Abstract: The most common thoracic cage abnormality is pectus excavatum (PE), which is assessed using three CT indices: Haller, correction, and asymmetry. Manually performing this analysis is time-consuming and variable. This paper proposes a fully…
Abstract: AutoAtlas, a novel neural network architecture, partitions and learns 3D brain MRI volumes unsupervised. Two neural networks perform multi-label partitioning based on local texture in the volume and compress the information within each partition…
Abstract: BAA can diagnose endocrine and metabolic disorders in children. Deep learning-based bone age classification methods use global images or local information from bounding boxes or key points. Training with the global image underutilizes discriminative…
Abstract: Deep neural networks increasingly use attention to interpret and improve model performance, but little research has examined how attention progresses to complete a task and whether it is reasonable. We propose an Attention with…
Abstract: This paper generalizes the Attention in Attention (AiA) mechanism in P. Fang et al., 2019 by using explicit mapping in reproducing kernel Hilbert spaces to generate input feature map attention values. Inner and outer…
Abstract: Clinical criteria—bradykinesia, rest tremor, rigidity—diagnose Parkinson’s disease (PD). Clinical rating scales for PD severity are subject to inter-rater variability. This paper proposes a deep learning-based automatic PD diagnosis method using videos for clinical use….
Abstract: Colonoscopy has developed many automated image analysis methods in recent decades. Several recent multi-center clinical trials have tested real-time implementation of the most promising colonoscopy methods. All trials suggest colorectal cancer prevention. We review…
Abstract: Deep learning (DL) requires massive data collection. DL is rarely used in applications without a large dataset. Getting enough data takes time, money, and effort. The scientific community uses synthetic images and real data…
Abstract: Heart failure is difficult to predict. Deep learning models on rich electronic health records may improve prediction but are unexplainable, limiting their use in medicine. A deep-learning framework for accurate and explainable 6-month incident…
Abstract: The Iteratively Optimized Patch Label Inference Network (IOPLIN) is a novel deep learning framework for automatically detecting pavement distresses beyond cracks and potholes. IOPLIN can infer patch labels from pavement images using the Expectation-Maximization…
Abstract: This paper presents an end-to-end neural network model for autonomous driving road detection. We use road boundary and deep convolutional network multi-task learning. To improve performance, we reassign the label and rebalance road pixel…
Abstract: Visual change detection is a fundamental task in computer vision and video analytics. Anomaly detection, object tracking, traffic monitoring, human-machine interaction, behavior analysis, action recognition, and visual surveillance are change detection applications. Change detection…
WhatsApp us