3D G-CNNs for Pulmonary Nodule Detection

Abstract

Convolutional Neural Networks (CNNs) require a large amount of annotated data to learn from, which is often difficult to obtain in the medical domain. In this paper we show that the sample complexity of CNNs can be significantly improved by using 3D roto-translation group convolutions (G-Convs) instead of the more conventional translational convolutions. These 3D G-CNNs were applied to the problem of false positive reduction for pulmonary nodule detection, and proved to be substantially more effective in terms of performance, sensitivity to malignant nodules, and speed of convergence compared to a strong and comparable baseline architecture with regular convolutions, data augmentation and a similar number of parameters. For every dataset size tested, the G-CNN achieved a FROC score close to the CNN trained on ten times more data.

References in corpus (6)

• Adam: A Method for Stochastic Optimization
• Validation, comparison, and combination of algorithms for automatic detection of pulmonary nodules in computed tomography images: the LUNA16 challenge
• On the Generalization of Equivariance and Convolution in Neural Networks to the Action of Compact Groups
• Spherical CNNs
• Covariant Compositional Networks For Learning Graphs
• Intertwiners between Induced Representations (with Applications to the Theory of Equivariant Neural Networks)

Cited by in corpus (30)

• SE(3)-Transformers: 3D Roto-Translation Equivariant Attention Networks
• 3D Steerable CNNs: Learning Rotationally Equivariant Features in Volumetric Data
• Gauge Equivariant Convolutional Networks and the Icosahedral CNN
• PDE-based Group Equivariant Convolutional Neural Networks
• Three-dimensional microstructure generation using generative adversarial neural networks in the context of continuum micromechanics
• On Learning Sets of Symmetric Elements
• MDP Homomorphic Networks: Group Symmetries in Reinforcement Learning
• Gauge Equivariant Mesh CNNs: Anisotropic convolutions on geometric graphs
• A General Theory of Equivariant CNNs on Homogeneous Spaces
• Attentive Group Equivariant Convolutional Networks
• Neural Networks Enforcing Physical Symmetries in Nonlinear Dynamical Lattices: The Case Example of the Ablowitz-Ladik Model
• Theoretical Aspects of Group Equivariant Neural Networks
• B-Spline CNNs on Lie Groups
• Rotation-Equivariant Deep Learning for Diffusion MRI
• A Data and Compute Efficient Design for Limited-Resources Deep Learning
• Covariance in Physics and Convolutional Neural Networks
• Geometric and Physical Quantities Improve E(3) Equivariant Message Passing
• Scaling-Translation-Equivariant Networks with Decomposed Convolutional Filters
• SE(3)-Equivariant and Noise-Invariant 3D Rigid Motion Tracking in Brain MRI
• Provably Strict Generalisation Benefit for Equivariant Models
• Spin-Weighted Spherical CNNs
• Coordinate Independent Convolutional Networks -- Isometry and Gauge Equivariant Convolutions on Riemannian Manifolds
• Rotational 3D Texture Classification Using Group Equivariant CNNs
• Understanding the Generalization Benefit of Model Invariance from a Data Perspective
• Deep Learning for 2D and 3D Rotatable Data: An Overview of Methods
• Affine Self Convolution
• Finite Group Equivariant Neural Networks for Games
• Toward Building Science Discovery Machines
• Provably Strict Generalisation Benefit for Invariance in Kernel Methods
• Exploiting Redundancy: Separable Group Convolutional Networks on Lie Groups