Enhancing signal-to-noise ratio in real-time LED-based photoacoustic imaging: A comparative study of CNN-based deep learning architectures
Recent advances in Light Emitting Diode (LED) technology have enabled a more affordable high frame rate photoacoustic imaging (PA) alternative to traditional laser-based PA systems that are costly and have slow pulse repetition rate. However, a major disadvantage with LEDs is the low energy outputs...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-02-01
|
| Series: | Photoacoustics |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2213597924000910 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841526312036990976 |
|---|---|
| author | Avijit Paul Srivalleesha Mallidi |
| author_facet | Avijit Paul Srivalleesha Mallidi |
| author_sort | Avijit Paul |
| collection | DOAJ |
| description | Recent advances in Light Emitting Diode (LED) technology have enabled a more affordable high frame rate photoacoustic imaging (PA) alternative to traditional laser-based PA systems that are costly and have slow pulse repetition rate. However, a major disadvantage with LEDs is the low energy outputs that do not produce high signal-to-noise ratio (SNR) PA images. There have been recent advancements in integrating deep learning methodologies aimed to address the challenge of improving SNR in LED-PA images, yet comprehensive evaluations across varied datasets and architectures are lacking. In this study, we systematically assess the efficacy of various Encoder-Decoder-based CNN architectures for enhancing SNR in real-time LED-based PA imaging. Through experimentation with in vitro phantoms, ex vivo mouse organs, and in vivo tumors, we compare basic convolutional autoencoder and U-Net architectures, explore hierarchical depth variations within U-Net, and evaluate advanced variants of U-Net. Our findings reveal that while U-Net architectures generally exhibit comparable performance, the Dense U-Net model shows promise in denoising different noise distributions in the PA image. Notably, hierarchical depth variations did not significantly impact performance, emphasizing the efficacy of the standard U-Net architecture for practical applications. Moreover, the study underscores the importance of evaluating robustness to diverse noise distributions, with Dense U-Net and R2 U-Net demonstrating resilience to Gaussian, salt and pepper, Poisson, and Speckle noise types. These insights inform the selection of appropriate deep learning architectures based on application requirements and resource constraints, contributing to advancements in PA imaging technology. |
| format | Article |
| id | doaj-art-eb416691eb1141f9ba8f268f48f9d3e1 |
| institution | Kabale University |
| issn | 2213-5979 |
| language | English |
| publishDate | 2025-02-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Photoacoustics |
| spelling | doaj-art-eb416691eb1141f9ba8f268f48f9d3e12025-01-17T04:49:31ZengElsevierPhotoacoustics2213-59792025-02-0141100674Enhancing signal-to-noise ratio in real-time LED-based photoacoustic imaging: A comparative study of CNN-based deep learning architecturesAvijit Paul0Srivalleesha Mallidi1Department of Biomedical Engineering, Tufts University, Medford, MA 02155, USACorresponding author.; Department of Biomedical Engineering, Tufts University, Medford, MA 02155, USARecent advances in Light Emitting Diode (LED) technology have enabled a more affordable high frame rate photoacoustic imaging (PA) alternative to traditional laser-based PA systems that are costly and have slow pulse repetition rate. However, a major disadvantage with LEDs is the low energy outputs that do not produce high signal-to-noise ratio (SNR) PA images. There have been recent advancements in integrating deep learning methodologies aimed to address the challenge of improving SNR in LED-PA images, yet comprehensive evaluations across varied datasets and architectures are lacking. In this study, we systematically assess the efficacy of various Encoder-Decoder-based CNN architectures for enhancing SNR in real-time LED-based PA imaging. Through experimentation with in vitro phantoms, ex vivo mouse organs, and in vivo tumors, we compare basic convolutional autoencoder and U-Net architectures, explore hierarchical depth variations within U-Net, and evaluate advanced variants of U-Net. Our findings reveal that while U-Net architectures generally exhibit comparable performance, the Dense U-Net model shows promise in denoising different noise distributions in the PA image. Notably, hierarchical depth variations did not significantly impact performance, emphasizing the efficacy of the standard U-Net architecture for practical applications. Moreover, the study underscores the importance of evaluating robustness to diverse noise distributions, with Dense U-Net and R2 U-Net demonstrating resilience to Gaussian, salt and pepper, Poisson, and Speckle noise types. These insights inform the selection of appropriate deep learning architectures based on application requirements and resource constraints, contributing to advancements in PA imaging technology.http://www.sciencedirect.com/science/article/pii/S2213597924000910LED based photoacoustic imagingDeep-learningConvolutional neural networksU-Net architecturesSignal-to-noise ratio |
| spellingShingle | Avijit Paul Srivalleesha Mallidi Enhancing signal-to-noise ratio in real-time LED-based photoacoustic imaging: A comparative study of CNN-based deep learning architectures Photoacoustics LED based photoacoustic imaging Deep-learning Convolutional neural networks U-Net architectures Signal-to-noise ratio |
| title | Enhancing signal-to-noise ratio in real-time LED-based photoacoustic imaging: A comparative study of CNN-based deep learning architectures |
| title_full | Enhancing signal-to-noise ratio in real-time LED-based photoacoustic imaging: A comparative study of CNN-based deep learning architectures |
| title_fullStr | Enhancing signal-to-noise ratio in real-time LED-based photoacoustic imaging: A comparative study of CNN-based deep learning architectures |
| title_full_unstemmed | Enhancing signal-to-noise ratio in real-time LED-based photoacoustic imaging: A comparative study of CNN-based deep learning architectures |
| title_short | Enhancing signal-to-noise ratio in real-time LED-based photoacoustic imaging: A comparative study of CNN-based deep learning architectures |
| title_sort | enhancing signal to noise ratio in real time led based photoacoustic imaging a comparative study of cnn based deep learning architectures |
| topic | LED based photoacoustic imaging Deep-learning Convolutional neural networks U-Net architectures Signal-to-noise ratio |
| url | http://www.sciencedirect.com/science/article/pii/S2213597924000910 |
| work_keys_str_mv | AT avijitpaul enhancingsignaltonoiseratioinrealtimeledbasedphotoacousticimagingacomparativestudyofcnnbaseddeeplearningarchitectures AT srivalleeshamallidi enhancingsignaltonoiseratioinrealtimeledbasedphotoacousticimagingacomparativestudyofcnnbaseddeeplearningarchitectures |