Medicine and Health Sciences Commons^™

Enhancing Clinical Concept Extraction With Contextual Embeddings., Yuqi Si, Jingqi Wang, Hua Xu, Kirk Roberts

Journal Articles

OBJECTIVE: Neural network-based representations ("embeddings") have dramatically advanced natural language processing (NLP) tasks, including clinical NLP tasks such as concept extraction. Recently, however, more advanced embedding methods and representations (eg, ELMo, BERT) have further pushed the state of the art in NLP, yet there are no common best practices for how to integrate these representations into clinical tasks. The purpose of this study, then, is to explore the space of possible options in utilizing these new models for clinical concept extraction, including comparing these to traditional word embedding methods (word2vec, GloVe, fastText).

MATERIALS AND METHODS: Both off-the-shelf, open-domain embeddings and …

Exposures At Day Labor Corners: Using Existing Georeferenced Data To Describe Features Of Urban Environments, Maria Eugenia Fernández-Esquer, Amy E Hughes, Sandi L Pruitt

Journal Articles

PURPOSE: Latino day laborers are male immigrants from mainly Mexico and Central America who congregate at corners, that is, informal hiring sites, to solicit short-term employment. Studies describing the occupational environment of Latino day laborers traditionally measure jobsite exposures, not corner exposures. We sought to elucidate exposures at corners by describing their demographic, socioeconomic, occupational, business, built, and physical environmental characteristics and by comparing corner characteristics with other locations in a large urban county in Texas.

METHODS: We used multiple publicly available data sets from the U.S. Census, local tax authority, Google's Nearby Places Application Programming Interface, and Environmental Protection …

Ankle And Midtarsal Joint Quasi-Stiffness During Walking With Added Mass, Andrew M. Kern, Nikolaos Papachatzis, Jeffrey M. Patterson, Dustin A. Bruening, Kota Z. Takahashi

Journal Articles

Examination of how the ankle and midtarsal joints modulate stiffness in response to increased force demand will aid understanding of overall limb function and inform the development of bio-inspired assistive and robotic devices. The purpose of this study is to identify how ankle and midtarsal joint quasi-stiffness are affected by added body mass during over-ground walking. Healthy participants walked barefoot over-ground at 1.25 m/s wearing a weighted vest with 0%, 15% and 30% additional body mass. The effect of added mass was investigated on ankle and midtarsal joint range of motion (ROM), peak moment and quasi-stiffness. Joint quasi-stiffness was broken …

Unraveling The Functional Role Of The Orphan Solute Carrier, Slc22a24 In The Transport Of Steroid Conjugates Through Metabolomic And Genome-Wide Association Studies, Sook Wah Yee, Adrian Stecula, Huan-Chieh Chien, Ling Zou, Elena V Feofanova, Marjolein Van Borselen, Kit Wun Kathy Cheung, Noha A Yousri, Karsten Suhre, Jason M Kinchen, Eric Boerwinkle, Roshanak Irannejad, Bing Yu, Kathleen M Giacomini

Journal Articles

Variation in steroid hormone levels has wide implications for health and disease. The genes encoding the proteins involved in steroid disposition represent key determinants of interindividual variation in steroid levels and ultimately, their effects. Beginning with metabolomic data from genome-wide association studies (GWAS), we observed that genetic variants in the orphan transporter, SLC22A24 were significantly associated with levels of androsterone glucuronide and etiocholanolone glucuronide (sentinel SNPs p-value

Pm, Amal Rammah, Kristina W Whitworth, Inkyu Han, Wenyaw Chan, Elaine Symanski

Journal Articles

There is limited evidence on the role of exposure to chemical constituents of fine particulate matter (PM

Deep Patient Representation Of Clinical Notes Via Multi-Task Learning For Mortality Prediction., Yuqi Si, Kirk Roberts

Journal Articles

We propose a deep learning-based multi-task learning (MTL) architecture focusing on patient mortality predictions from clinical notes. The MTL framework enables the model to learn a patient representation that generalizes to a variety of clinical prediction tasks. Moreover, we demonstrate how MTL enables small but consistent gains on a single classification task (e.g., in-hospital mortality prediction) simply by incorporating related tasks (e.g., 30-day and 1-year mortality prediction) into the MTL framework. To accomplish this, we utilize a multi-level Convolutional Neural Network (CNN) associated with a MTL loss component. The model is evaluated with 3, 5, and 20 tasks and is …