Abstract:Background Deep learning technology has been widely applied to medical image analysis. But due to the limitations of its own imaging principle, ultrasound image has the disadvantages of low resolution and high Speckle Noise density, which not only hinder the diagnosis of patients’ conditions but also affect the extraction of ultrasound image features by computer technology. Objective In this study, we investigate the robustness of deep convolutional neural network (CNN) for classification, segmentation, and target detection of breast ultrasound image through random Salt & Pepper Noise and Gaussian Noise. Methods We trained and validated 9 CNN architectures in 8617 breast ultrasound images, but tested the models with noisy test set. Then, we trained and validated 9 CNN architectures with different levels of noise in these breast ultrasound images, and tested the models with noisy test set. Diseases of each breast ultrasound image in our dataset were annotated and voted by three sonographers based on their malignancy suspiciousness. we use evaluation indexes to evaluate the robustness of the neural network algorithm respectively. Results There is a moderate to high impact (The accuracy of the model decreased by about 5%-40%) on model accuracy when Salt and Pepper Noise, Speckle Noise, or Gaussian Noise is introduced to the images respectively. Consequently, DenseNet, UNet++ and Yolov5 were selected as the most robust model based on the selected index. When any two of these three kinds of noise are introduced into the image at the same time, the accuracy of the model will be greatly affected. Conclusions Our experimental results reveal new insights: The variation trend of accuracy with the noise level in Each network used for classification tasks and object detection tasks has some unique characteristics. This finding provides us with a method to reveal the black-box architecture of computer-aided diagnosis (CAD) systems. On the other hand, the purpose of this study is to explore the impact of adding noise directly to the image on the performance of neural networks, which is different from the existing articles on robustness in the field of medical image processing. Consequently, it provides a new way to evaluate the robustness of CAD systems in the future.

Algorithm and Application of a Noise-Tolerant Neural Network

Noise is the Fatal Poison: A Noise-aware Network for Noisy Dataset Classification

Activity-dependent Neuron Model for Noise Resistance

A Noise-Sensitivity-Analysis-Based Test Prioritization Technique for Deep Neural Networks

Noise Optimization in Artificial Neural Networks

Robust Stability and Robust Periodicity of Delayed Recurrent Neural Networks with Noise Disturbance

Exploiting Noise As a Resource for Computation and Learning in Spiking Neural Networks

Robustness Enhancement in Neural Networks with Alpha-Stable Training Noise

A Detection Algorithm of Robust Neural Network for Heavy-tailed Noise

Analysis of Winning Experience and Technical Training Effect of Badminton Match Based on BP Neural Network

Enhance DNN Adversarial Robustness and Efficiency via Injecting Noise to Non-Essential Neurons

Training Robust Deep Neural Networks via Adversarial Noise Propagation

Noise-robustness test for ultrasound breast nodule neural network models as medical devices

Analyzing the Noise Robustness of Deep Neural Networks

Towards Understanding the Importance of Noise in Training Neural Networks

Noise-Resilient Designs for Optical Neural Networks

Parametric Noise Injection: Trainable Randomness to Improve Deep Neural Network Robustness against Adversarial Attack

A New BP Neural Network Model Based on the Random Fuzzy Theory

Research on Application of Artificial Intelligence in Computer Network Technology

Robustness of Physics-Informed Neural Networks to Noise in Sensor Data

A noise based novel strategy for faster SNN training