Abstract:Background Most existing models for CT kernel conversion take images reconstructed with a single predetermined source kernel as input and convert them to images that are reconstructed with a target kernel. However, these models can achieve even better performance if they leverage complementary information obtained from images reconstructed with multiple different kernels. In many clinical practice scenarios, only images with one kernel can be acquired. Purpose We propose a privileged knowledge learning framework that learns privileged knowledge of other source kernels available only in the training data (called privileged information) to guide the conversion from a specific single source kernel to the target kernel, via a joint prediction (JP) task. Methods We construct an ensemble of kernel‐specific (KS) tasks where a KS network (KSNet) takes images reconstructed with a specific source kernel as input and converts them to images reconstructed with the target kernel. Then, a JP task is designed to provide extra regularization, which helps each KSNet learn more informative feature representations for kernel conversion, such as detail and structure representations. Meanwhile, we use a cross‐shaped window‐based attention mechanism in the JP task to highlight the most relevant features to strengthen privileged knowledge learning, thereby alleviating the problems of redundant noise unrelated to images reconstructed with target kernel and inconsistent features that arise from images reconstructed with different kernels. All KSNets can be trained collaboratively by using a JP task to improve the performance of each individual KSNet. Results We extensively evaluate our method on a clinical dataset with scanners from three manufacturers, that is, Siemens, GE and Philips. The experimental results demonstrate that our privileged knowledge learning framework is effective in improving CT kernel conversion. Conclusions Through both quantitative and qualitative research, our privileged knowledge learning framework improves the kernel conversion results, thereby contributing to the improvement of diagnostic accuracy and the advancement of comparative research in quantitative measurements.

Deep Kernel Calibration

CKR-Calibrator: Convolution Kernel Robustness Evaluation and Calibration

Deep Clustered Convolutional Kernels

Improving Deep Neural Network Performance by Integrating Kernelized Min-Max Objective

Kernel Product Neural Networks

How to Scale Up Kernel Methods to Be As Good As Deep Neural Nets

Selective Kernel Networks

Deep Feature Response Discriminative Calibration

Revisit the Essence of Distilling Knowledge Through Calibration

Optimizing Kernel Machines using Deep Learning

Deep $k$-Means: Re-Training and Parameter Sharing with Harder Cluster Assignments for Compressing Deep Convolutions

Deep Neural Networks for Sensor-Based Human Activity Recognition Using Selective Kernel Convolution

Calibrating Deep Neural Network using Euclidean Distance

Deep Kernel Learning via Random Fourier Features

Proximity-Informed Calibration for Deep Neural Networks

Learning with privileged knowledge of multiple kernels via joint prediction for CT Kernel conversion

Discriminative Kernel Feature Extraction and Learning for Object Recognition and Detection.

Calibrating Deep Neural Networks using Focal Loss

Complex Convolution Kernel for Deep Networks

The Deep Kernelized Autoencoder

Calibration Bottleneck: Over-compressed Representations Are Less Calibratable