•A machine learning based method to automatically segment 3D breast ultrasound image.•Might help assist clinical diagnose breast cancer in the future.•Segmentation results demonstrate good ...consistency with ground truth.
Breast cancer is the most commonly diagnosed cancer, which alone accounts for 30% all new cancer diagnoses for women, posing a threat to women’s health. Segmentation of breast ultrasound images into functional tissues can aid tumor localization, breast density measurement, and assessment of treatment response, which is important to the clinical diagnosis of breast cancer. However, manually segmenting the ultrasound images, which is skill and experience dependent, would lead to a subjective diagnosis; in addition, it is time-consuming for radiologists to review hundreds of clinical images. Therefore, automatic segmentation of breast ultrasound images into functional tissues has received attention in recent years, amidst the more numerous studies of detection and segmentation of masses. In this paper, we propose to use convolutional neural networks (CNNs) for segmenting breast ultrasound images into four major tissues: skin, fibroglandular tissue, mass, and fatty tissue, on three-dimensional (3D) breast ultrasound images. Quantitative metrics for evaluation of segmentation results including Accuracy, Precision, Recall, and F1measure, all reached over 80%, which indicates that the method proposed has the capacity to distinguish functional tissues in breast ultrasound images. Another metric called the Jaccard similarity index (JSI) yields an 85.1% value, outperforming our previous study using the watershed algorithm with 74.54% JSI value. Thus, our proposed method might have the potential to provide the segmentations necessary to assist the clinical diagnosis of breast cancer and improve imaging in other modes in medical ultrasound.
Delay and Sum (DAS) is one of the most common beamforming algorithms for photoacoustic imaging (PAI) reconstruction. Based on calculating beamformed signal with simple delaying and summing, DAS can ...function in a quick response and is quite suitable for real-time PAI. However, high sidelobes and intense artifacts may appear when using DAS due to summing with unnecessary data. In this paper, a beamforming algorithm called Multiple Delay and Sum with Enveloping (multi-DASE) is introduced to solve this problem. Compared to DAS, the multi-DASE algorithm calculates not only the initial value of the beamformed signal but also the complete N-shaped photoacoustic signal for each pixel. Through computer simulation, a phantom experiment and experiment on human finger joint, the multi-DASE algorithm is compared with other beamforming methods in removing artifacts by evaluating the quality of the reconstructed images. Furthermore, by rearranging the calculation sequences, the multi-DASE algorithm can be computing in parallel using GPU acceleration to meet the needs of real-time clinical application.
To investigate the use of photoacoustic (PA) spectrum analysis (PASA) to identify microstructural changes corresponding to fat accumulation in mouse livers ex vivo and in situ.
The laboratory animal ...protocol for this work was approved by the university committee on use and care of animals. Six mice with normal livers and six mice with fatty livers were examined ex vivo with a PA system at 1200 nm, and nine similar pairs of mice were examined at 532 nm. To explore the feasibility of this technique for future study in an in vivo mouse model, an additional pair of normal and fatty mouse livers was scanned in situ with an ultrasonographic (US) and PA dual-modality imaging system. The PA signals acquired were analyzed by using the proposed PASA method. Results of the groups were compared by using the Student t test.
Prominent differences between the PASA parameters from the fatty and normal mouse livers were observed. The analysis of the PASA parameters from six normal and six fatty mouse livers indicates that there are differences of up to 5 standard deviations between the PASA parameters of the normal livers and those of the fatty livers at 1200 nm; for parameters from nine normal and nine fatty mouse livers at 532 nm, the differences were approximately 2 standard deviations (P < .05) for each PASA parameter.
The results supported our hypothesis that the PASA allows quantitative identification of the microstructural changes that differentiate normal from fatty livers. Compared with that at 532 nm, PASA at 1200 nm is more reliable for fatty liver diagnosis. Online supplemental material is available for this article.
•The proposed method segments ultrasound images into major tissue components.•Our method eliminates differences and inconsistencies in ultrasound interpretation.•Comparison between our method and ...manual segmentation demonstrated good consistency.
Segmentation of an ultrasound image into functional tissues is of great importance to clinical diagnosis of breast cancer. However, many studies are found to segment only the mass of interest and not all major tissues. Differences and inconsistencies in ultrasound interpretation call for an automated segmentation method to make results operator-independent. Furthermore, manual segmentation of entire three-dimensional (3D) ultrasound volumes is time-consuming, resource-intensive, and clinically impractical. Here, we propose an automated algorithm to segment 3D ultrasound volumes into three major tissue types: cyst/mass, fatty tissue, and fibro-glandular tissue. To test its efficacy and consistency, the proposed automated method was employed on a database of 21 cases of whole breast ultrasound. Experimental results show that our proposed method not only distinguishes fat and non-fat tissues correctly, but performs well in classifying cyst/mass. Comparison of density assessment between the automated method and manual segmentation demonstrates good consistency with an accuracy of 85.7%. Quantitative comparison of corresponding tissue volumes, which uses overlap ratio, gives an average similarity of 74.54%, consistent with values seen in MRI brain segmentations. Thus, our proposed method exhibits great potential as an automated approach to segment 3D whole breast ultrasound volumes into functionally distinct tissues that may help to correct ultrasound speed of sound aberrations and assist in density based prognosis of breast cancer.
