scholarly journals Sounding out the hidden data: A concise review of deep learning in photoacoustic imaging

2021 ◽  
pp. 153537022110003
Author(s):  
Anthony DiSpirito ◽  
Tri Vu ◽  
Manojit Pramanik ◽  
Junjie Yao
Keyword(s):  
Artificial Intelligence ◽  
Deep Learning ◽  
Structural Information ◽  
Photoacoustic Imaging ◽  
Functional Brain Imaging ◽  
Photoacoustic Tomography ◽  
Model Parameters ◽  
Processing Unit ◽  
Concise Review

The rapidly evolving field of photoacoustic tomography utilizes endogenous chromophores to extract both functional and structural information from deep within tissues. It is this power to perform precise quantitative measurements in vivo—with endogenous or exogenous contrast —that makes photoacoustic tomography highly promising for clinical translation in functional brain imaging, early cancer detection, real-time surgical guidance, and the visualization of dynamic drug responses. Considering photoacoustic tomography has benefited from numerous engineering innovations, it is of no surprise that many of photoacoustic tomography’s current cutting-edge developments incorporate advances from the equally novel field of artificial intelligence. More specifically, alongside the growth and prevalence of graphical processing unit capabilities within recent years has emerged an offshoot of artificial intelligence known as deep learning. Rooted in the solid foundation of signal processing, deep learning typically utilizes a method of optimization known as gradient descent to minimize a loss function and update model parameters. There are already a number of innovative efforts in photoacoustic tomography utilizing deep learning techniques for a variety of purposes, including resolution enhancement, reconstruction artifact removal, undersampling correction, and improved quantification. Most of these efforts have proven to be highly promising in addressing long-standing technical obstacles where traditional solutions either completely fail or make only incremental progress. This concise review focuses on the history of applied artificial intelligence in photoacoustic tomography, presents recent advances at this multifaceted intersection of fields, and outlines the most exciting advances that will likely propagate into promising future innovations.

In vivo demonstration of reflection artifact reduction in LED-based photoacoustic imaging using deep learning

2021 ◽  
Author(s):  
Sumit Agrawal ◽  
Kerrick Johnstonbaugh ◽  
Thaarakh Suresh ◽  
Ankit Garikipati ◽  
Mithun Kuniyil Ajith Singh ◽  
...  
Keyword(s):  
Deep Learning ◽  
Photoacoustic Imaging ◽  
Artifact Reduction

scholarly journals Y-Net: Hybrid deep learning image reconstruction for photoacoustic tomography in vivo

2020 ◽  
Vol 20 ◽  
pp. 100197 ◽  
Author(s):  
Hengrong Lan ◽  
Daohuai Jiang ◽  
Changchun Yang ◽  
Feng Gao ◽  
Fei Gao
Keyword(s):  
Deep Learning ◽  
Image Reconstruction ◽  
Photoacoustic Tomography

scholarly journals Characterization of a Fiber Bundle-Based Real-Time Ultrasound/Photoacoustic Imaging System and Its In Vivo Functional Imaging Applications

Micromachines ◽  
2019 ◽  
Vol 10 (12) ◽  
pp. 820
Author(s):  
He Leng ◽  
Yuhling Wang ◽  
De-Fu Jhang ◽  
Tsung-Sheng Chu ◽  
Chia-Hui Tsao ◽  
...  
Keyword(s):  
User Interface ◽  
Real Time ◽  
In Vivo Imaging ◽  
Fiber Bundle ◽  
Imaging System ◽  
Structural Information ◽  
Photoacoustic Imaging ◽  
Numerical Aperture ◽  
Biological Tissues

