Ahmed M. Mahmoud

Classification of voluntary cough sound and airflow patterns for detecting abnormal pulmonary function

BackgroundInvoluntary cough is a classic symptom of many respiratory diseases. The act of coughing serves a variety of functions such as clearing the airways in response to respiratory irritants or aspiration of foreign materials. It has been pointed out that a cough results in substantial stresses on the body which makes voluntary cough a useful tool in physical diagnosis.MethodsIn the present study, fifty-two normal subjects and sixty subjects with either obstructive or restrictive lung disorders were asked to perform three individual voluntary coughs. The objective of the study was to evaluate if the airflow and sound characteristics of a voluntary cough could be used to distinguish between normal subjects and subjects with lung disease. This was done by extracting a variety of features from both the cough airflow and acoustic characteristics and then using a classifier that applied a reconstruction algorithm based on principal component analysis.ResultsResults showed that the proposed method for analyzing voluntary coughs was capable of achieving an overall classification performance of 94% and 97% for identifying abnormal lung physiology in female and male subjects, respectively. An ROC analysis showed that the sensitivity and specificity of the cough parameter analysis methods were equal at 98% and 98% respectively, for the same groups of subjects.ConclusionA novel system for classifying coughs has been developed. This automated classification system is capable of accurately detecting abnormal lung function based on the combination of the airflow and acoustic properties of voluntary cough.

Noninvasive detection of lipids in atherosclerotic plaque using ultrasound thermal strain imaging: in vivo animal study

OBJECTIVES This study sought to examine the feasibility of in vivo detection of lipids in atherosclerotic plaque (AP) by ultrasound (US) thermal (or temporal) strain imaging (TSI). BACKGROUND Intraplaque lipid content is thought to contribute to plaque stability. Lipid exhibits a distinctive physical characteristic of temperature-dependent US speed compared with water-bearing tissues. As tissue temperature changes, US radiofrequency (RF) echoes shift in time of flight, which produces an apparent strain (thermal or temporal strain [TS]). METHODS US heating-imaging pulse sequences and transducers were designed and integrated into commercial US scanners for US-TSI of arterial segments. US-RF data were collected while gradually increasing tissue temperature. Phase-sensitive speckle tracking was applied to reconstruct TS maps coregistered to B-scans. Segments from injured atherosclerotic and uninjured nonatherosclerotic common femoral arteries (CFA) in cholesterol-fed New Zealand rabbits, and segments from control normal diet-fed rabbits (N =14) were scanned in vivo at different time points up to 12 weeks. RESULTS Lipid-rich atherosclerotic lesions exhibited distinct positive TS (+0.19 ± 0.08%) compared with that in nonatherosclerotic (-0.10 ± 0.13%) and control (-0.09 ± 0.09%) segments (p < 0.001). US-TSI enabled serial monitoring of lipids during atherosclerosis development. The coregistered set of morphological and compositional information of US-TSI showed good agreement with histology. CONCLUSIONS US-TSI successfully detected and longitudinally monitored lipid progression in atherosclerotic CFA. US-TSI of relatively superficial arteries may be a modality that could be integrated into a commercial US system for noninvasive lipid detection in AP.

Detecting hepatic steatosis using ultrasound-induced thermal strain imaging: an ex vivo animal study

