Reconstruction of Optical Coherence Tomography Images from Wavelength Space Using Deep Learning
Conventional Fourier domain Optical Coherence Tomography (FD-OCT) systems depend on resampling into a wavenumber (<i>k</i>) domain to extract the depth profile. This either necessitates additional hardware resources or amplifies the existing computational complexity. Moreover, the OCT im...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-12-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/25/1/93 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841548915565920256 |
|---|---|
| author | Maryam Viqar Erdem Sahin Elena Stoykova Violeta Madjarova |
| author_facet | Maryam Viqar Erdem Sahin Elena Stoykova Violeta Madjarova |
| author_sort | Maryam Viqar |
| collection | DOAJ |
| description | Conventional Fourier domain Optical Coherence Tomography (FD-OCT) systems depend on resampling into a wavenumber (<i>k</i>) domain to extract the depth profile. This either necessitates additional hardware resources or amplifies the existing computational complexity. Moreover, the OCT images also suffer from speckle noise, due to systemic reliance on low-coherence interferometry. We propose a streamlined and computationally efficient approach based on Deep Learning (DL) which enables reconstructing speckle-reduced OCT images directly from the wavelength (λ) domain. For reconstruction, two encoder–decoder styled networks, namely Spatial Domain Convolution Neural Network (SD-CNN) and Fourier Domain CNN (FD-CNN), are used sequentially. The SD-CNN exploits the highly degraded images obtained by Fourier transforming the (λ) domain fringes to reconstruct the deteriorated morphological structures along with suppression of unwanted noise. The FD-CNN leverages this output to enhance the image quality further by optimization in the Fourier domain (FD). We quantitatively and visually demonstrate the efficacy of the method in obtaining high-quality OCT images. Furthermore, we illustrate the computational complexity reduction by harnessing the power of DL models. We believe that this work lays the framework for further innovations in the realm of OCT image reconstruction. |
| format | Article |
| id | doaj-art-47b0f2a0f331442296de0ca28b0a6fc3 |
| institution | Kabale University |
| issn | 1424-8220 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj-art-47b0f2a0f331442296de0ca28b0a6fc32025-01-10T13:20:51ZengMDPI AGSensors1424-82202024-12-012519310.3390/s25010093Reconstruction of Optical Coherence Tomography Images from Wavelength Space Using Deep LearningMaryam Viqar0Erdem Sahin1Elena Stoykova2Violeta Madjarova3Faculty of Information Technology and Communication Sciences, Tampere University, 33720 Tampere, FinlandFaculty of Information Technology and Communication Sciences, Tampere University, 33720 Tampere, FinlandInstitute of Optical Materials and Technologies, Bulgarian Academy of Sciences, 1113 Sofia, BulgariaInstitute of Optical Materials and Technologies, Bulgarian Academy of Sciences, 1113 Sofia, BulgariaConventional Fourier domain Optical Coherence Tomography (FD-OCT) systems depend on resampling into a wavenumber (<i>k</i>) domain to extract the depth profile. This either necessitates additional hardware resources or amplifies the existing computational complexity. Moreover, the OCT images also suffer from speckle noise, due to systemic reliance on low-coherence interferometry. We propose a streamlined and computationally efficient approach based on Deep Learning (DL) which enables reconstructing speckle-reduced OCT images directly from the wavelength (λ) domain. For reconstruction, two encoder–decoder styled networks, namely Spatial Domain Convolution Neural Network (SD-CNN) and Fourier Domain CNN (FD-CNN), are used sequentially. The SD-CNN exploits the highly degraded images obtained by Fourier transforming the (λ) domain fringes to reconstruct the deteriorated morphological structures along with suppression of unwanted noise. The FD-CNN leverages this output to enhance the image quality further by optimization in the Fourier domain (FD). We quantitatively and visually demonstrate the efficacy of the method in obtaining high-quality OCT images. Furthermore, we illustrate the computational complexity reduction by harnessing the power of DL models. We believe that this work lays the framework for further innovations in the realm of OCT image reconstruction.https://www.mdpi.com/1424-8220/25/1/93image reconstructionoptical coherence tomographyspeckle noisetime complexity |
| spellingShingle | Maryam Viqar Erdem Sahin Elena Stoykova Violeta Madjarova Reconstruction of Optical Coherence Tomography Images from Wavelength Space Using Deep Learning Sensors image reconstruction optical coherence tomography speckle noise time complexity |
| title | Reconstruction of Optical Coherence Tomography Images from Wavelength Space Using Deep Learning |
| title_full | Reconstruction of Optical Coherence Tomography Images from Wavelength Space Using Deep Learning |
| title_fullStr | Reconstruction of Optical Coherence Tomography Images from Wavelength Space Using Deep Learning |
| title_full_unstemmed | Reconstruction of Optical Coherence Tomography Images from Wavelength Space Using Deep Learning |
| title_short | Reconstruction of Optical Coherence Tomography Images from Wavelength Space Using Deep Learning |
| title_sort | reconstruction of optical coherence tomography images from wavelength space using deep learning |
| topic | image reconstruction optical coherence tomography speckle noise time complexity |
| url | https://www.mdpi.com/1424-8220/25/1/93 |
| work_keys_str_mv | AT maryamviqar reconstructionofopticalcoherencetomographyimagesfromwavelengthspaceusingdeeplearning AT erdemsahin reconstructionofopticalcoherencetomographyimagesfromwavelengthspaceusingdeeplearning AT elenastoykova reconstructionofopticalcoherencetomographyimagesfromwavelengthspaceusingdeeplearning AT violetamadjarova reconstructionofopticalcoherencetomographyimagesfromwavelengthspaceusingdeeplearning |