NOISE FILTERING AND DEEP LEARNING-BASED RECONSTRUCTION METHODS FOR IMPROVING IMAGE QUALITY IN MOBILE X-RAY EQUIPMENT UNDER LOW-DOSE RADIATION CONDITIONS
Fuente:
Zenodo
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Recurso digital |
| Language: | English |
| Published: |
Zenodo
2026
|
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866902246392856576 |
|---|---|
| author | 1D.A. Umarova, 1N.S. Yusupova, 1F.Q. Shakarov 2O.E. Jiyanbayev, 2I.N. Abdullayev, |
| author_facet | 1D.A. Umarova, 1N.S. Yusupova, 1F.Q. Shakarov 2O.E. Jiyanbayev, 2I.N. Abdullayev, |
| contents | <h2>This study presents an integrated approach for improving image quality in mobile X-ray equipment under low-dose radiation conditions using noise filtering and deep learning-based reconstruction methods. Mobile radiography systems are widely used in intensive care units, emergency departments, operating rooms, and bedside diagnostics, where patient mobility is limited and rapid imaging is required. However, reducing radiation dose in mobile X-ray imaging often leads to increased quantum noise, reduced contrast, and degradation of anatomical detail visibility. These limitations can negatively affect diagnostic accuracy, especially in chest radiography and pediatric imaging.</h2> <p class="p">To address these challenges, this study proposes a multi-stage image enhancement framework that combines conventional noise suppression, adaptive preprocessing, and deep learning-based reconstruction. The proposed methodology includes low-dose image acquisition, noise pattern estimation, adaptive filtering, convolutional neural network-based image reconstruction, edge-preserving detail enhancement, and final image quality evaluation. The approach is designed to improve signal-to-noise ratio, preserve anatomical structures, and maintain diagnostic informativeness while supporting dose reduction.</p> <p class="p">The results demonstrate that deep learning-based noise reduction and reconstruction methods can significantly improve the visual quality of low-dose mobile X-ray images without excessive smoothing or loss of clinically important details. The proposed framework provides a technically efficient solution for mobile radiography systems and can contribute to safer diagnostic imaging by reducing radiation exposure while preserving image quality.</p> |
| format | Recurso digital |
| id | zenodo_https___doi_org_10_5281_zenodo_20061489 |
| institution | Zenodo |
| language | eng |
| publishDate | 2026 |
| publisher | Zenodo |
| record_format | zenodo |
| spellingShingle | NOISE FILTERING AND DEEP LEARNING-BASED RECONSTRUCTION METHODS FOR IMPROVING IMAGE QUALITY IN MOBILE X-RAY EQUIPMENT UNDER LOW-DOSE RADIATION CONDITIONS 1D.A. Umarova, 1N.S. Yusupova, 1F.Q. Shakarov 2O.E. Jiyanbayev, 2I.N. Abdullayev, <h2>This study presents an integrated approach for improving image quality in mobile X-ray equipment under low-dose radiation conditions using noise filtering and deep learning-based reconstruction methods. Mobile radiography systems are widely used in intensive care units, emergency departments, operating rooms, and bedside diagnostics, where patient mobility is limited and rapid imaging is required. However, reducing radiation dose in mobile X-ray imaging often leads to increased quantum noise, reduced contrast, and degradation of anatomical detail visibility. These limitations can negatively affect diagnostic accuracy, especially in chest radiography and pediatric imaging.</h2> <p class="p">To address these challenges, this study proposes a multi-stage image enhancement framework that combines conventional noise suppression, adaptive preprocessing, and deep learning-based reconstruction. The proposed methodology includes low-dose image acquisition, noise pattern estimation, adaptive filtering, convolutional neural network-based image reconstruction, edge-preserving detail enhancement, and final image quality evaluation. The approach is designed to improve signal-to-noise ratio, preserve anatomical structures, and maintain diagnostic informativeness while supporting dose reduction.</p> <p class="p">The results demonstrate that deep learning-based noise reduction and reconstruction methods can significantly improve the visual quality of low-dose mobile X-ray images without excessive smoothing or loss of clinically important details. The proposed framework provides a technically efficient solution for mobile radiography systems and can contribute to safer diagnostic imaging by reducing radiation exposure while preserving image quality.</p> |
| title | NOISE FILTERING AND DEEP LEARNING-BASED RECONSTRUCTION METHODS FOR IMPROVING IMAGE QUALITY IN MOBILE X-RAY EQUIPMENT UNDER LOW-DOSE RADIATION CONDITIONS |
| url | https://doi.org/10.5281/zenodo.20061489 |