Xiaolei Qu

Novel automatic first-arrival picking method for ultrasound sound-speed tomography

Ultrasound sound-speed tomography (USST) is a promising technique for breast cancer diagnosis that is currently under investigation. Compared with two-dimensional X-ray mammography, it not only provides three-dimensional images but also avoids radiation exposure. However, the image quality of USST is highly dependent on the accuracy of travel time map (TTM). To improve the accuracy, a novel automatic first-arrival picking method is proposed in this study. With this method, Akaike information criterion is used to obtain travel time roughly, then cross-correlation of neighboring traces is employed to correct the obtained travel time. Simulation, phantom, and ex vivo experiments are implemented. The simulation experiments showed that the absolute errors of the proposed method were 52 and 98 ns for simple and complex structure data, respectively. The phantom and ex vivo experiments demonstrated the feasibility of the proposed method. In this study, a novel and robust first-arrival picking method was proposed for USST.

Phase aberration correction by multi-stencils fast marching method using sound speed image in ultrasound computed tomography

Reflection image from ultrasound computed tomography (USCT) system can be obtained by synthetic aperture technique, however its quality is decreased by phase aberration caused by inhomogeneous media. Therefore, phase aberration correction is important to improve image quality. In this study, multi-stencils fast marching method (MSFMM) is employed for phase correction. The MSFMM is an accurate and fast solution of Eikonal equation which considers the refraction. The proposed method includes two steps. First, the MSFMM is used to compute sound propagation time from each element to each image gird point using sound speed image of USCT. Second, synthetic aperture technique is employed to obtain reflection image using the computed propagation time. To evaluate the proposed method, both numerical simulation and phantom experiment were conducted. With regard to numerical simulation, both quantitative and qualitative comparisons between reflection images with and without phase aberration correction were given. In the quantitative comparison, the diameters of point spread function (PSF) in reflection images of a two layer structure were presented. In the qualitative comparison, reflection images of simple circle and complex breast modes with phase aberration correction show higher quality than that without the correction. In respect to phantom experiment, a piece of breast phantom with artificial glandular structure inside was scanned by a USCT prototype, and the artificial glandular structure is able to be visible more clearly in the reflection image with phase aberration correction than in that without the correction. In this study, a phase aberration correction method by the MSFMM are proposed for reflection image of the USCT.

Bent ray ultrasound tomography reconstruction using virtual receivers for reducing time cost

Bent ray ultrasound sound speed tomography reconstruction can improve image quality comparing to straight ray. However, it suffers from time consuming ray linking, which finds bent ray to link a pair of given emitter and receiver. Currently, multi ray tracing always be required for single ray linking, but all of traced rays will be discarded excepting one which links the given emitter and receiver. It is important for reducing time cost to avoid the discarding and decrease ray tracing number. For this purpose, a novel bent ray reconstruction method (BRRM) using virtual receiver was proposed in this study. Single reconstruction iteration of proposed method includes five steps. Firstly, travel time difference map (TTDM) is picked by first peak method. Secondly, launch angles for straight rays are obtained. Thirdly, ray tracing for each obtained launch angle is implemented and their arrival positions in transducer ring are recorded. Fourthly, TTDM for virtual receivers, which are placed in each bent ray arrival position, is estimated by interpolation of picked TTDM. Fifthly, simultaneous algebraic reconstruction technique (SART) is employed for reconstruction. To evaluated proposed method, ultrasound tomography RF data of simple and complex sound speed models are simulated by PZFlex. Reconstruction results show that proposed method can reduce ray tracing number to be about 20% and time cost to be one third of previous BRRM with similar image quality. In this study, a novel BRRM using virtual receiver is proposed to reduce ray tracing number and time cost of BRRM without image quality decreasing.

