Embedding algorithms are increasingly used to represent clinical concepts in healthcare for improving machine learning tasks such as clinical phenotyping and disease prediction. Recent studies have adapted state-of-the-art bidirectional encoder representations from transformers (BERT) architecture to structured electronic health records (EHR) data for the generation of contextualized concept embeddings, yet do not fully incorporate temporal data across multiple clinical domains. Therefore we developed a new BERT adaptation, CEHR-BERT, to incorporate temporal information using a hybrid approach by augmenting the input to BERT using artificial time tokens, incorporating time, age, and concept embeddings, and introducing a new second learning objective for visit type. CEHR-BERT was trained on a subset of Columbia University Irving Medical Center-York Presbyterian Hospital's clinical data, which includes 2.4M patients, spanning over three decades, and tested using 4-fold cross-validation on the following prediction tasks: hospitalization, death, new heart failure (HF) diagnosis, and HF readmission. Our experiments show that CEHR-BERT outperformed existing state-of-the-art clinical BERT adaptations and baseline models across all 4 prediction tasks in both ROC-AUC and PR-AUC. CEHR-BERT also demonstrated strong transfer learning capability, as our model trained on only 5% of data outperformed comparison models trained on the entire data set. Ablation studies to better understand the contribution of each time component showed incremental gains with every element, suggesting that CEHR-BERT's incorporation of artificial time tokens, time and age embeddings with concept embeddings, and the addition of the second learning objective represents a promising approach for future BERT-based clinical embeddings.
Emerging applications of collaborative autonomy, such as Multi-Target Tracking, Unknown Map Exploration, and Persistent Surveillance, require robots plan paths to navigate an environment while maximizing the information collected via on-board sensors. In this paper, we consider such information acquisition tasks but in adversarial environments, where attacks may temporarily disable the robots' sensors. We propose the first receding horizon algorithm, aiming for robust and adaptive multi-robot planning against any number of attacks, which we call Resilient Active Information acquisitioN (RAIN). RAIN calls, in an online fashion, a Robust Trajectory Planning (RTP) subroutine which plans attack-robust control inputs over a look-ahead planning horizon. We quantify RTP's performance by bounding its suboptimality. We base our theoretical analysis on notions of curvature introduced in combinatorial optimization. We evaluate RAIN in three information acquisition scenarios: Multi-Target Tracking, Occupancy Grid Mapping, and Persistent Surveillance. The scenarios are simulated in C++ and a Unity-based simulator. In all simulations, RAIN runs in real-time, and exhibits superior performance against a state-of-the-art baseline information acquisition algorithm, even in the presence of a high number of attacks. We also demonstrate RAIN's robustness and effectiveness against varying models of attacks (worst-case and random), as well as, varying replanning rates.
An automatic encoder (AE) extreme learning machine (ELM)-AE-ELM model is proposed to predict the NOx emission concentration based on the combination of mutual information algorithm (MI), AE, and ELM. First, the importance of practical variables is computed by the MI algorithm, and the mechanism is analyzed to determine the variables related to the NOx emission concentration. Then, the time delay correlations between the selected variables and NOx emission concentration are further analyzed to reconstruct the modeling data. Subsequently, the AE is applied to extract hidden features within the input variables. Finally, an ELM algorithm establishes the relationship between the NOx emission concentration and deep features. The experimental results on practical data indicate that the proposed model shows promising performance compared to state-of-art models.
This essay is based on the joint project with 111, Inc. The pharmacy e-Commerce business grows rapidly in recent years with the ever-increasing medical demand during the pandemic. A big challenge for online pharmacy platforms is drug product matching. The e-Commerce platform usually collects drug product information from multiple data sources such as the warehouse or retailers. Therefore, the data format is inconsistent, making it hard to identify and match the same drug product. This paper creates an integrated system for matching drug products from two data sources. Besides, the system would correct some inconsistent drug approval numbers based on a Naive-Bayes drug type (Chinese or Non-Chinese Drug) classifier. Our integrated system achieves 98.3% drug matching accuracy, with 99.2% precision and 97.5% recall
In the field of deep-sea exploration, sonar is presently the only efficient long-distance sensing device. The complicated underwater environment, such as noise interference, low target intensity or background dynamics, has brought many negative effects on sonar imaging. Among them, the problem of nonlinear intensity is extremely prevalent. It is also known as the anisotropy of acoustic sensor imaging, that is, when autonomous underwater vehicles (AUVs) carry sonar to detect the same target from different angles, the intensity variation between image pairs is sometimes very large, which makes the traditional matching algorithm almost ineffective. However, image matching is the basis of comprehensive tasks such as navigation, positioning, and mapping. Therefore, it is very valuable to obtain robust and accurate matching results. This paper proposes a combined matching method based on phase information and deep convolution features. It has two outstanding advantages: one is that the deep convolution features could be used to measure the similarity of the local and global positions of the sonar image; the other is that local feature matching could be performed at the key target position of the sonar image. This method does not need complex manual designs, and completes the matching task of nonlinear intensity sonar images in a close end-to-end manner. Feature matching experiments are carried out on the deep-sea sonar images captured by AUVs, and the results show that our proposal has preeminent matching accuracy and robustness.
