Search

Automatic medical event prediction (MEP), e.g. diagnosis prediction, medication prediction, using electronic health records (EHRs) is a popular research direction in health informatics. In many cases, MEP relies on the determinations from different types of medical events, which demonstrates the heterogeneous nature of EHRs. However, most existing methods for MEP fail to distinguishingly model the type of event that is highly associated with the prediction task, i.e. task-wise event, which usually plays a more significant role than other events. In this paper, we proposed a Long Short-Term Memory network (LSTM)-based method for MEP, named Multi-Channel Fusion LSTM (MCF-LSTM), which models the correlations between different types of medical events using multiple network channels. To this end, we designed a task-wise fusion module, in which a gated network is applied to select how much information can be transferred between events. Furthermore, the irregular temporal interval between adjacent medical visits is also modeled in an individual channel, which is combined with other events in a unified manner. We compared MCF-LSTM with state-of-the-art methods on four MEP tasks on two public datasets: MIMIC-III and eICU. Experimental results show that MCF-LSTM achieves promising results on AUC(receiver operating characteristic curve), AUPR (area under the precision-recall curve), and top-k recall, and outperforms other methods with high stability.

Complex networks as a powerful way to represent complex systems have been widely studied during the past several years. One of the most important tasks of complex network analysis is to detect communities embedded in networks. In the real world, weighted networks are very common and may contain overlapping communities where a node is allowed to belong to multiple communities. In this paper, we propose a novel Bayesian approach, called the Bayesian mixture network (BMN) model, to detect overlapping communities in weighted networks. The advantages of our method are (i) providing soft-partition solutions in weighted networks; (ii) providing soft memberships, which quantify ‘how strongly’ a node belongs to a community. Experiments on a large number of real and synthetic networks show that our model has the ability in detecting overlapping communities in weighted networks and is competitive with other state-of-the-art models at shedding light on community partition.

Complex networks provide a powerful way to represent complex systems and have been widely studied during the past several years. One of the most important tasks of network analysis is to detect structures (also called structural regularities) embedded in networks by determining group number and group partition. Most of network structure exploration models only consider network links. However, in real world networks, nodes may have attributes that are useful for network structure exploration. In this paper, we propose a novel Bayesian nonparametric (BNP) model to explore structural regularities in networks with node attributes, called Bayesian nonparametric attribute (BNPA) model. This model does not only take full advantage of both links between nodes and node attributes for group partition via shared hidden variables, but also determine group number automatically via the Bayesian nonparametric theory. Experiments conducted on a number of real and synthetic networks show that our BNPA model is able to automatically explore structural regularities in networks with node attributes and is competitive with other state-of-the-art models.

In recent years, there have been researches on using semantic knowledge and global statistical features to guide the learning of word embeddings. Though the syntax knowledge also plays an very important role in natural language understanding, its effectiveness on the word embedding learning is still far from well investigated. Inspired by the principle of the dissociation between nouns and verbs (DNV) in language acquisition observed in neuropsychology, we propose a novel model for word embeddings learning using DNV (named Continuous Dissociation between Nouns and Verbs Model, CDNV). CDNV uses a three-layer feed forward neural network to integrate DNV generated by auto-tagged noun/verb information into the word embeddings learning process, which can still preserve the word order of local context. The advantage of the CDNV lies in that it is able to learn high-quality word embeddings with relatively low time complexity. Experimental results show that: (1) CDNV takes about 1.5h to learn word embeddings on a corpus of billions of words, which is comparable with CBOW and Skip-gram and more efficient than other models; (2) the nearest neighbors of some representative words derived from the word embeddings learnt by CDNV are more reasonable than other word embeddings; (3) the performance improvement on F1 measure from CDNV word embeddings is greater than other word embeddings on NER and Chunking.

Display Omitted We proposed a hybrid system to automatically de-identify electronic medical records.PHIs are identified by token-level and character-level conditional random fields.The character-level CRFs is used to avoid boundary errors caused by tokenization.Our system achieves an F-score of 91.24% on 2014 i2b2 corpus, which is top-ranked. De-identification, identifying and removing all protected health information (PHI) present in clinical data including electronic medical records (EMRs), is a critical step in making clinical data publicly available. The 2014 i2b2 (Center of Informatics for Integrating Biology and Bedside) clinical natural language processing (NLP) challenge sets up a track for de-identification (track 1). In this study, we propose a hybrid system based on both machine learning and rule approaches for the de-identification track. In our system, PHI instances are first identified by two (token-level and character-level) conditional random fields (CRFs) and a rule-based classifier, and then are merged by some rules. Experiments conducted on the i2b2 corpus show that our system submitted for the challenge achieves the highest micro F-scores of 94.64%, 91.24% and 91.63% under the "token", "strict" and "relaxed" criteria respectively, which is among top-ranked systems of the 2014 i2b2 challenge. After integrating some refined localization dictionaries, our system is further improved with F-scores of 94.83%, 91.57% and 91.95% under the "token", "strict" and "relaxed" criteria respectively.

Display Omitted We proposed a hybrid system to automatically identify heart disease risk factors.We divided different types of risk factors into three categories according to their descriptions.Our system achieves an F-score of 92.86% on 2014 i2b2 corpus, which is top-ranked. Despite recent progress in prediction and prevention, heart disease remains a leading cause of death. One preliminary step in heart disease prediction and prevention is risk factor identification. Many studies have been proposed to identify risk factors associated with heart disease; however, none have attempted to identify all risk factors. In 2014, the National Center of Informatics for Integrating Biology and Beside (i2b2) issued a clinical natural language processing (NLP) challenge that involved a track (track 2) for identifying heart disease risk factors in clinical texts over time. This track aimed to identify medically relevant information related to heart disease risk and track the progression over sets of longitudinal patient medical records. Identification of tags and attributes associated with disease presence and progression, risk factors, and medications in patient medical history were required. Our participation led to development of a hybrid pipeline system based on both machine learning-based and rule-based approaches. Evaluation using the challenge corpus revealed that our system achieved an F1-score of 92.68%, making it the top-ranked system (without additional annotations) of the 2014 i2b2 clinical NLP challenge.