Robust contrast source inversion method with automatic choice rule of regularization parameters for ultrasound waveform tomography

We consider ultrasound waveform tomography using an ultrasound prototype equipped with the ring-array transducers. For this purpose, we use robust contrast source inversion (robust CSI), viz extended contrast source inversion, to reconstruct the sound-speed image from the wave-field data. The robust CSI method is implemented by the alternating minimization method. An automatic choice rule is employed into the alternating minimization method in order to heuristically determine a suitable regularization parameter while iterating. We prove the convergence of this algorithm. The numerical examples show that the robust CSI method with the automatic choice rule improves the spatial resolution of medical images and enhances the robustness, even when the wave-field data of a wavelength of 6.16 mm contaminated by 5% noise are used. The numerical results also show that the images reconstructed by the proposed method yield a spatial resolution of approximately half the wavelength that may be adequate for imaging a breast tumor at Stage I.

Limb muscle sound speed estimation by ultrasound computed tomography excluding receivers in bone shadow

Sarcopenia is the degenerative loss of skeletal muscle ability associated with aging. One reason is the increasing of adipose ratio of muscle, which can be estimated by the speed of sound (SOS), since SOSs of muscle and adipose are different (about 7%). For SOS imaging, the conventional bent-ray method iteratively finds ray paths and corrects SOS along them by travel-time. However, the iteration is difficult to converge for soft tissue with bone inside, because of large speed variation. In this study, the bent-ray method is modified to produce SOS images for limb muscle with bone inside. The modified method includes three steps. First, travel-time is picked up by a proposed Akaike Information Criterion (AIC) with energy term (AICE) method. The energy term is employed for detecting and abandoning the transmissive wave through bone (low energy wave). It results in failed reconstruction for bone, but makes iteration convergence and gives correct SOS for skeletal muscle. Second, ray paths are traced using Fermat’s principle. Finally, simultaneous algebraic reconstruction technique (SART) is employed to correct SOS along ray paths, but excluding paths with low energy wave which may pass through bone. The simulation evaluation was implemented by k-wave toolbox using a model of upper arm. As the result, SOS of muscle was 1572.0±7.3 m/s, closing to 1567.0 m/s in the model. For vivo evaluation, a ring transducer prototype was employed to scan the cross sections of lower arm and leg of a healthy volunteer. And the skeletal muscle SOSs were 1564.0±14.8 m/s and 1564.1±18.0 m/s, respectively.

A general method for cupping artifact correction of cone-beam breast computed tomography images

PurposeCone-beam breast computed tomography (CBBCT), a promising breast cancer diagnostic technique, has been under investigation for the past decade. However, owing to scattered radiation and beam hardening, CT numbers are not uniform on CBBCT images. This is known as cupping artifact, and it presents an obstacle for threshold-based volume segmentation. In this study, we proposed a general post-reconstruction method for cupping artifact correction.MethodsThere were four steps in the proposed method. First, three types of local region histogram peaks were calculated: adipose peaks with low CT numbers, glandular peaks with high CT numbers, and unidentified peaks. Second, a linear discriminant analysis classifier, which was trained by identified adipose and glandular peaks, was employed to identify the unidentified peaks as adipose or glandular peaks. Third, adipose background signal profile was fitted according to the adipose peaks using the least squares method. Finally, the adipose background signal profile was subtracted from original image to obtain cupping corrected imageResultsIn experimental study, standard deviation of adipose tissue CT numbers was obviously reduced and the CT numbers were more uniform after cupping correction by proposed method; in simulation study, root-mean-square errors were significantly reduced for both symmetric and asymmetric cupping artifacts, indicating that the proposed method was effective to both artifacts.ConclusionsA general method without a circularly symmetric assumption was proposed to correct cupping artifacts in CBBCT images for breast. It may be properly applied to images of real patient breasts with natural pendent geometry.