Despite significant academic and corporate efforts, autonomous driving under adverse visual conditions still proves challenging. As neuromorphic technology has matured, its application to robotics and autonomous vehicle systems has become an area of active research. Low-light and latency-demanding situations can benefit. To enable event cameras to operate alongside staple sensors like lidar in perception tasks, we propose a direct, temporally-decoupled calibration method between event cameras and lidars. The high dynamic range and low-light operation of event cameras are exploited to directly register lidar laser returns, allowing information-based correlation methods to optimize for the 6-DoF extrinsic calibration between the two sensors. This paper presents the first direct calibration method between event cameras and lidars, removing dependencies on frame-based camera intermediaries and/or highly-accurate hand measurements. Code will be made publicly available.
Transformers have made progress in miscellaneous tasks, but suffer from quadratic computational and memory complexities. Recent works propose sparse Transformers with attention on sparse graphs to reduce complexity and remain strong performance. While effective, the crucial parts of how dense a graph needs to be to perform well are not fully explored. In this paper, we propose Normalized Information Payload (NIP), a graph scoring function measuring information transfer on graph, which provides an analysis tool for trade-offs between performance and complexity. Guided by this theoretical analysis, we present Hypercube Transformer, a sparse Transformer that models token interactions in a hypercube and shows comparable or even better results with vanilla Transformer while yielding $O(N\log N)$ complexity with sequence length $N$. Experiments on tasks requiring various sequence lengths lay validation for our graph function well.
Attention-deficit/hyperactivity disorder (ADHD) is a neurodevelopmental disorder that is highly prevalent and requires clinical specialists to diagnose. It is known that an individual's viewing behavior, reflected in their eye movements, is directly related to attentional mechanisms and higher-order cognitive processes. We therefore explore whether ADHD can be detected based on recorded eye movements together with information about the video stimulus in a free-viewing task. To this end, we develop an end-to-end deep learning-based sequence model which we pre-train on a related task for which more data are available. We find that the method is in fact able to detect ADHD and outperforms relevant baselines. We investigate the relevance of the input features in an ablation study. Interestingly, we find that the model's performance is closely related to the content of the video, which provides insights for future experimental designs.
Introducing biomedical informatics (BMI) students to natural language processing (NLP) requires balancing technical depth with practical know-how to address application-focused needs. We developed a set of three activities introducing introductory BMI students to information retrieval with NLP, covering document representation strategies and language models from TF-IDF to BERT. These activities provide students with hands-on experience targeted towards common use cases, and introduce fundamental components of NLP workflows for a wide variety of applications.
Synthesizing natural interactions between virtual humans and their 3D environments is critical for numerous applications, such as computer games and AR/VR experiences. Our goal is to synthesize humans interacting with a given 3D scene controlled by high-level semantic specifications as pairs of action categories and object instances, e.g., "sit on the chair". The key challenge of incorporating interaction semantics into the generation framework is to learn a joint representation that effectively captures heterogeneous information, including human body articulation, 3D object geometry, and the intent of the interaction. To address this challenge, we design a novel transformer-based generative model, in which the articulated 3D human body surface points and 3D objects are jointly encoded in a unified latent space, and the semantics of the interaction between the human and objects are embedded via positional encoding. Furthermore, inspired by the compositional nature of interactions that humans can simultaneously interact with multiple objects, we define interaction semantics as the composition of varying numbers of atomic action-object pairs. Our proposed generative model can naturally incorporate varying numbers of atomic interactions, which enables synthesizing compositional human-scene interactions without requiring composite interaction data. We extend the PROX dataset with interaction semantic labels and scene instance segmentation to evaluate our method and demonstrate that our method can generate realistic human-scene interactions with semantic control. Our perceptual study shows that our synthesized virtual humans can naturally interact with 3D scenes, considerably outperforming existing methods. We name our method COINS, for COmpositional INteraction Synthesis with Semantic Control. Code and data are available at https://github.com/zkf1997/COINS.