The role of inertial cavitation in acoustic droplet vaporization Fabiilli, M.L.; Haworth, K.J.; Fakhri, N.H. ...
IEEE transactions on ultrasonics, ferroelectrics and frequency control/IEEE transactions on ultrasonics, ferroelectrics, and frequency control,
05/2009, Letnik:
56, Številka:
5
Journal Article
Recenzirano
Odprti dostop
The vaporization of a superheated droplet emulsion into gas bubbles using ultrasound-termed acoustic droplet vaporization (ADV)-has potential therapeutic applications in embolotherapy and drug ...delivery. The optimization of ADV for therapeutic applications can be enhanced by understanding the physical mechanisms underlying ADV, which are currently not clearly elucidated. Acoustic cavitation is one possible mechanism. This paper investigates the relationship between ADV and inertial cavitation (IC) thresholds (measured as peak rarefactional pressures) by studying parameters that are known to influence the IC threshold. These parameters include bulk fluid properties such as gas saturation, temperature, viscosity, and surface tension; droplet parameters such as degree of superheat, surfactant type, and size; and acoustic properties such as pulse repetition frequency and pulse width. In all cases the ADV threshold occurred at a lower rarefactional pressure than the IC threshold, indicating that the phase transition occurs before IC events. The viscosity and temperature of the bulk fluid are shown to influence both thresholds directly and inversely, respectively. An inverse trend is observed between threshold and diameter for droplets in the 1 to 2.5 mum range. Based on a choice of experimental parameters, it is possible to achieve ADV with or without IC.
Background
As of 2022, breast cancer continues to be the most diagnosed cancer worldwide. This problem persists within the United States as well, as the American Cancer Society has reported that ...∼12.5% of women will be diagnosed with invasive breast cancer over the course of their lifetime. Therefore, a clinical need continues to exist to address this disease from a treatment and therapeutic perspective. Current treatments for breast cancer and cancers more broadly include surgery, radiation, and chemotherapy. Adjuncts to these methods have been developed to improve the clinical outcomes for patients. One such adjunctive treatment is mild hyperthermia therapy (MHTh), which has been shown to be successful in the treatment of cancers by increasing effectiveness and reduced dosage requirements for radiation and chemotherapies. MHTh‐assisted treatments can be performed with invasive thermal devices, noninvasive microwave induction, heating and recirculation of extracted patient blood, or whole‐body hyperthermia with hot blankets.
Purpose
One common method for inducing MHTh is by using microwave for heat induction and magnetic resonance imaging for temperature monitoring. However, this leads to a complex, expensive, and inaccessible therapy platform. Therefore, in this work we aim to show the feasibility of a novel all‐acoustic MHTh system that uses focused ultrasound (US) to induce heating while also using US tomography (UST) to provide temperature estimates. Changes in sound speed (SS) have been shown to be strongly correlated with temperature changes and can therefore be used to indirectly monitor heating throughout the therapy. Additionally, these SS estimates allow for heterogeneous SS‐corrected phase delays when heating complex and heterogeneous tissue structures.
Methods
Feasibility to induce localized heat in tissue was investigated in silico with a simulated breast model, including an embedded tumor using continuous wave US. Here, both heterogenous acoustic and thermal properties were modeled in addition to blood perfusion. We further demonstrate, with ex vivo tissue phantoms, the feasibility of using ring‐based UST to monitor temperature by tracking changes in SS. Two phantoms (lamb tissue and human abdominal fat) with latex tubes containing varied temperature flowing water were imaged. The measured SS of the water at each temperature were compared against values that are reported in literature.
Results
Results from ex vivo tissue studies indicate successful tracking of temperature under various phantom configurations and ranges of water temperature. The results of in silico studies show that the proposed system can heat an acoustically and thermally heterogenous breast model to the clinically relevant temperature of 42°C while accounting for a reasonable time needed to image the current cross section (200 ms). Further, we have performed an initial in silico study demonstrating the feasibility of adjusting the transmit waveform frequency to modify the effective heating height at the focused region. Lastly, we have shown in a simpler 2D breast model that MHTh level temperatures can be maintained by adjusting the transmit pressure intensity of the US ring.
