FFTMed: leveraging fast-fourier transform for a lightweight and adversarial-resilient medical image segmentation framework.
Gespeichert in:
| Titel: | FFTMed: leveraging fast-fourier transform for a lightweight and adversarial-resilient medical image segmentation framework. |
|---|---|
| Autoren: | Pham VT; Department of Computer Science, The University of Alabama at Birmingham, Birmingham, USA., Ha MH; Department of Computer Science, School of Information and Communication Technology, Hanoi University of Science and Technology, Hanoi, Vietnam., Bui BVQ; Department of Computer Science, Ho Chi Minh City University of Technology, Ho Chi Minh City, Vietnam., Hy TS; Department of Computer Science, The University of Alabama at Birmingham, Birmingham, USA. thy@uab.edu. |
| Quelle: | Scientific reports [Sci Rep] 2025 Oct 29; Vol. 15 (1), pp. 37879. Date of Electronic Publication: 2025 Oct 29. |
| Publikationsart: | Journal Article |
| Sprache: | English |
| Info zur Zeitschrift: | Publisher: Nature Publishing Group Country of Publication: England NLM ID: 101563288 Publication Model: Electronic Cited Medium: Internet ISSN: 2045-2322 (Electronic) Linking ISSN: 20452322 NLM ISO Abbreviation: Sci Rep Subsets: MEDLINE |
| Imprint Name(s): | Original Publication: London : Nature Publishing Group, copyright 2011- |
| MeSH-Schlagworte: | Fourier Analysis* , Image Processing, Computer-Assisted*/methods , Diagnostic Imaging*/methods, Humans ; Neural Networks, Computer ; Algorithms |
| Abstract: | Accurate and reliable medical image segmentation is essential for computer-aided diagnosis and formulating appropriate treatment plans. However, noise often significantly reduces diagnostic accuracy and complicates treatment planning. Therefore, noise reduction in medical imaging is paramount, as it not only improves diagnostic accuracy but also contributes to enhanced treatment efficacy and minimizes patient risk. Prior methods have explored frequency-domain approaches to accelerate convolutional operations or combine frequency-based features with spatial convolutions. However, most only partially integrate Fourier-based processing and thus fail to fully exploit its advantages. We propose a novel neural architecture, FFTMed, that operates directly in the frequency domain, harnessing its resilience to noise and uneven brightness while also reducing computational overhead. Notably, FFTMed requires no additional noise augmentation during training yet remains resilient when confronted with noisy test images, demonstrating its effectiveness in real-world medical image segmentation tasks. Additionally, we propose a new benchmark incorporating various levels of noise to assess susceptibility to noise attacks. The experimental results demonstrate that FFTMed not only effectively eliminates noise and consistently achieves accurate image segmentation but also shows robust resistance to imperceptible adversarial attacks compared to other baseline models. The datasets generated and analysed during this study have been deposited in the Zenodo repository and are openly accessible at https://zenodo.org/records/15310397 . The source code to reproduce all experiments is publicly available at https://github.com/HySonLab/LightMed . (© 2025. The Author(s).) |
| References: | Comput Methods Programs Biomed. 2025 Apr;261:108611. (PMID: 39892086) Med Image Comput Comput Assist Interv. 2023 Oct;14222:561-571. (PMID: 38840671) IEEE Trans Pattern Anal Mach Intell. 2024 Sep 13;PP:. (PMID: 39269798) Deep Learn Med Image Anal Multimodal Learn Clin Decis Support (2018). 2018 Sep;11045:3-11. (PMID: 32613207) Med Image Anal. 2021 Oct;73:102141. (PMID: 34246850) Adv Neural Inf Process Syst. 2022 Dec;35:29582-29596. (PMID: 37533756) Med Phys. 2024 Jan;51(1):209-223. (PMID: 37966121) IEEE J Biomed Health Inform. 2023 Oct;27(10):4816-4827. (PMID: 37796719) Quant Imaging Med Surg. 2014 Dec;4(6):475-7. (PMID: 25525580) Med Image Comput Comput Assist Interv. 2023 Oct;14223:194-205. (PMID: 38813456) Science. 2019 Mar 22;363(6433):1287-1289. (PMID: 30898923) Med Image Comput Comput Assist Interv. 2022 Sep;13434:639-652. (PMID: 37465615) J Med Imaging (Bellingham). 2023 Jan;10(1):014005. (PMID: 36820234) Med Image Anal. 2021 Aug;72:102117. (PMID: 34161914) IEEE J Biomed Health Inform. 2024 Dec 26;PP:. (PMID: 40030826) Adv Neural Inf Process Syst. 2023 Dec;36:9984-10021. (PMID: 38813114) IEEE Trans Med Imaging. 2022 Sep;41(9):2228-2237. (PMID: 35320095) Entropy (Basel). 2023 Oct 05;25(10):. (PMID: 37895539) |
| Contributed Indexing: | Keywords: Adversarial attack; Fast Fourier transform; Medical image segmentation |
| Entry Date(s): | Date Created: 20251030 Date Completed: 20251030 Latest Revision: 20251102 |
| Update Code: | 20251102 |
| PubMed Central ID: | PMC12572134 |
| DOI: | 10.1038/s41598-025-21799-5 |
| PMID: | 41162533 |
| Datenbank: | MEDLINE |
Full Text 
Full Text Finder 
Nájsť tento článok vo Web of Science | Abstract: | Accurate and reliable medical image segmentation is essential for computer-aided diagnosis and formulating appropriate treatment plans. However, noise often significantly reduces diagnostic accuracy and complicates treatment planning. Therefore, noise reduction in medical imaging is paramount, as it not only improves diagnostic accuracy but also contributes to enhanced treatment efficacy and minimizes patient risk. Prior methods have explored frequency-domain approaches to accelerate convolutional operations or combine frequency-based features with spatial convolutions. However, most only partially integrate Fourier-based processing and thus fail to fully exploit its advantages. We propose a novel neural architecture, FFTMed, that operates directly in the frequency domain, harnessing its resilience to noise and uneven brightness while also reducing computational overhead. Notably, FFTMed requires no additional noise augmentation during training yet remains resilient when confronted with noisy test images, demonstrating its effectiveness in real-world medical image segmentation tasks. Additionally, we propose a new benchmark incorporating various levels of noise to assess susceptibility to noise attacks. The experimental results demonstrate that FFTMed not only effectively eliminates noise and consistently achieves accurate image segmentation but also shows robust resistance to imperceptible adversarial attacks compared to other baseline models. The datasets generated and analysed during this study have been deposited in the Zenodo repository and are openly accessible at https://zenodo.org/records/15310397 . The source code to reproduce all experiments is publicly available at https://github.com/HySonLab/LightMed .<br /> (© 2025. The Author(s).) |
|---|---|
| ISSN: | 2045-2322 |
| DOI: | 10.1038/s41598-025-21799-5 |