Pathology reports contain useful information such as the main involved organ, diagnosis, etc. These information can be identified from the free text reports and used for large-scale statistical analysis or serve as annotation for other modalities such as pathology slides images. However, manual classification for a huge number of reports on multiple tasks is labor-intensive. In this paper, we have developed an automatic text classifier based on BERT and we propose a human-centric metric to evaluate the model. According to the model confidence, we identify low-confidence cases that require further expert annotation and high-confidence cases that are automatically classified. We report the percentage of low-confidence cases and the performance of automatically classified cases. On the high-confidence cases, the model achieves classification accuracy comparable to pathologists. This leads a potential of reducing 80% to 98% of the manual annotation workload.

Clinical notes contain a large amount of clinically valuable information that is ignored in many clinical decision support systems due to the difficulty that comes with mining that information. Recent work has found success leveraging deep learning models for the prediction of clinical outcomes using clinical notes. However, these models fail to provide clinically relevant and interpretable information that clinicians can utilize for informed clinical care. In this work, we augment a popular convolutional model with an attention mechanism and apply it to unstructured clinical notes for the prediction of ICU readmission and mortality. We find that the addition of the attention mechanism leads to competitive performance while allowing for the straightforward interpretation of predictions. We develop clear visualizations to present important spans of text for both individual predictions and high-risk cohorts. We then conduct a qualitative analysis and demonstrate that our model is consistently attending to clinically meaningful portions of the narrative for all of the outcomes that we explore.

The enormous amount of data and computation required to train DNNs have led to the rise of various parallelization strategies. Broadly, there are two strategies: 1) Data-Parallelism -- replicating the DNN on multiple processes and training on different training samples, and 2) Model-Parallelism -- dividing elements of the DNN itself into partitions across different processes. While data-parallelism has been extensively studied and developed, model-parallelism has received less attention as it is non-trivial to split the model across processes. In this paper, we propose HyPar-Flow: a framework for scalable and user-transparent parallel training of very large DNNs (up to 5,000 layers). We exploit TensorFlow's Eager Execution features and Keras APIs for model definition and distribution. HyPar-Flow exposes a simple API to offer data, model, and hybrid (model + data) parallel training for models defined using the Keras API. Under the hood, we introduce MPI communication primitives like send and recv on layer boundaries for data exchange between model-partitions and allreduce for gradient exchange across model-replicas. Our proposed designs in HyPar-Flow offer up to 3.1x speedup over sequential training for ResNet-110 and up to 1.6x speedup over Horovod-based data-parallel training for ResNet-1001; a model that has 1,001 layers and 30 million parameters. We provide an in-depth performance characterization of the HyPar-Flow framework on multiple HPC systems with diverse CPU architectures including Intel Xeon(s) and AMD EPYC. HyPar-Flow provides 110x speed up on 128 nodes of the Stampede2 cluster at TACC for hybrid-parallel training of ResNet-1001.

The advent of the Internet era has led to an explosive growth in the Electronic Health Records (EHR) in the past decades. The EHR data can be regarded as a collection of clinical events, including laboratory results, medication records, physiological indicators, etc, which can be used for clinical outcome prediction tasks to support constructions of intelligent health systems. Learning patient representation from these clinical events for the clinical outcome prediction is an important but challenging step. Most related studies transform EHR data of a patient into a sequence of clinical events in temporal order and then use sequential models to learn patient representations for outcome prediction. However, clinical event sequence contains thousands of event types and temporal dependencies. We further make an observation that clinical events occurring in a short period are not constrained by any temporal order but events in a long term are influenced by temporal dependencies. The multi-scale temporal property makes it difficult for traditional sequential models to capture the short-term co-occurrence and the long-term temporal dependencies in clinical event sequences. In response to the above challenges, this paper proposes a Multi-level Representation Model (MRM). MRM first uses a sparse attention mechanism to model the short-term co-occurrence, then uses interval-based event pooling to remove redundant information and reduce sequence length and finally predicts clinical outcomes through Long Short-Term Memory (LSTM). Experiments on real-world datasets indicate that our proposed model largely improves the performance of clinical outcome prediction tasks using EHR data.