Hepatic steatosis or fatty liver disease occurs when lipids accumulate within the liver and can lead to steatohepatitis, cirrhosis, liver cancer and eventual liver failure requiring liver transplant. Conventional brightness mode (B-mode) ultrasound (US) is the most common noninvasive diagnostic imaging modality used to diagnose hepatic steatosis in clinics. However, it is mostly subjective or requires a reference organ such as the kidney or spleen with which to compare. This comparison can be problematic when the reference organ is diseased or absent. The current work presents an alternative approach to noninvasively detecting liver fat content using US-induced thermal strain imaging (US-TSI). This technique is based on the difference in the change in the speed of sound as a function of temperature between water- and lipid-based tissues. US-TSI was conducted using two system configurations including a mid-frequency scanner with a single linear array transducer (5-14 MHz) for both imaging and heating and a high-frequency (13-24 MHz) small animal imaging system combined with a separate custom-designed US heating transducer array. Fatty livers (n = 10) with high fat content (45.6 ± 11.7%) from an obese mouse model and control livers (n = 10) with low fat content (4.8 ± 2.9%) from wild-type mice were embedded in gelatin. Then, US imaging was performed before and after US induced heating. Heating time periods of ∼ 3 s and ∼ 9.2 s were used for the mid-frequency imaging and high-frequency imaging systems, respectively, to induce temperature changes of approximately 1.5 °C. The apparent echo shifts that were induced as a result of sound speed change were estimated using 2D phase-sensitive speckle tracking. Following US-TSI, histology was performed to stain lipids and measure percentage fat in the mouse livers. Thermal strain measurements in fatty livers (-0.065 ± 0.079%) were significantly (p < 0.05) higher than those measured in control livers (-0.124 ± 0.037%). Using histology as a gold standard to classify mouse livers, US-TSI had a sensitivity and specificity of 70% and 90%, respectively. The area under the receiver operating characteristic curve was 0.775. This ex vivo study demonstrates the feasibility of using US-TSI to detect fatty livers and warrants further investigation of US-TSI as a diagnostic tool for hepatic steatosis.

In vivo vascular wall tissue characterization using a strain tensor measuring (STM) technique for flow-mediated vasodilation analyses

Endothelial dysfunction is considered to be a key factor in the development of atherosclerosis, and the measurement of flow-mediated vasodilation (FMD) in brachial and other conduit arteries has become a common method to assess the status of endothelial function in vivo. Based on the direct relationship between the FMD response and local shear stress on the conduit brachial artery endothelium, we hypothesize that measuring relevant changes in the brachial wall strain tensor would provide a non-invasive tool for assessing vascular mechanics during post-occlusion reactive hyperemia. Direct measurement of the wall strain tensor due to FMD has not yet been reported in the literature. In this work, a noninvasive direct ultrasound-based strain tensor measuring (STM) technique is presented to assess changes in the mechanical parameters of the vascular wall during post-occlusion reactive hyperemia and/or FMD, including local velocities and displacements, diameter change, local strain tensor and strain rates. The STM technique utilizes sequences of B-mode ultrasound images as its input with no extra hardware requirement, and its algorithm starts with segmenting a region of interest within the artery and providing the acquisition parameters. Then a block matching technique based on speckle tracking is employed to measure the frame-to-frame local velocities. Displacements, diameter change, local strain tensor and strain rates are then calculated by integrating or differentiating velocity components. The accuracy of the STM algorithm was assessed in vitro using phantom studies, where an average error of 7% was reported using different displacement ranging from 100 microm to 1000 microm. Furthermore, in vivo studies using human subjects were performed to test the STM algorithm during pre- and post-occlusion. Good correlations (|r| >0.5, P < 0.05) were found between the post-occlusion responses of diameter change and local wall strains. Results indicate the validity and versatility of the STM algorithm and describe how parameters other than the diameter change are sensitive to reactive hyperemia following occlusion. This work suggests that parameters such as local strains and strain rates within the arterial wall are promising metrics for the assessment of endothelial function, which can then be used for accurate assessment of atherosclerosis. In summary, this study describes a simple and computationally efficient algorithm that can be integrated with ultrasound machines for vascular research. Moreover, it suggests that monitoring the local strain and strain rates of the brachial artery wall can replace or augment the measurement of arterial diameter in FMD studies.

Flexible integration of high-imaging- resolution and high-power arrays for ultrasound-induced thermal strain imaging (US-TSI)

Ultrasound-induced thermal strain imaging (USTSI) for carotid artery plaque detection requires both high imaging resolution (<;100 μm) and sufficient US-induced heating to elevate the tissue temperature (~1°C to 3°C within 1 to 3 cardiac cycles) to produce a noticeable change in sound speed in the targeted tissues. Because the optimization of both imaging and heating in a monolithic array design is particularly expensive and inflexible, a new integrated approach is presented which utilizes independent ultrasound arrays to meet the requirements for this particular application. This work demonstrates a new approach in dual-array construction. A 3-D printed manifold was built to support both a high-resolution 20 MHz commercial imaging array and 6 custom heating elements operating in the 3.5 to 4 MHz range. For the application of US-TSI in carotid plaque characterization, the tissue target site is 20 to 30 mm deep, with a typical target volume of 2 mm (elevation) × 8 mm (azimuthal) × 5 mm (depth). The custom heating array performance was fully characterized for two design variants (flat and spherical apertures), and can easily deliver 30 W of total acoustic power to produce intensities greater than 15 W/cm2 in the tissue target region.

