arXiv subjects

Junjie Yao

Publications and source records attributed to Junjie Yao.

26 records · Page 2Linked to original sources

arXiv2021

Deep image prior for undersampling high-speed photoacoustic microscopy

Photoacoustic microscopy (PAM) is an emerging imaging method combining light and sound. However, limited by the laser's repetition rate, state-of-the-art high-speed PAM technology often sacrifices spatial sampling density (i.e., undersampling) for increased imaging speed over a large field-of-view. Deep learning (DL) methods have recently been used to improve sparsely sampled PAM images; however, these methods often require time-consuming pre-training and large training dataset with ground truth. Here, we propose the use of deep image prior (DIP) to improve the image quality of undersampled PAM images. Unlike other DL approaches, DIP requires neither pre-training nor fully-sampled ground truth, enabling its flexible and fast implementation on various imaging targets. Our results have demonstrated substantial improvement in PAM images with as few as 1.4$\%$ of the fully sampled pixels on high-speed PAM. Our approach outperforms interpolation, is competitive with pre-trained supervised DL method, and is readily translated to other high-speed, undersampling imaging modalities.

eess.IV↗

arXiv2020

Tri-modality Cavitation Mapping in Shock Wave Lithotripsy

Shock wave lithotripsy (SWL) has been widely used for non-invasive treatment of kidney stones. Cavitation plays an important role in stone fragmentation, yet may also contribute to renal injury during SWL. It is therefore crucial to determine the spatiotemporal distributions of cavitation activities to maximize stone fragmentation while minimizing tissue injury. Traditional cavitation detection methods include high-speed optical imaging, active cavitation mapping (ACM), and passive cavitation mapping (PCM). While each of the three methods provides unique information about the dynamics of the bubbles, PCM has most practical applications in biological tissues. To image the dynamics of cavitation bubble collapse, we previously developed a sliding-window PCM (SW-PCM) method to identify each bubble collapse with high temporal and spatial resolution. To further validate and optimize the SW-PCM method, in this work, we have developed tri-modality cavitation imaging that includes 3D high-speed optical imaging, ACM, and PCM seamlessly integrated in a single system. Using the tri-modality system, we imaged and analyzed laser-induced single cavitation bubbles in both free and constricted space and shockwave-induced cavitation clusters. Collectively, our results have demonstrated the high reliability and spatial-temporal accuracy of the SW-PCM approach, which paves the way for future in vivo applications on large animals and humans in SWL.

physics.app-ph↗

arXiv2020

Internal-Illumination Photoacoustic Tomography Enhanced by a Graded-scattering Fiber Diffuser

The penetration depth of photoacoustic imaging in biological tissues has been fundamentally limited by the strong optical attenuation when light is delivered externally through the tissue surface. To address this issue, we previously reported internal-illumination photoacoustic imaging using a customized radial-emission optical fiber diffuser, which, however, has complex fabrication, high cost, and non-uniform light emission. To overcome these shortcomings, we have developed a new type of low-cost fiber diffusers based on a graded-scattering method in which the optical scattering of the fiber diffuser is gradually increased as the light travels. The graded scattering can compensate for the optical attenuation and provide relatively uniform light emission along the diffuser. We performed Monte Carlo numerical simulations to optimize several key design parameters, including the number of scattering segments, scattering anisotropy factor, divergence angle of the optical fiber, and reflective index of the surrounding medium. These optimized parameters collectively result in uniform light emission along the fiber diffuser and can be flexibly adjusted to accommodate different applications. We fabricated and characterized the prototype fiber diffuser made of agarose gel and intralipid. Equipped with the new fiber diffuser, we performed thorough proof-of-concept studies on ex vivo tissue phantoms and an in vivo swine model to demonstrate the deep-imaging capability (~10 cm achieved ex vivo) of photoacoustic tomography. We believe that the internal light delivery via the optimized fiber diffuser is an effective strategy to image deep targets (e.g., kidney) in large animals or humans.

physics.med-ph↗

arXiv2020

3D Monte Carlo Simulation of Light Distribution in Mouse Brain in Quantitative Photoacoustic Computed Tomography

Photoacoustic computed tomography (PACT) detects light-induced ultrasound waves to reconstruct the optical absorption contrast of the biological tissues. Due to its relatively deep penetration (several centimeters in soft tissue), high spatial resolution, and inherent functional sensitivity, PACT has great potential for imaging mouse brains with endogenous and exogenous contrasts, which is of immense interest to the neuroscience community. However, conventional PACT either assumes homogenous optical fluence within the brain or uses a simplified attenuation model for optical fluence estimation. Both approaches underestimate the complexity of the fluence heterogeneity and can result in poor quantitative imaging accuracy. To optimize the quantitative performance of PACT, we explore for the first time 3D Monte Carlo simulation to study the optical fluence distribution in a complete mouse brain model. We apply the MCX Monte Carlo simulation package on a digital mouse (Digimouse) brain atlas that has complete anatomy information. To evaluate the impact of the brain vasculature on light delivery, we also incorporate the whole-brain vasculature in the Digimouse atlas. The simulation results clearly show that the optical fluence in the mouse brain is heterogeneous at the global level and can decrease by a factor of five with increasing depth. Moreover, the strong absorption and scattering of the brain vasculature also induce the fluence disturbance at the local level. Our results suggest that both global and local fluence heterogeneity contributes to the reduced quantitative accuracy of the reconstructed PACT images of mouse brain.