Photoacoustic (PA) imaging is an attractive technology for imaging biological tissues because it can capture both functional and structural information with satisfactory spatial resolution. Current commercially available PA imaging systems are limited by their bulky size or inflexible user interface. We present a new handheld real-time ultrasound/photoacoustic imaging system (HARP) consisting of a detachable, high-numerical-aperture (NA) fiber bundle-based illumination system integrated with an array-based ultrasound (US) transducer and a data acquisition platform. In this system, different PA probes can be used for different imaging applications by switching the transducers and the corresponding jackets to combine the fiber pads and transducer into a single probe. The intuitive user interface is a completely programmable MATLAB-based platform. In vitro phantom experiments were conducted to test the imaging performance of the developed PA system. Furthermore, we demonstrated (1) in vivo brain vasculature imaging, (2) in vivo imaging of real-time stimulus-evoked cortical hemodynamic changes during forepaw electrical stimulation, and (3) in vivo imaging of real-time cerebral pharmacokinetics in rats using the developed PA system. The overall purpose of this design concept for a customizable US/PA imaging system is to help overcome the diverse challenges faced by medical researchers performing both preclinical and clinical PA studies.

scholarly journals Artificial Intelligence & Machine Learning in Computer Vision Applications

2020 ◽  
Vol 7 (1) ◽  
pp. 2-3
Author(s):  
Shadi Saleh
Keyword(s):  
Artificial Intelligence ◽  
Machine Learning ◽  
Neural Networks ◽  
Computer Vision ◽  
Deep Learning ◽  
Large Scale ◽  
Short Term Memory ◽  
Graphics Processing Unit ◽  
Multimedia Data ◽  
Processing Unit

Deep learning and machine learning innovations are at the core of the ongoing revolution in Artificial Intelligence for the interpretation and analysis of multimedia data. The convergence of large-scale datasets and more affordable Graphics Processing Unit (GPU) hardware has enabled the development of neural networks for data analysis problems that were previously handled by traditional handcrafted features. Several deep learning architectures such as Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), Long Short Term Memory (LSTM)/Gated Recurrent Unit (GRU), Deep Believe Networks (DBN), and Deep Stacking Networks (DSNs) have been used with new open source software and libraries options to shape an entirely new scenario in computer vision processing.

scholarly journals An Adjustable Dark-Field Acoustic-Resolution Photoacoustic Imaging System with Fiber Bundle-Based Illumination

Biosensors ◽  
2021 ◽  
Vol 11 (8) ◽  
pp. 262
Author(s):  
Yuhling Wang ◽  
De-Fu Jhang ◽  
Tsung-Sheng Chu ◽  
Chia-Hui Tsao ◽  
Chia-Hua Tsai ◽  
...  
Keyword(s):  
Real Time ◽  
Imaging System ◽  
Structural Information ◽  
Photoacoustic Imaging ◽  
Biological Tissues ◽  
Dark Field ◽  
Design Concept ◽  
High Frequency Ultrasound

Photoacoustic (PA) imaging has become one of the major imaging methods because of its ability to record structural information and its high spatial resolution in biological tissues. Current commercialized PA imaging instruments are limited to varying degrees by their bulky size (i.e., the laser or scanning stage) or their use of complex optical components for light delivery. Here, we present a robust acoustic-resolution PA imaging system that consists of four adjustable optical fibers placed 90° apart around a 50 MHz high-frequency ultrasound (US) transducer. In the compact design concept of the PA probe, the relative illumination parameters (i.e., angles and fiber size) can be adjusted to fit different imaging applications in a single setting. Moreover, this design concept involves a user interface built in MATLAB. We first assessed the performance of our imaging system using in vitro phantom experiments. We further demonstrated the in vivo performance of the developed system in imaging (1) rat ear vasculature, (2) real-time cortical hemodynamic changes in the superior sagittal sinus (SSS) during left-forepaw electrical stimulation, and (3) real-time cerebral indocyanine green (ICG) dynamics in rats. Collectively, this alignment-free design concept of a compact PA probe without bulky optical lens systems is intended to satisfy the diverse needs in preclinical PA imaging studies.