Noninvasive assessment of human jawbone using ultrasonic guided waves

The problem of detecting defects in jawbones is an important problem. Existing methods based on X-rays are invasive and constrain the achievable image quality. They also may carry known risks of cancer generation or may be limited in accurate diagnosis scope. This work is motivated by the lack of current imaging modalities to accurately predict the mechanical properties and defects in jawbone. Ultrasonic guided waves are sensitive to changes in microstructural properties and thus have been widely used for noninvasive material characterization. Using these waves may provide means for early diagnosis of marrow ischemic disorders via detecting focal osteoporotic marrow defect, chronic nonsuppurative osteomyelitis, and cavitations in the mandible (jawbone). Guided waves propagating along the mandibles may exhibit dispersion behavior that depends on material properties, geometry, and embedded cavities. In this work, we present the first study in the theoretical and experimental analysis of guided wave propagation in jawbone. Semianalytical, finite-element (SAFE) method is used to analyze dispersion behavior of guided waves propagating in human mandibles. The geometry of the cross section is obtained by segmenting the computed tomography (CT) images of the jawbone. The cross section of the mandible is divided in two regions representing the cortical and trabecular bones. Each region is modeled as a linear Hookean material. The material properties for both regions are adopted from the literature. The experimental setup for the guided waves experiment is described. The results from both numerical analysis and guided waves experiment exhibit variations in the group velocity of the first arrival signal and in the dispersion behavior of healthy and defected mandibles. These results shall provide a means to noninvasively characterize the jawbone and accurately assess the bone mechanical properties. Our study is not aimed at characterizing the bone density in human mandibles. Rather, it is aimed to assess bone mechanical properties and defects that cannot be diagnosed by X-ray or other imaging modalities. This work may pave the way to the development of inexpensive noninvasive devices to detect small defects in human mandibles.

A System for Recording High Fidelity Cough Sound and Airflow Characteristics

Cough is considered an early sign of many respiratory diseases. Recently, there has been increased interest in measuring, analyzing, and characterizing the acoustical properties of a cough. In most cases the main focus of those studies was to distinguish between involuntary coughs and ambient sounds over a specified time period. The objective of this study was to develop a system to measure high fidelity voluntary cough sounds to detect lung diseases. To further augment the analysis capability of the system, a non-invasive flow measurement was also incorporated into the design. One of the main design considerations was to increase the fidelity of the recorded sound characteristics by increasing the signal to noise ratio of cough sounds and to minimize acoustical reflections from the environment. To accomplish this goal, a system was designed with a mouthpiece connected to a cylindrical tube. A microphone was attached near the mouthpiece so that its diaphragm was tangent to the inner surface of the cylinder. A pneumotach at the end of the tube measured the airflow generated by the cough. The system was terminated with an exponential horn to minimize sound reflections. Custom software was developed to read, process, display, record, and analyze cough sound and airflow characteristics. The system was optimized by comparing acoustical reflections and total signal to background noise ratios across different designs. Cough measurements were also collected from volunteer subjects to assess the viability of the system. Results indicate that analysis of cough characteristics has the potential to detect lung disease.

Classification of Voluntary Cough Airflow Patterns for Prediction of Abnormal Spirometry

