as a high-dimensional point process time-series. Because neural spike trains are point processes, standard signal processing techniques for continuous data are of limited utility in the analysis of neural systems. Accurate processing of neural signals requires the development of quantitative techniques to characterize correctly the point process nature of neural spiking activity. We discuss our research on the use of the state-space paradigm for point process observation to characterize neural systems and discuss three applications characterizing: the response threshold of neurons in primary auditory cortex; the dynamics of subthalamic nuclei neurons in patients suffering from Parkinson's disease; and how ensemble spiking activity of rat hippocampal neurons maintain a dynamic representation of the animal's position in its environment. Brain dynamics can also be measured indirectly at the scalp through the brain's electrical and magnetic fields. As a final example, we present a high-dimensional EM algorithm for a state-space model and use it to compute simultaneous state (dipole sources) and parameter estimates for magnetoencephalography (MEG) inverse problems. (60 pages)Blind source separation to extract foetal heart sounds from noisy abdominal phonograms: a single channel method
http://dl-live.theiet.org/content/conferences/10.1049/cp_20080431
A method for extracting foetal heart sounds (FHS) from noisy single channel abdominal phonograms is proposed. First, an appropriate matrix of delays is constructed; then multiple independent components are calculated using FastICA; finally, components are projected back onto the measurement space and those associated to FHS are subjectively selected. Three single channel phonograms, obtained from different subjects were analysed. Preliminary results are promising and showed successfully extractions of FHS (S1, S2). Future work will increase the number of subjects, evaluate the extraction quality, look for more information about foetal well-being, find an objective way to select FHS, and explore ICA implementations that utilise temporal structure such as Temporal Decorrelation source SEParation (TDSEP). (4 pages)Touching tooth segmentation from CT image sequences using coupled level set method
http://dl-live.theiet.org/content/conferences/10.1049/cp_20080343
We focus on the technique to construct and visualize the individual tooth model from CT image sequences for dental diagnosis and treatment. One challenging problem for dental CT images is that tooth crown regions are touching with other teeth in some slices. The common boundary for two adjacent teeth is missing. We propose a coupled variational level set method with region competition of two adjacent teeth to generate the virtual common boundary for them. Moreover the tooth region presents un-uniform intensity level and leads to the suspicious inner edges. We use the normal direction of the evolving contour and the gradient direction of the image to determine true or false outer boundaries and remove the suspicious inner edges. Finally we use the similar tooth shape feature from the CT image sequences to segment out all the tooth contours in different slices starting from the specified initial slice and initial contours. Both segmentation result in 2D images and visualization of each individual tooth model in 3D space verify our method.Multi-resolution rigid image registration using bacterial multiple colony chemotaxis
http://dl-live.theiet.org/content/conferences/10.1049/cp_20080371
A novel approach for image registration based on mutual information is presented in this paper. Firstly, normalized mutual information is introduced as the fitness function to supersede the traditional mutual information. Meanwhile, the extrema in fitness function resulted from interpolation is analyzed and the solution is presented as adding a weighted parameter into the variables set. Secondly, search model is improved. The 2nd generation wavelet is utilized to decompose the original image into multi-resolution sub-images to search optimal solution in different precision layer, which harmonizes the contradiction between search precision and convergence rate. Finally, we introduce a novel optimization means-bacterial multiple colony chemotaxis. Experiments demonstrate the legitimacy and efficiency of our betterment.Self organizing feature maps for the fiber tracking in diffusion tensor MR images
http://dl-live.theiet.org/content/conferences/10.1049/cp_20080466
Self-Organizing Maps (SOM) can be utilized for the purpose of discovering underlying structures in a distributed system. In the proposed study, SOM is employed to track neuronal pathways in brain Diffusion Tensor Magnetic Resonance (DTMR) images. DTMR Imaging (DTMRI) provides local directional information along the nerve fiber bundles in a vector form. The directional information is obtained by solving for the eigensystem of the tensors produced by the diffusion of the water molecules during the imaging process in MRI. The idea is to map a network of connected nodes of a SOM structure onto the investigated neuronal tracks. First, the imaging matrix is simulated with random eigenvectors and SOM is implemented and verified on these synthetic diffusivity tracks. Then the analysis follows on real DTMR images. The aim of the study is to propose an alternate method for the DTMRI tractography with SOM. Preliminary results on two dimensional synthetic diffusion images are encouraging to pursue the idea in 3D real images. In this study we discuss the proposed method in detail. (4 pages)Spectral analysis of backscattered ultrasound field from hydroxyapatite granules
