Izvorni znanstveni članak
https://doi.org/10.1080/00051144.2024.2409552
A wavelet CNN with appropriate feed-allocation and PSO optimized activations for diabetic retinopathy grading
Chandrasekaran Raja
; Department of ECE, Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology, Chennai, India
B V Santhosh Krishna
; Department of Computer Science and Engineering, New Horizon College of Engineering, Bengaluru, India
*
Balaji Loganathan
; Department of ECE, Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology, Chennai, India
Sanjay Kumar Suman
; Department of ECE, St. Martin’s Engineering College, Secunderabad, India
L. Bhagyalakshmi
; Department of ECE, Rajalakshmi Engineering College, Chennai, India
Mubarak Alrashoud
; Department of Software Engineering, College of Computer and Information Sciences, King Saud University, Riyadh, Saudi Arabia
Jayant Giri
; Department of Mechanical Engineering, Yeshwantrao Chavan College of Engineering, Nagpur, India
T. Sathish
; Saveetha School of Engineering, SIMATS, Chennai, India
* Dopisni autor.
Sažetak
This work modifies the architecture of conventional CNN with the integration of Multi-resolution
Analysis (MRA) in a CNN framework for Diabetic Retinopathy (DR) diagnosis and grading. Here,
the HF sub-bands are subjected to optimized activations and are directly fed to the fully connected layers, as it encompasses edge features. Unlike FD-Relu, the proposed function preserves
significant negative coefficients, compared to the S-Relu, the proposed third-order S-Relu is optimized such that it sustains the activations in the range suitable for the wavelet coefficients. The
coefficients of higher-order terms of the proposed 3rd-order S-Relu are optimized with PSO,
fitting the maximum energy of the wavelet sub-bands to ensure High Frequency (HF) edge
preservation. The authors re-architecture 3 different CNNs published in the Retinal Image analysis field, with spatial and wavelet inputs with optimized activations. The highest accuracy of
96% is attained with the AlexNet re-architecture, with 35,126 fundus images secured from the
Kaggle dataset. As we can infer the proposed re-architecture wavelet CNN outperformed the
multiscale shallow CNNs, multiscale attention net, and stacked CNNs with a 6.6, 0.3, 0.7 per cent
increase in accuracy. The entire implementation of the wavelet CNN is made available under
source code.
Ključne riječi
WaveletCNN; activation function; ResNet; AlexNet; wavelet; diabetic retinopathy
Hrčak ID:
326438
URI
Datum izdavanja:
15.10.2024.
Posjeta: 0 *