Improved highly accurate localized motion imaging for monitoring high-intensity focused ultrasound therapy

Visualizing an area subjected to high-intensity focused ultrasound (HIFU) therapy is necessary for controlling the amount of HIFU exposure. One of the promising monitoring methods is localized motion imaging (LMI), which estimates coagulation length by detecting the change in stiffness. In this study, we improved the accuracy of our previous LMI by dynamic cross-correlation window (DCCW) and maximum vibration amount (MVA) methods. The DCCW method was used to increase the accuracy of estimating vibration amplitude, and the MVA method was employed to increase signal–noise ratio of the decrease ratio at the coagulated area. The qualitative comparison of results indicated that the two proposed methods could suppress the effect of noise. Regarding the results of the quantitative comparison, coagulation length was estimated with higher accuracy by the improved LMI method, and the root-mean-square error (RMSE) was reduced from 2.51 to 1.69 mm.

An improved noise robust localized motion imaging for monitoring HIFU treatment

High intensity focused ultrasound (HIFU) is a minimally invasive treatment modality for cancerous tumors. The monitoring of HIFU treatment is extremely important. Localized motion imaging (LMI) is a HIFU monitoring method basing on Harmonic motion method (HMI). It estimates the coagulation length by detecting the change of tissue stiffness. However, the previous LMI is not robust to noise, thus two techniques are employed to improve its noise robustness in this study. First, dynamic cross correlation window (DCCW) is used to obtain vibration amplitude with noise suppression. Second, vibration-frequency band pass filter (VBPF) is employed to calculate the amount of vibration with expected frequency. The qualitative evaluation indicates that the two techniques are able to suppress the affection of noise on the vibration amplitude and vibration amount, respectively. Furthermore, the quantitative comparison demonstrates that the improved LMI is able to give higher accuracy of coagulation length estimation than the previous LMI, and it reduces the root mean square error (RMSE) from 2.52 to 1.84 mm. In this article, an improved noise robust LMI is proposed for monitoring HIFU treatment.

Temperature distribution analysis for high intensity focused ultrasound breast cancer treatment by numerical simulation

Breast cancer incidence is dramatically increasing in Japan and remaining high over the world. High intensity focused ultrasound (HIFU) is a low-invasive treatment method for breast cancer. It could produce irreversible tissue thermal coagulation in focus position. Direct reason for coagulation is temperature-rise. This study tried to numerically analyze temperature-rise distribution for HIFU breast cancer treatment. The breast model is constructed from MRI images of real patient, and a total of 4 patient cases were studied. Temporal study indicates that temperature rises with irradiation time, and possible unexpected burn could be anticipated outside of targeted focus due to heterogeneous breast tissue. Spatial observation reveals noticeable temperature-rise positions outside of focus. Temperature-rise ratio was calculated and quantitative influence of high temperature-rise positions outside of focus was shown. To avoid unexpected burns, focus control turns to be necessary.

Multi-frequency accelerating strategy for the contrast source inversion method of ultrasound waveform tomography using pulse data

In this work, we construct a multi-frequency accelerating strategy for the contrast source inversion (CSI) method using pulse data in the time domain. CSI is a frequency-domain inversion method for ultrasound waveform tomography that does not require the forward solver through the process of reconstruction. Several prior researches show that the CSI method has a good performance of convergence and accuracy in the low-center-frequency situation. In contrast, utilizing the high-center-frequency data leads to a high-resolution reconstruction but slow convergence on large numbers of grid. Our objective is to take full advantage of all low frequency components from pulse data with the high-center-frequency data measured by the diagnostic device. First we process the raw data in the frequency domain. Then multi-frequency accelerating strategy helps restart CSI in the current frequency using the last iteration result obtained from the lower frequency component. The merit of multi- frequency accelerating strategy is that computational burden decreases at the first few iterations. Because the low frequency component of dataset computes on the coarse grid with assuming a fixed number of points per wavelength. In the numerical test, the pulse data were generated by the K-wave simulator and have been processed to meet the computation of the CSI method. We investigate the performance of the multi-frequency and single-frequency reconstructions and conclude that the multi-frequency accelerating strategy significantly enhances the quality of the reconstructed image and simultaneously reduces the average computational time for any iteration step.