Abstract: Diabetes ICU patients have a higher risk of complications and in-hospital death. Due to many factors, estimating death risk is difficult and time-consuming. Healthcare providers want to identify high-risk ICU patients to reduce risk…
Abstract: Due to its many real-world applications—evacuation situation analysis, intelligent transport system deployment, traffic operations—human trajectory forecasting has been a field of active research for decades. We model human trajectory forecasting as learning a representation…
Abstract: With the development of various sensors in smartphones or wearable devices, human activity recognition (HAR) has been widely researched and has many applications in healthcare, smart cities, etc. Sensor-based HAR uses many hand-crafted feature…
Abstract: Smart traffic management systems forecast traffic. Missing data issues hinder urban road network traffic flow forecasting due to the widespread use of massive vehicle trajectory data. This paper investigates urban network-wide short-term traffic speed…
Abstract: Human-centric computer vision tasks require pedestrian detection, which is difficult. Handcrafted and deep features have improved over the past decade. We review recent pedestrian detection advances. We first review single-spectral pedestrian detection using handcrafted…
Abstract: Due to its importance in airline and airport efficiency, flight delay prediction has become popular. Most predictions use the single-airport scenario, which ignores airport networks’ time-varying spatial interactions. This paper examines flight delay prediction…
Abstract: Traffic flow prediction is essential to intelligent transportation systems and various traffic-related applications. Urban residents depend on fixed-route buses, which creates latent travel regularity. This fixed mobility mode hides human mobility patterns, particularly bus…
Abstract: Deep learning has been successful in classifying the label and severity stage of certain diseases, but few of them explain how to make predictions. We propose to use deep learning in medical diagnosis to…
Abstract: Objective. Intracranial pathologies are treated using mean intracranial pressure (ICP). The ICP pulse waveform, the shape of the ICP signal over one cardiac cycle, also indicates craniospinal space state. This study proposed an end-to-end…
Abstract: This paper trains a deep convolutional neural network with low-bitwidth weights and activations. The quantizer’s non-differentiability makes optimizing a low-precision network difficult, resulting in accuracy loss. Progressive quantization, stochastic precision, and joint knowledge distillation…
Abstract: AI healthcare applications use sensitive EHRs that are rarely labeled and distributed across symbiont institutions. Such data makes machine learning model training difficult. To address these challenges, we propose dynamic neural graphs-based federated learning…
Abstract: This paper uses Deep Learning (DL) on smartphone tri-axial accelerometer signals to identify and verify drivers. ResNet-50 followed by two SGRUs is the driver identification architecture. Shortcut connections allow ResNet to extract rich accelerometer…
WhatsApp us