physics.med-ph↗

arXiv2020

Reconstructing undersampled photoacoustic microscopy images using deep learning

One primary technical challenge in photoacoustic microscopy (PAM) is the necessary compromise between spatial resolution and imaging speed. In this study, we propose a novel application of deep learning principles to reconstruct undersampled PAM images and transcend the trade-off between spatial resolution and imaging speed. We compared various convolutional neural network (CNN) architectures, and selected a fully dense U-net (FD U-net) model that produced the best results. To mimic various undersampling conditions in practice, we artificially downsampled fully-sampled PAM images of mouse brain vasculature at different ratios. This allowed us to not only definitively establish the ground truth, but also train and test our deep learning model at various imaging conditions. Our results and numerical analysis have collectively demonstrated the robust performance of our model to reconstruct PAM images with as few as 2% of the original pixels, which may effectively shorten the imaging time without substantially sacrificing the image quality.

eess.IV↗

arXiv2015

The Collaborate Calibration and Alignment of Button-type BPM

Beam position monitor (BPM) can easily reinforce the handling of beam orbits and measure the absolute beam position [1]. Its data can be used to optimize and correct beam in both first turn and closed orbit mode. In order to set the absolute center position of Button-type BPM, and formulate the offset between mechanic and electronic center precisely, we mounted BPM together with solenoid on a vertical rotated test-bench when its calibration takes out, and developed transform software to calculate the offset. This paper describes the method and process of collaborate calibration: the assembly and alignment of BPM itself on the designed work-bench; the mechanic calibration of bundle BPM-Solenoid, and the alignment of mechanic to the wire center used by Laser Tracker and Portable coordinate measurement machine (CMM) jointly; the connection of coaxial cable and read-out for electronics; the electronic calibration of bundle BPM-Solenoid. Form the above four steps, the author analyses the error sources, measures and subtracts the offset values, and formulate the position calculation formula of offset. At last, the residual offset is nearly 20um.The total accuracy of calibration is 0.1mm.

physics.acc-ph↗

arXiv2013

Community Specific Temporal Topic Discovery from Social Media

Studying temporal dynamics of topics in social media is very useful to understand online user behaviors. Most of the existing work on this subject usually monitors the global trends, ignoring variation among communities. Since users from different communities tend to have varying tastes and interests, capturing community-level temporal change can improve the understanding and management of social content. Additionally, it can further facilitate the applications such as community discovery, temporal prediction and online marketing. However, this kind of extraction becomes challenging due to the intricate interactions between community and topic, and intractable computational complexity. In this paper, we take a unified solution towards the community-level topic dynamic extraction. A probabilistic model, CosTot (Community Specific Topics-over-Time) is proposed to uncover the hidden topics and communities, as well as capture community-specific temporal dynamics. Specifically, CosTot considers text, time, and network information simultaneously, and well discovers the interactions between community and topic over time. We then discuss the approximate inference implementation to enable scalable computation of model parameters, especially for large social data. Based on this, the application layer support for multi-scale temporal analysis and community exploration is also investigated. We conduct extensive experimental studies on a large real microblog dataset, and demonstrate the superiority of proposed model on tasks of time stamp prediction, link prediction and topic perplexity.

cs.SI↗

arXiv2012

Challenging the Long Tail Recommendation

The success of "infinite-inventory" retailers such as Amazon.com and Netflix has been largely attributed to a "long tail" phenomenon. Although the majority of their inventory is not in high demand, these niche products, unavailable at limited-inventory competitors, generate a significant fraction of total revenue in aggregate. In addition, tail product availability can boost head sales by offering consumers the convenience of "one-stop shopping" for both their mainstream and niche tastes. However, most of existing recommender systems, especially collaborative filter based methods, can not recommend tail products due to the data sparsity issue. It has been widely acknowledged that to recommend popular products is easier yet more trivial while to recommend long tail products adds more novelty yet it is also a more challenging task. In this paper, we propose a novel suite of graph-based algorithms for the long tail recommendation. We first represent user-item information with undirected edge-weighted graph and investigate the theoretical foundation of applying Hitting Time algorithm for long tail item recommendation. To improve recommendation diversity and accuracy, we extend Hitting Time and propose efficient Absorbing Time algorithm to help users find their favorite long tail items. Finally, we refine the Absorbing Time algorithm and propose two entropy-biased Absorbing Cost algorithms to distinguish the variation on different user-item rating pairs, which further enhances the effectiveness of long tail recommendation. Empirical experiments on two real life datasets show that our proposed algorithms are effective to recommend long tail items and outperform state-of-the-art recommendation techniques.

cs.DB↗