scholarly journals PFDLIS: Privacy-Preserving and Fair Deep Learning Inference Service under Publicly Verifiable Covert Security Setting

Electronics ◽  
2019 ◽  
Vol 8 (12) ◽  
pp. 1488
Author(s):  
Fengyi Tang ◽  
Jialu Hao ◽  
Jian Liu ◽  
Huimei Wang ◽  
Ming Xian
Keyword(s):  
Artificial Intelligence ◽  
Deep Learning ◽  
Prediction Accuracy ◽  
Data Privacy ◽  
Input Data ◽  
Computational Cost ◽  
Privacy Preserving ◽  
Model Parameters ◽  
Privacy Model ◽  
Security Properties

The recent popularity and widespread use of deep learning heralds an era of artificial intelligence. Thanks to the emergence of a deep learning inference service, non-professional clients can enjoy the improvements and profits brought by artificial intelligence as well. However, the input data of the client may be sensitive so that the client does not want to send its input data to the server. Similarly, the pre-trained model of the server is valuable and the server is unwilling to make the model parameters public. Therefore, we propose a privacy-preserving and fair scheme for a deep learning inference service based on secure three-party computation and making commitments under the publicly verifiable covert security setting. We demonstrate that our scheme has the following desirable security properties—input data privacy, model privacy and defamation freeness. Finally, we conduct extensive experiments to evaluate the performance of our scheme on MNIST dataset. The experimental results verify that our scheme can achieve the same prediction accuracy as the pre-trained model with acceptable extra computational cost.

scholarly journals Full-view in vivo skin and blood vessels profile segmentation in photoacoustic imaging based on deep learning

2021 ◽  
pp. 100310
Author(s):  
Cao Duong Ly ◽  
Van Tu Nguyen ◽  
Tan Hung Vo ◽  
Sudip Mondal ◽  
Sumin Park ◽  
...  
Keyword(s):  
Deep Learning ◽  
Blood Vessels ◽  
Photoacoustic Imaging ◽  
Full View ◽  
Profile Segmentation

scholarly journals cDeepbind: A context sensitive deep learning model of RNA-protein binding

2018 ◽  
Author(s):  
Shreshth Gandhi ◽  
Leo J. Lee ◽  
Andrew Delong ◽  
David Duvenaud ◽  
Brendan J. Frey
Keyword(s):  
Deep Learning ◽  
Secondary Structure ◽  
Rna Binding ◽  
Short Term Memory ◽  
Structural Information ◽  
Genetic Disorders ◽  
In Vitro Test ◽  
Convolutional Network

AbstractMotivationDetermining RNA binding protein(RBP) binding specificity is crucial for understanding many cellular processes and genetic disorders. RBP binding is known to be affected by both the sequence and structure of RNAs. Deep learning can be used to learn generalizable representations of raw data and has improved state of the art in several fields such as image classification, speech recognition and even genomics. Previous work on RBP binding has either used shallow models that combine sequence and structure or deep models that use only the sequence. Here we combine both abilities by augmenting and refining the original Deepbind architecture to capture structural information and obtain significantly better performance.ResultsWe propose two deep architectures, one a lightweight convolutional network for transcriptome wide inference and another a Long Short-Term Memory(LSTM) network that is suitable for small batches of data. We incorporate computationally predicted secondary structure features as input to our models and show its effectiveness in boosting prediction performance. Our models achieved significantly higher correlations on held out in-vitro test data compared to previous approaches, and generalise well to in-vivo CLIP-SEQ data achieving higher median AUCs than other approaches. We analysed the output from our model for VTS1 and CPO and provided intuition into its working. Our models confirmed known secondary structure preferences for some proteins as well as found new ones where secondary structure might play a role. We also demonstrated the strengths of our model compared to other approaches such as the ability to combine information from long distances along the input.AvailabilitySoftware and models are available at https://github.com/shreshthgandhi/[email protected], [email protected]

Sign in / Sign up

Export Citation Format

Share Document