http://dl-live.theiet.org/content/conferences/10.1049/cp_20080469
Microcalcifications (MCs) are of great importance in breast cancer early diagnosis and sometimes they do represent the unique evidence of cancer disease (up to 47%). Their visualization in ultrasound (US) is limited by a number of factors the most important of which are speckle and system limited spatial resolution. MCs are modelled as hydroxyapatite (HA) microgranules and sound scattering from HA microgranules can be solved by mean of Faran model. Agar based phantoms have been manufactured to simulate human soft tissues with cylindrical shape and a commercial echo-scanner (Technos Mpx di ESAOTE s.r.l., Genova, Italia) has been used to perform the scan of the phantom. This work concerns with phantoms and the analysis of US data for the visualization of HA microgranules under multiple angle scan. A reconstruction method was hence used to get a 2D map of reflectivity from the target and further Fourier analysis was applied to radio frequency (RF) data. (4 pages)New features for classification of cancerous masses in mammograms based on morphological dilation
http://dl-live.theiet.org/content/conferences/10.1049/cp_20080293
In the current research in the field of analysis of mammograms by means of image processing tools the task of malignancy and specularity assessment is investigated extensively. Early detection of malignant masses may significantly lower the risk of metastasis. It is known that malignancy is closely related to the shape of a mass (existence of spicules emanating from the center of the mass). Therefore the tasks of malignancy and specularity analysis are very often treated jointly. In this paper we introduce a new set of features useful in performing this task. For a contour of a cancerous mass a sequence of dilations is computed, the number of pixels on the outer contour of each dilation is counted, and this number is plotted against the size of the disk-shaped structuring element. Next, the linear trend is removed, and after denoising, the proposed features are calculated. The crucial point is that the proposed features are zero iff the input contour is circular and that all the features are invariant under translation, rotation, and scaling. These distinctive properties ensure successful classification irrespective to location, orientation and scale of the mass with the A<sub xmlns="http://pub2web.metastore.ingenta.com/ns/">z</sub> values of the ROC curve higher than for features given in the literature. The additional advantage of our approach is the relative simplicity of the proposed features. In contrast to many traditional features, no sophisticated algorithms are employed, so reimplementation of the new features is easy.Segmentation and analysis of the glomerular basement membrane using active contour models
http://dl-live.theiet.org/content/conferences/10.1049/cp_20080442
Some renal diseases are associated with significant alterations in the structure of the glomerular basement membranes (GBM). Increased thickness is commonly seen in diabetic nephropathy, where it may be an early sign of renal involvement. Abnormally thin GBMs are associated with the passing of blood in the urine, or hematuria. Measurement of the GBM thickness is carried out on images obtained from transmission electron microscopy (TEM). We propose image processing methods for the detection and measurement of the GBM. The methods include edge detection, morphological image processing, active contour modeling, skeletonization, and statistical analysis of the width of the GBM. The proposed methods were applied to 34 TEM images of six patients. The mean and standard deviation of the GBM width for a patient with normal GBM were estimated to be 348 ± 135 nm; those for a patient with thin GBMs associated with familial hematuria were 227 ± 94 nm; and those for a patient with diabetic nephropathy were 1152 ± 411 nm. Comparative analysis of the results of image processing with manual measurements by an experienced renal pathologist indicated low error in the range of 36 ± 11 nm. (4 pages)Nasopharyngeal carcinoma lesion extraction using clustering via semi-supervised metric learning with side-information
http://dl-live.theiet.org/content/conferences/10.1049/cp_20080373
In this paper, we consider the extraction of nasopharyngeal carcinoma lesion from magnetic resonance images as a clustering problem. The metric used by the clustering algorithm in our proposed method is a new spatially weighted metric, which is learned by semi-supervised metric learning with side-information. Several experiments have been conducted to compare the performance of the proposed metric with similar metrics for the tumor extraction.Accelerometer and footswitch evaluation of movement in three elderly patient groups
