Deep learning methods hold promise for light fluence compensation in three-dimensional optoacoustic imaging

Madasamy, Arumugaraj; Gujrati, Vipul; Ntziachristos, Vasilis; Prakash, Jaya

doi:10.1117/1.jbo.27.10.106004

Cited by 20 publications

(22 citation statements)

References 74 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Compensating for the distribution of light fluence can improve the accuracy of qPAI 129 – 131 To this end, Madasamy et al 132 . compared the compensation performance of different DL models.…”

Section: Challenges In Pai and Solutions Through DLmentioning

confidence: 99%

Recent advances in deep-learning-enhanced photoacoustic imaging

et al. 2023

View full text Add to dashboard Cite

Photoacoustic imaging (PAI), recognized as a promising biomedical imaging modality for preclinical and clinical studies, uniquely combines the advantages of optical and ultrasound imaging. Despite PAI's great potential to provide valuable biological information, its wide application has been hindered by technical limitations, such as hardware restrictions or lack of the biometric information required for image reconstruction. We first analyze the limitations of PAI and categorize them by seven key challenges: limited detection, low-dosage light delivery, inaccurate quantification, limited numerical reconstruction, tissue heterogeneity, imperfect image segmentation/classification, and others. Then, because deep learning (DL) has increasingly demonstrated its ability to overcome the physical limitations of imaging modalities, we review DL studies from the past five years that address each of the seven challenges in PAI. Finally, we discuss the promise of future research directions in DL-enhanced PAI.

show abstract

Section: Challenges In Pai and Solutions Through DLmentioning

confidence: 99%

Recent advances in deep-learning-enhanced photoacoustic imaging

et al. 2023

View full text Add to dashboard Cite

show abstract

“…Also, it will be important to explore fluence-compensated oxygen saturation estimation for obtaining more quantitative information. 17,41,42,43…”

Section: Real-time Imaging Of Tail-vein So2 Using Acousticxmentioning

confidence: 99%

Oxygen saturation imaging using dual-wavelength LED-based photoacoustics

Singh¹,

Sato²,

Tsunoi³

et al. 2023

Photons Plus Ultrasound: Imaging and Sensing 2023

View full text Add to dashboard Cite

Since hypoxia is an early marker for cancer and several microvascular disorders, imaging oxygen saturation with high resolution has profound importance in multiple preclinical and clinical imaging studies. In this work, we demonstrate the potential of dual-wavelength (690/850 nm) LED-based photoacoustic imaging in high resolution real-time oxygen saturation imaging in vivo. We performed two live rat imaging experiments in which the oxygen saturation of the bloodvessels in the left hindlimb was imaged for 20 seconds while the animal was breathing pure nitrogen gas to induce hypoxia. For a comparison, we continuously monitored the oxygen level using a pulse oximeter connected to the forelimb. In another experiment, we performed imaging of oxygen saturation changes in the tail vein of the rat while hypoxia was induced over the period of 40 seconds. In this case also, pulse oximeter readings were recorded for a comparison. In all the experiments, photoacoustic-based oxygen saturation values measured over time followed the same trend as the reference values provided by the pulse oximeter. Also, our two wavelength LED-based photoacoustic imaging approach was found to be sensitive even for 3% change in oxygen saturation. Results give a direct confirmation about the potential of LED-based photoacoustic imaging in detection of oxygen saturation with high spatio-temporal resolution, making it an ideal tool for hypoxia detection in longitudinal preclinical and clinical imaging studies.

show abstract

“…Recently, data-driven approaches have been proposed as a potential solution because they do not require explicit prior information about the object and can be computationally efficient. 6,[14][15][16][17][18][19][20][21][22][23][24] However, the effectiveness of such methods has not been established on full-scale problems in which clinically relevant variability in both anatomy and physiological parameters is considered. 6,[14][15][16][17][18][19][20][21]24 As such, there exists an important need to investigate the efficacy of data-driven methods under realistic conditions.…”

Section: Introductionmentioning

confidence: 99%

Investigation of a learned image reconstruction method for three-dimensional quantitative photoacoustic tomography of the breast

Cam,

Park,

Villa

et al. 2024

Photons Plus Ultrasound: Imaging and Sensing 2024

View full text Add to dashboard Cite

Quantitative photoacoustic computed tomography (qPACT) holds great promise to advance a variety of important clinical applications with its potential to estimate vital physiological properties such as oxygen saturation. However, the qPACT reconstruction problem is highly nonlinear and ill-posed. Conventional spectral unmixing methods often oversimplify the problem, resulting in suboptimal accuracy. Alternatively, more principled image reconstruction approaches that comprehensively model the imaging physics are computationally burdensome and require the design of effective regularization strategies. To overcome these limitations, learning-based methods have been proposed. To date, however, the effectiveness of such methods on full-scale problems in which clinically relevant variability in anatomy and physiological parameters is considered has not been established. To address this, we investigated the use of a convolutional neural network with spatial and channel attention modules to estimate the three-dimensional (3D) distribution of tissue oxygenation within vessels and lesions in the female breast. The network was provided with input data comprising noise-corrupted 3D initial pressure distributions corresponding to three wavelengths (757, 800, 850 nm). An additional novel aspect of our study was the use of realistic 3D numerical breast phantoms that described stochastic variations in breast anatomy and functional properties, which enabled a meaningful, quantitative, and systematic evaluation of the proposed method. This study represents an important contribution to the field of qPACT and will guide the exploration of learning-based methods to help translate this important technology by delineating potential prospects and limitations.

show abstract

Deep learning methods hold promise for light fluence compensation in three-dimensional optoacoustic imaging

Cited by 20 publications

References 74 publications

Recent advances in deep-learning-enhanced photoacoustic imaging

Recent advances in deep-learning-enhanced photoacoustic imaging

Oxygen saturation imaging using dual-wavelength LED-based photoacoustics

Investigation of a learned image reconstruction method for three-dimensional quantitative photoacoustic tomography of the breast

Contact Info

Product

Resources

About