Search

In recent years, deep convolutional neural networks (DCNNs) have shown promising performance in medical image analysis, including breast lesion classification in 2D ultrasound (US) images. Despite the outstanding performance of DCNN solutions, explaining their decisions remains an open investigation. Yet, the explainability of DCNN models has become essential for healthcare systems to accept and trust the models. This paper presents a novel framework for explaining DCNN classification decisions of lesions in ultrasound images using the saliency maps linking the DCNN decisions to known cancer characteristics in the medical domain. The proposed framework consists of three main phases. First, DCNN models for classification in ultrasound images are built. Next, selected methods for visualization are applied to obtain saliency maps on the input images of the DCNN models. In the final phase, the visualization outputs and domain-known cancer characteristics are mapped. The paper then demonstrates the use of the framework for breast lesion classification from ultrasound images. We first follow the transfer learning approach and build two DCNN models. We then analyze the visualization outputs of the trained DCNN models using the EGrad-CAM and Ablation-CAM methods. We map the DCNN model decisions of benign and malignant lesions through the visualization outputs to the characteristics such as echogenicity, calcification, shape, and margin. A retrospective dataset of 1298 US images collected from different hospitals is used to evaluate the effectiveness of the framework. The test results show that these characteristics contribute differently to the benign and malignant lesions' decisions. Our study provides the foundation for other researchers to explain the DCNN classification decisions of other cancer types. [ABSTRACT FROM AUTHOR]

The automated analysis of interacting objects or people from video has many uses, including the recognition of activities, and the identification of prototypical or unusual behaviors. Existing techniques generally use temporal sequences of quantifiable real-valued features, such as object position or orientation; however, more recently, qualitative representations have been proposed. In this paper, we present a novel and robust qualitative method, which can be used for both the classification and the clustering of pair-activities. We use qualitative trajectory calculus ( $QTC$ ) to represent the relative motion between two objects and encode their interactions as a trajectory of $QTC$ states. A key element is a general and robust means of determining the sequence similarity, which we term Normalized Weighted Sequence Alignment; we show that this is an effective metric for both recognition and clustering problems. We have evaluated our method across three different data sets, and have shown that it outperforms the state-of-the-art quantitative methods, achieving an error rate of no more than 4.1% for recognition, and cluster purities higher than 90%. Our motivation originates from an interest in automated analysis of animal behaviors, and we present a comprehensive video data set of fish behaviors (Gasterosteus aculeatus), collected from lab-based experiments. [ABSTRACT FROM AUTHOR]