Quadratic Autoencoder (Q-AE) for Low-dose CT Denoising

2019/01/17 by Fenglei Fan, Fan, Fenglei, Hongming Shan +15 · 4 citations
Computer Science · Engineering · Mathematics · Medicine · #Advanced X-ray and CT Imaging #Algorithm #Artificial intelligence #Artificial neural network #Autoencoder #Computer science #Deep learning #FOS: Computer and information sciences #FOS: Electrical engineering #Interpretability #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Mathematical optimization #Mathematics #Medical Imaging Techniques and Applications #Noise reduction #Pattern recognition (psychology) #Quadratic equation #Quadratic function #Quadratic programming #Radiomics and Machine Learning in Medical Imaging #Representation (politics) #Signal Processing (eess.SP) #cs.LG #eess.SP #electronic engineering #information engineering #stat.ML

paper · pdf · doi:10.48550/arxiv.1901.05593

published in arXiv (Cornell University) (Cornell University)

openalex publication_date 2019/01/17 · arxiv created 2019/10/30 · arxiv updated 2019/11/01 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/08

Abstract

Inspired by complexity and diversity of biological neurons, our group proposed quadratic neurons by replacing the inner product in current artificial neurons with a quadratic operation on input data, thereby enhancing the capability of an individual neuron. Along this direction, we are motivated to evaluate the power of quadratic neurons in popular network architectures, simulating human-like learning in the form of quadratic-neuron-based deep learning. Our prior theoretical studies have shown important merits of quadratic neurons and networks in representation, efficiency, and interpretability. In this paper, we use quadratic neurons to construct an encoder-decoder structure, referred as the quadratic autoencoder, and apply it to low-dose CT denoising. The experimental results on the Mayo low-dose CT dataset demonstrate the utility of quadratic autoencoder in terms of image denoising and model efficiency. To our best knowledge, this is the first time that the deep learning approach is implemented with a new type of neurons and demonstrates a significant potential in the medical imaging field.

Citations

• Auto-Encoding Variational Bayes
• Penalized weighted least-squares image reconstruction for dual energy X-ray transmission tomography
• rm K-SVD: An Algorithm for Designing Overcomplete Dictionaries for Sparse Representation
• ImageNet classification with deep convolutional neural networks
• Deep Convolutional Neural Network for Inverse Problems in Imaging

Cited by

• PatchDenoiser: Parameter-efficient multi-scale patch learning and fusion denoiser for Low-dose CT imaging
• Low-Dose CT Imaging Using a Regularization-Enhanced Efficient Diffusion Probabilistic Model
• Poisson Flow Consistency Training
• MAN: Latent Diffusion Enhanced Multistage Anti-Noise Network for Efficient and High-Quality Low-Dose CT Image Denoising

Related

• On Expressivity and Trainability of Quadratic Networks
• Computational and Storage Efficient Quadratic Neurons for Deep Neural Networks
• QuadEnhancer: Leveraging Quadratic Transformations to Enhance Deep Neural Networks
• Universal Approximation with Quadratic Deep Networks
• Deep Learning for Low-Dose CT Denoising