http://dl-live.theiet.org/content/conferences/10.1049/cp_20080647
We examined anterior-posterior (AP) and medio-lateral (ML) head and trunk movements during gait using accelerometers, and a footswitch evaluation of temporal parameters of gait during walking of two distinct elderly faller groups. Elderly fallers with a primary diagnosis of orthostatic hypotension, elderly fallers without a diagnosis of Orthostatic Hypotension and a control group of healthy elderly non-fallers were examined to determine if both sets of measures can be used to differentiate between the three groups. Using different parameters of AP and ML trunk movement and certain temporal parameters of gait we were able to significantly differentiate between the three groups to the same extent using both sets of measures. These results are exploratory and need to be confirmed in a definitive more high powered study.Motion compensated complementary coding for medical ultrasound
http://dl-live.theiet.org/content/conferences/10.1049/cp_20080646
Ultrasonic imaging using complementary coded pulses offers the SNR improvements of signal coding without the filter side-lobes introduced by single transmit codes. The effects of the transducer and motion in the medium, however, can introduce mismatch artefacts and high side-lobes due to misalignment. A method for filtering and motion compensation of complementary coded signals appropriate for use in medical imaging scenarios is presented in this paper. The method has been shown by simulation to reduce side-lobes to levels that compare favourably to systems using FM-coded signals of similar length and bandwidth while increasing coding gain and range resolution.Reconstruction of breast composition in a free space utilizing 2-D forward-backward time-stepping for breast cancer detection
http://dl-live.theiet.org/content/conferences/10.1049/cp_20080454
In this paper, we present Forward-Backward Time-Stepping (FBTS) technique to determine the presence and location of malignant tumours in the breast model. We demonstrate 2-D FBTS technique utilizing the numerical breast model in a free space instead of immersed in coupling liquid. Numerical simulation results show the FBTS algorithm has the potential to provide useful quantitative information of the breast's internal composition. (4 pages)Low frequency phase synchronisation analysis of MEG recordings from children with ADHD and controls using single channel ICA
http://dl-live.theiet.org/content/conferences/10.1049/cp_20080428
It has been suggested that the human brain is intrinsically organised into dynamic, anti-correlated functional networks. This paper presents a study on the so-called default mode network - which is active when the brain is apparently at rest - and on brain activity related to a given task. This work involves the analysis of low frequency magnetoencephalographic recordings of children with Attention Deficit Hyperactivity Disorder (ADHD) and controls performing both attentional as well as perceptual tasks. Single channel independent component analysis is used to isolate low frequency brain signals within the data in the presence of higher frequency brain activity and artifacts. Phase synchrony analysis is then carried out between the components of channels of interest to quantify any interaction between distant brain regions within the default-mode network. Preliminary results show variations in the phase locking between ADHD and controls, and indicate a corresponding change in phase synchrony between the corresponding brain regions at periods of rest and when tasks are being performed. (4 pages)Simulating and predicting blood glucose levels for improved diabetes healthcare
http://dl-live.theiet.org/content/conferences/10.1049/cp_20080433
Expert management of blood glucose levels (BGLs) can be difficult to achieve for many people with diabetes, since BGLs are affected in a very complex and non-linear manner by carbohydrate intake (diet), medication (tablets or insulin) and exercise. This paper discusses some of the expert management diabetes software systems that have been developed in the last decade which can accommodate the day- to-day complexities of BGL control, through a BGL prediction / therapy optimization strategy. In particular, the popular AIDA educational diabetes simulator is reviewed, together with a novel BGL simulation system based on artificial neural networks. (4 pages)Robust automated vertebra motion tracking from videofluoroscopy sequences
http://dl-live.theiet.org/content/conferences/10.1049/cp_20080441
This paper proposed a robust and reliable automated tracking technique using a particle filter. A constant velocity random walk model is adopted in the state transition model. The observation model is an amalgam of histogram matching from Markov random fields segmentation and gradient intensity measurements. The accuracy of tracking the vertebrae from a calibration model is of 1° and the variability among 5 initialisations is less than 0.5°. (4 pages)