Your browser doesn't support javascript.
loading
Multi-Scale-Denoising Residual Convolutional Network for Retinal Disease Classification Using OCT.
Peng, Jinbo; Lu, Jinling; Zhuo, Junjie; Li, Pengcheng.
Afiliación
  • Peng J; State Key Laboratory of Digital Medical Engineering, School of Biomedical Engineering, Hainan University, Haiko 570228, China.
  • Lu J; Key Laboratory of Biomedical Engineering of Hainan Province, One Health Institute, Hainan University, Haiko 570228, China.
  • Zhuo J; Research Unit of Multimodal Cross Scale Neural Signal Detection and Imaging, Chinese Academy of Medical Science, HUST-Suzhou Institute for Brainsmatics, Jiangsu Industrial Technology Research Institute (JITRI), Suzhou 215100, China.
  • Li P; State Key Laboratory of Digital Medical Engineering, School of Biomedical Engineering, Hainan University, Haiko 570228, China.
Sensors (Basel) ; 24(1)2023 Dec 27.
Article en En | MEDLINE | ID: mdl-38203011
ABSTRACT
Macular pathologies can cause significant vision loss. Optical coherence tomography (OCT) images of the retina can assist ophthalmologists in diagnosing macular diseases. Traditional deep learning networks for retinal disease classification cannot extract discriminative features under strong noise conditions in OCT images. To address this issue, we propose a multi-scale-denoising residual convolutional network (MS-DRCN) for classifying retinal diseases. Specifically, the MS-DRCN includes a soft-denoising block (SDB), a multi-scale context block (MCB), and a feature fusion block (FFB). The SDB can determine the threshold for soft thresholding automatically, which removes speckle noise features efficiently. The MCB is designed to capture multi-scale context information and strengthen extracted features. The FFB is dedicated to integrating high-resolution and low-resolution features to precisely identify variable lesion areas. Our approach achieved classification accuracies of 96.4% and 96.5% on the OCT2017 and OCT-C4 public datasets, respectively, outperforming other classification methods. To evaluate the robustness of our method, we introduced Gaussian noise and speckle noise with varying PSNRs into the test set of the OCT2017 dataset. The results of our anti-noise experiments demonstrate that our approach exhibits superior robustness compared with other methods, yielding accuracy improvements ranging from 0.6% to 2.9% when compared with ResNet under various PSNR noise conditions.
Asunto(s)
Palabras clave

Texto completo: 1 Colección: 01-internacional Base de datos: MEDLINE Asunto principal: Enfermedades de la Retina / Tomografía de Coherencia Óptica Tipo de estudio: Prognostic_studies Límite: Humans Idioma: En Revista: Sensors (Basel) Año: 2023 Tipo del documento: Article País de afiliación: China Pais de publicación: Suiza

Texto completo: 1 Colección: 01-internacional Base de datos: MEDLINE Asunto principal: Enfermedades de la Retina / Tomografía de Coherencia Óptica Tipo de estudio: Prognostic_studies Límite: Humans Idioma: En Revista: Sensors (Basel) Año: 2023 Tipo del documento: Article País de afiliación: China Pais de publicación: Suiza