Measurement of partial expiratory flow-volume curves has become an important technique in diagnosing lung disease, particularly in children and in the elderly. The objective of this study was to investigate the feasibility of predicting abnormal spirometry using the partial flow-volume curve generated during a voluntary cough. Here, abnormal spirometry is defined as less than the lower limit of normal (LLN) predicted by standard reference equations. Cough airflow signals of 107 subjects (56 male, 51 female) were previously collected from patients performing spirometry in a pulmonary function clinic. A variety of features were extracted from the airflow signal. A support vector machine (SVM) classifier was developed to predict abnormal spirometry. Airflow signal features and SVM parameters were selected using a genetic algorithm. The ability of the classifier to distinguish between normal and abnormal spirometry based on cough flow was evaluated by comparing the classifiers decisions with the LLN for the given subjects spirometry, including forced expiratory volume in 1 s (FEV1), forced vital capacity (FVC), and their ratio (FEV1/FVC%). Findings indicated that it was possible to classify patients whose spirometry results were less than the LLN with an overall accuracy of 76% for FEV1, 65% for FVC, and 76% for the ratio FEV1/FVC%. Accuracies were determined by repeated double cross-validation. This study demonstrates the potential of using airflow measured during voluntary coughing to identify test subjects with abnormal spirometry.

A new gradient-based algorithm for edge detection in ultrasonic carotid artery images

In human clinical studies, digital B-Mode ultrasound images of carotid and femoral artery walls are used to measure Intima-Media Thickness (IMT). IMT represents the arterial intima-media complex and is a validated surrogate parameter for atherosclerosis and cardiovascular disease risk. Conventionally, IMT is obtained by tracing the ultrasound interfaces of the arterial far walls manually. The manual tracing, however, may be replaced by an automated approach in order to decrease image analysis variability and improve consistency and efficiency of the imaging laboratory. In this paper, we present and test a novel automated edge detection method which employs a multi-step gradient based algorithm. The new method principally uses intensity, intensity gradient, and interface continuity of pixels to determine the ultrasound interfaces. In our investigations, we used the far wall of the common carotid artery to test the proposed algorithm. As our results show, the novel algorithm greatly eliminates subjectivity associated with conventional manual tracing and semi-automated gradient methods that employ seed point selection. The new method can therefore have a great potential in atherosclerosis studies and clinical trials.

High-resolution vascular tissue characterization in mice using 55MHz ultrasound hybrid imaging.

Ultrasound and Duplex ultrasonography in particular are routinely used to diagnose cardiovascular disease (CVD), which is the leading cause of morbidity and mortality worldwide. However, these techniques may not be able to characterize vascular tissue compositional changes due to CVD. This work describes an ultrasound-based hybrid imaging technique that can be used for vascular tissue characterization and the diagnosis of atherosclerosis. Ultrasound radiofrequency (RF) data were acquired and processed in time, frequency, and wavelet domains to extract six parameters including time integrated backscatter (T(IB)), time variance (T(var)), time entropy (T(E)), frequency integrated backscatter (F(IB)), wavelet root mean square value (W(rms)), and wavelet integrated backscatter (W(IB)). Each parameter was used to reconstruct an image co-registered to morphological B-scan. The combined set of hybrid images were used to characterize vascular tissue in vitro and in vivo using three mouse models including control (C57BL/6), and atherosclerotic apolipoprotein E-knockout (APOE-KO) and APOE/A(1) adenosine receptor double knockout (DKO) mice. The technique was tested using high-frequency ultrasound including single-element (center frequency=55 MHz) and commercial array (center frequency=40 MHz) systems providing superior spatial resolutions of 24 μm and 40 μm, respectively. Atherosclerotic vascular lesions in the APOE-KO mouse exhibited the highest values (contrast) of -10.11±1.92 dB, -12.13±2.13 dB, -7.54±1.45 dB, -5.10±1.06 dB, -5.25±0.94 dB, and -10.23±2.12 dB in T(IB), T(var), T(E), F(IB), W(rms), W(IB) hybrid images (n=10, p<0.05), respectively. Control segments of normal vascular tissue showed the lowest values of -20.20±2.71 dB, -22.54±4.54 dB, -14.94±2.05 dB, -9.64±1.34 dB, -10.20±1.27 dB, and -19.36±3.24 dB in same hybrid images (n=6, p<0.05). Results from both histology and optical images showed good agreement with ultrasound findings within a maximum error of 3.6% in lesion estimation. This study demonstrated the feasibility of a high-resolution hybrid imaging technique to diagnose atherosclerosis and characterize plaque components in mouse. In the future, it can be easily implemented on commercial ultrasound systems and eventually translated into clinics as a screening tool for atherosclerosis and the assessment of vulnerable plaques.