Curriculum learning for improved femur fracture classification: Scheduling data with prior knowledge and uncertainty

•We identify common curriculum learning elements among different data scheduling strategies, and present them within a unified formulation.•We propose two types of novel ranking functions guiding the prioritization of the training data.•We leverage domain-specific clinical knowledge to define the first scoring function.•In absence of domain knowledge, we propose to estimate the ranking of the training samples by dynamically quantifying the uncertainty of the model predictions.•We validate our strategies on a clinical dataset for the multi-class classification of proximal femur fractures.•With a controlled experimental setting, we confirm that our method is useful in reducing the classification error under limited amounts of data, imbalance in the class distribution, and unreliable annotations. We give recommendations about the best approaches for each scenario. [Display omitted] An adequate classification of proximal femur fractures from X-ray images is crucial for the treatment choice and the patients’ clinical outcome. We rely on the commonly used AO system, which describes a hierarchical knowledge tree classifying the images into types and subtypes according to the fracture’s location and complexity. In this paper, we propose a method for the automatic classification of proximal femur fractures into 3 and 7 AO classes based on a Convolutional Neural Network (CNN). As it is known, CNNs need large and representative datasets with reliable labels, which are hard to collect for the application at hand. In this paper, we design a curriculum learning (CL) approach that improves over the basic CNNs performance under such conditions. Our novel formulation reunites three curriculum strategies: individually weighting training samples, reordering the training set, and sampling subsets of data. The core of these strategies is a scoring function ranking the training samples. We define two novel scoring functions: one from domain-specific prior knowledge and an original self-paced uncertainty score. We perform experiments on a clinical dataset of proximal femur radiographs. The curriculum improves proximal femur fracture classification up to the performance of experienced trauma surgeons. The best curriculum method reorders the training set based on prior knowledge resulting into a classification improvement of 15%. Using the publicly available MNIST dataset, we further discuss and demonstrate the benefits of our unified CL formulation for three controlled and challe