Conclusions
This work has demonstrated the feasibility of using a 256‐element ring array transducer for temperature monitoring; however, future work will investigate minimizing the difference between measured SS and the values shown in literature. A hypothesis attributes this bias to potential volumetric average artifacts from the ray‐based SS inversion algorithm that was used, and that moving to a waveform‐based SS inversion algorithm will greatly improve the SS estimates. Additionally, we have shown that an all‐acoustic MHTh system is feasible via in silico studies. These studies have indicated that the proposed system can heat a tumor within a heterogenous breast model to 42°C within a narrow time frame. This holds great promise for increasing the accessibility and reducing the complexity of a future all‐acoustic MHTh system.
Pure optical photoacoustic microscopy Xie, Zhixing; Chen, Sung-Liang; Ling, Tao ...
Optics express,
05/2011, Letnik:
19, Številka:
10
Journal Article
Recenzirano
Odprti dostop
The concept of pure optical photoacoustic microscopy(POPAM) was proposed based on optical rastering of a focused excitation beam and optically sensing the photoacoustic signal using a microring ...resonator fabricated by a nanoimprinting technique. After the refinements of the microring's working wavelength and in the resonator structure and mold fabrication, an ultrahigh Q factor of 3.0×10(5) was achieved which provided high sensitivity with a noise equivalent detectable pressure(NEDP) value of 29 Pa. This NEDP is much lower than the hundreds of Pascals achieved with existing optical resonant structures such as etalons, fiber gratings and dielectric multilayer interference filters available for acoustic measurement. The featured high sensitivity allowed the microring resonator to detect the weak photoacoustic signals from micro- or submicroscale objects. The inherent superbroad bandwidth of the optical microring resonator combined with an optically focused scanning beam provided POPAM with high resolution in the axial as well as both lateral directions while the axial resolution of conventional photoacoustic microscopy (PAM) suffers from the limited bandwidth of PZT detectors. Furthermore, the broadband microring resonator showed similar sensitivity to that of our most sensitive PZT detector. The current POPAM system provides a lateral resolution of 5 μm and an axial resolution of 8 μm, comparable to that achieved by optical microscopy while presenting the unique contrast of optical absorption and functional information complementing other optical modalities. The 3D structure of microvasculature, including capillary networks, and even individual red blood cells have been discerned successfully in the proof-of-concept experiments on mouse bladders ex vivo and mouse ears in vivo. The potential of approximately GHz bandwidth of the microring resonator also might allow much higher resolution than shown here in microscopy of optical absorption and acoustic propagation properties at depths in unfrozen tissue specimens or thicker tissue sections, which is not now imageable with current optical or acoustic microscopes of comparable resolution.
Purpose:
A first‐arrival travel‐time sound speed algorithm presented by Tarantola Inverse Problem Theory and Methods for Model Parameter Estimation (SIAM, Philadelphia, PA, 2005) is adapted to the ...medical ultrasonics setting. Through specification of a covariance matrix for the object model, the algorithm allows for natural inclusion of physical a priori information of the object. The algorithm's ability to accurately and robustly reconstruct a complex sound speed distribution is demonstrated on simulation and experimental data using a limited aperture.
Methods:
The algorithm is first demonstrated generally in simulation with a numerical breast phantom imaged in different geometries. As this work is motivated by the authors' limited aperture dual sided ultrasound breast imaging system, experimental data are acquired with a Verasonics system with dual, 128 element, linear L7‐4 arrays. The transducers are automatically calibrated for usage in the eikonal forward model.A priori information such as knowledge of correlated regions within the object is obtained via segmentation of B‐mode images generated from synthetic aperture imaging.
Results:
As one illustration of the algorithm's facility for inclusion ofa priori information, physically grounded regularization is demonstrated in simulation. The algorithm's practicality is then demonstrated through experimental realization in limited aperture cases. Reconstructions of sound speed distributions of various complexity are improved through inclusion of a priori information. The sound speed maps are generally reconstructed with accuracy within a few m/s.
Conclusions:
This paper demonstrates the ability to form sound speed images using two opposed commercial linear arrays to mimic ultrasound image acquisition in the compressed mammographic geometry. The ability to create reasonably good speed of sound images in the compressed mammographic geometry allows images to be readily coregistered to tomosynthesis image volumes for breast cancer detection and characterization studies.
Ionizing radiation acoustic imaging (iRAI) allows online monitoring of radiation's interactions with tissues during radiation therapy, providing real-time, adaptive feedback for cancer treatments. We ...describe an iRAI volumetric imaging system that enables mapping of the three-dimensional (3D) radiation dose distribution in a complex clinical radiotherapy treatment. The method relies on a two-dimensional matrix array transducer and a matching multi-channel preamplifier board. The feasibility of imaging temporal 3D dose accumulation was first validated in a tissue-mimicking phantom. Next, semiquantitative iRAI relative dose measurements were verified in vivo in a rabbit model. Finally, real-time visualization of the 3D radiation dose delivered to a patient with liver metastases was accomplished with a clinical linear accelerator. These studies demonstrate the potential of iRAI to monitor and quantify the 3D radiation dose deposition during treatment, potentially improving radiotherapy treatment efficacy using real-time adaptive treatment.