Deep Learning-Based Noise Suppression and Feature Enhancement Algorithm for LED Medical Imaging Applications
Main Article Content
Abstract
Medical imaging systems employing light-emitting diodes are affected by signal degradation from photon noise, electronic interference, and wavelength-dependent tissue scattering. We present a deep learning framework integrating depthwise separable convolutions with dual-pathway attention mechanisms for noise suppression and feature enhancement in multi-spectral LED imaging. The network architecture incorporates physics-based constraints derived from LED emission profiles and tissue optical properties. Validation on 42,350 multi-spectral images from 847 patients demonstrates 34.7% signal-to-noise ratio improvement and 42.3% enhancement in diagnostic feature visibility. Processing speed reaches 28 frames per second on standard GPU hardware with 76% parameter reduction compared to baseline CNNs. Clinical evaluation shows diagnostic accuracy improvement from 76.3% to 89.7% across dermatological and vascular applications.
Article Details
Section
How to Cite
References
1. D. Shen, G. Wu, and H. I. Suk, “Deep learning in medical image analysis,” Annu. Rev. Biomed. Eng., vol. 19, no. 1, pp. 221–248, 2017, doi: 10.1146/annurev-bioeng-071516-044442.
2. J. M. Wolterink, T. Leiner, M. A. Viergever, and I. Išgum, “Generative adversarial networks for noise reduction in low-dose CT,” IEEE Trans. Med. Imaging, vol. 36, no. 12, pp. 2536–2545, 2017, doi: 10.1109/TMI.2017.2708987.
3. J. Ker, L. Wang, J. Rao, and T. Lim, “Deep learning applications in medical image analysis,” IEEE Access, vol. 6, pp. 9375–9389, 2017, doi:10.1109/ACCESS.2017.2788044.
4. J. A. Bhutto, L. Tian, Q. Du, Z. Sun, L. Yu, and M. F. Tahir, “CT and MRI medical image fusion using noise-removal and con-trast enhancement scheme with convolutional neural network,” Entropy, vol. 24, no. 3, p. 393, 2022, doi: 10.3390/e24030393.
5. K. Suzuki, “Overview of deep learning in medical imaging,” Radiol. Phys. Technol., vol. 10, no. 3, pp. 257–273, 2017, doi 10.1007/s12194-017-0406-5.
6. P. C. Tripathi and S. Bag, “CNN-DMRI: a convolutional neural network for denoising of magnetic resonance images,” Pattern Recogn. Lett., vol. 135, pp. 57–63, 2020, doi: 10.1016/j.patrec.2020.03.036.
7. J. G. Lee, S. Jun, Y. W. Cho, H. Lee, G. B. Kim, J. B. Seo, and N. Kim, “Deep learning in medical imaging: general overview,” Korean J. Radiol., vol. 18, no. 4, pp. 570–584, 2017, doi: :10.3348/kjr.2017.18.4.570.
8. Y. Chang, L. Yan, M. Chen, H. Fang, and S. Zhong, “Two-stage convolutional neural network for medical noise removal via image decomposition,” IEEE Trans. Instrum. Meas., vol. 69, no. 6, pp. 2707–2721, 2019, doi: 10.1109/TIM.2019.2925881.
9. H. Greenspan, B. Van Ginneken, and R. M. Summers, “Guest editorial deep learning in medical imaging: Overview and fu-ture promise of an exciting new technique,” IEEE Trans. Med. Imaging, vol. 35, no. 5, pp. 1153–1159, 2016, doi: 10.1109/TMI.2016.2553401.
10. J. Su, R. Pellicer-Guridi, T. Edwards, M. Fuentes, M. S. Rosen, V. Vegh, and D. Reutens, “A CNN based software gradiometer for electromagnetic background noise reduction in low field MRI applications,” IEEE Trans. Med. Imaging, vol. 41, no. 5, pp. 1007–1016, 2022, doi: 10.1109/TMI.2022.3147450.
11. M. Kim, J. Yun, Y. Cho, K. Shin, R. Jang, H. J. Bae, and N. Kim, “Deep learning in medical imaging,” Neurospine, vol. 17, no. 2, p. 471, 2020, doi: 10.14245/ns.1938396.198.
12. S. Pradeep and P. Nirmaladevi, “A review on speckle noise reduction techniques in ultrasound medical images based on spa-tial domain, transform domain and CNN methods,” in IOP Conf. Ser.: Mater. Sci. Eng., vol. 1055, no. 1, p. 012116, Feb. 2021, doi: 10.1088/1757-899X/1055/1/012116.
13. L. Gondara, “Medical image denoising using convolutional denoising autoencoders,” in Proc. IEEE 16th Int. Conf. Data Mining Workshops (ICDMW), 2016, pp. 241–246, doi: 10.1109/ICDMW.2016.0041.
14. R. Aggarwal, V. Sounderajah, G. Martin, D. S. Ting, A. Karthikesalingam, D. King, … A. Darzi, “Diagnostic accuracy of deep learning in medical imaging: a systematic review and meta-analysis,” NPJ Digit. Med., vol. 4, no. 1, p. 65, 2021, doi: 10.1038/s41746-021-00438-z.
15. M. Elhoseny and K. Shankar, “Optimal bilateral filter and convolutional neural network based denoising method of medical image measurements,” Measurement, vol. 143, pp. 125–135, 2019, doi: doi:10.1016/j.measurement.2019.04.072.
16. M. Tsuneki, “Deep learning models in medical image analysis,” J. Oral Biosci., vol. 64, no. 3, pp. 312–320, 2022, doi: 10.1016/j.job.2022.03.003.