Requests come in a variety of size and constraints, from the urgent monitoring of small areas to large area coverage. In this work we are particularly interested in the latter case, with requests covering whole countries or even continents. Depending on weather conditions, such requests may take several months to complete, even with multiple satellites. In order to shorten the time required to fulfill requests, the mission orchestrator shall schedule acquisitions with both a short and a long-term strategy. Determining a strategy robust to an uncertain environment is a complex task, this is why current solutions mainly consist of heuristics configured by human-experts. This paper demonstrates that Reinforcement Learning (RL) might be well-suited for such a challenge. RL has proven to be of great value since these algorithms have mastered several games such as Pong on Atari 2600 (Mnih et al. 2013), Go with AlphaGo (Silver et al. 2017) and more recently Starcraft (Arulkumaran, Cully, and Togelius 2019).c

Neural network pruning techniques have demonstrated it is possible to remove the majority of weights in a network with surprisingly little degradation to test set accuracy. However, this measure of performance conceals significant differences in how different classes and images are impacted by pruning. We find that certain examples, which we term pruning identified exemplars (PIEs), and classes are systematically more impacted by the introduction of sparsity. Removing PIE images from the test-set greatly improves top-1 accuracy for both pruned and non-pruned models. These hard-to-generalize-to images tend to be mislabelled, of lower image quality, depict multiple objects or require fine-grained classification. These findings shed light on previously unknown trade-offs, and suggest that a high degree of caution should be exercised before pruning is used in sensitive domains.

Robustness to capitalization errors is a highly desirable characteristic of named entity recognizers, yet we find standard models for the task are surprisingly brittle to such noise. Existing methods to improve robustness to the noise completely discard given orthographic information, mwhich significantly degrades their performance on well-formed text. We propose a simple alternative approach based on data augmentation, which allows the model to \emph{learn} to utilize or ignore orthographic information depending on its usefulness in the context. It achieves competitive robustness to capitalization errors while making negligible compromise to its performance on well-formed text and significantly improving generalization power on noisy user-generated text. Our experiments clearly and consistently validate our claim across different types of machine learning models, languages, and dataset sizes.

Charge prediction, determining charges for criminal cases by analyzing the textual fact descriptions, is a promising technology in legal assistant systems. In practice, the fact descriptions could exhibit a significant intra-class variation due to factors like nonnormative use of language, which makes the prediction task very challenging, especially for charge classes with too few samples to cover the expression variation. In this work, we explore to use the charge definitions from criminal law to alleviate this issue. The key idea is that the expressions in a fact description should have corresponding formal terms in charge definitions, and those terms are shared across classes and could account for the diversity in the fact descriptions. Thus, we propose to create auxiliary fact representations from charge definitions to augment fact descriptions representation. The generated auxiliary representations are created through the interaction of fact description with the relevant charge definitions and terms in those definitions by integrated sentence-and word-level attention scheme. Experimental results on two datasets show that our model achieves significant improvement than baselines, especially for classes with few samples. Introduction The task of charge prediction is to determine appropriate charges, such as theft, seizing or robbery, for criminal cases by analyzing the textual fact descriptions.

Task oriented language understanding in dialog systems is often modeled using intents (task of a query) and slots (parameters for that task). Intent detection and slot tagging are, in turn, modeled using sentence classification and word tagging techniques respectively. Similar to adversarial attack problems with computer vision models discussed in existing literature, these intent-slot tagging models are often over-sensitive to small variations in input -- predicting different and often incorrect labels when small changes are made to a query, thus reducing their accuracy and reliability. However, evaluating a model's robustness to these changes is harder for language since words are discrete and an automated change (e.g. adding `noise') to a query sometimes changes the meaning and thus labels of a query. In this paper, we first describe how to create an adversarial test set to measure the robustness of these models. Furthermore, we introduce and adapt adversarial training methods as well as data augmentation using back-translation to mitigate these issues. Our experiments show that both techniques improve the robustness of the system substantially and can be combined to yield the best results.

We use existing edge detection methods to construct simple edge diagrams, train a generative model to convert them into synthetic medical images, and construct a dataset of synthetic images with known segmentations using variations on extracted edge diagrams. This synthetic dataset is then used to train a supervised image segmentation model. We test our approach on a clinical dataset of kidney ultrasound images and the benchmark ISIC 2018 skin lesion dataset. We show that our unsupervised approach is more accurate than previous unsupervised methods, and performs reasonably compared to supervised image segmentation models. All code and trained models are available at https://github.com/kiretd/Unsupervised-MIseg . 1 Introduction In vivo medical imaging is one of the primary technologies available for clinical evaluation, diagnosis, and treatment planning.