In healthcare, applying deep learning models to electronic health records (EHRs) has drawn considerable attention. This sequential nature of EHR data make them wellmatched for the power of Recurrent Neural Network (RNN). In this poster, we propose “Deep Diabetologist” – using RNNs for EHR sequential data modeling to provide personalized hypoglycemic medication prediction for diabetic patients. Our results demonstrate improved performance compared with a baseline classifier using logistic regression.
In the collaborative research to support genomic medicine, we aim to improve the efficiency of operation such as prevention of hereditary cancer syndromes. In the present work, we built a prototype system to record a pedigree chart, clinical and genetic information on individuals during a genetic counseling session. In a mock examination, we were able to draw the pedigree chart of four generations in about four minutes and to record necessary information without disrupting conversation with counselees.
The aim of this study was to to estimate the sample size for the assumed feasibility study of the computer-assisted input support on a clinica trial. More than 1,500 observations were required for the feasibility study with assumed settings. Further study was required for more efficient research design.
Identifying important predicative indicators for prognosis is useful since these factors help for understanding diseases and determining treatments for patients. We extracted important factors for prognosis of cerebral infarction from EHR. We analyzed EHR data of 1,697 patients with 1,602 variables using gradient boosting decision tree. Extracted factors include not only well-known factors such as NIHSS but also new factors such as albumin-globulin ratio.
Jay Patel, Zasim Siddiqui, Anand Krishnan, Thankam Thyvalikakath
1281 - 1281
Smoking is a significant risk factor for initiation and progression of oral diseases. A patient's current smoking status and tobacco dependency can aid clinical decision making and treatment planning. The free-text nature of this data limits accessibility causing obstacles during the time of care and research utility. No studies exist on extracting patient's smoking status automatically from the Electronic Dental Record. This study reports the development and evaluation of an NLP system for this purpose.
Ramón García-Martínez, Sebastián Martins, Santiago Bianco, Hernán Navas
1282 - 1282
The evolution of Medical Data Mining has been possible from use of Health Information Systems. In Argentina, consumption of psychoactive substances is quantified through the National Survey on Prevalence of Psychoactive Substance of the National Institute of Statistics and Census. This paper presents the use of Information Mining Engineering as an alternative approach to identifying novel patterns.
Jose F. Rodriguez, Emilio S. Mole, Luciana Rubin, Sonia E. Benitez, Carlos M. Otero, Daniel R. Luna, Fernan Quiros
1283 - 1283
The infobuttons allows the solving of information needs. In our study, the use of Infobuttons is described, analyzing the number of queries to UpToDate® from the problem list of an Electronic Health Record. There were 26419 requests in 8 months. The highest average use occurred in June. The links to knowledge bases can help to solve information needs, even before they occur.
Continuing education is an imperative for professional nursing. e-Learning is one modality to support education and it has been extensively examined in a nursing academic context. An overview of quantitative, qualitative, and mixed-method systematic reviews were conducted to draw a broad picture of the effects of e-Learning and m-Learning used by registered nurses in a continuing education context.
Phenotypes are defined as observable characteristics of organisms. To facilitate the translation between genotype and phenotype, Human Phenotype Ontology (HPO) was developed as a semantically computable standardized vocabulary to capture phenotypic abnormalities found in human. In this study, we investigated the use of HPO to annotate phenotypic information in clinical domain by leveraging a corpus of 12.8 million clinical notes created from 2010 to 2015 for 729 thousand patients at Mayo Clinic Rochester campus.
A simulation modeling approach was used to determine the optimal human resource solutions for the main functions of the OPD by observing 384 conveniently selected patients at the National Hospital of Sri Lanka. This work seems to be optimized by increasing the morning shift doctors, causing minimal disturbance and inconvenience to the administration and the workers.
Priscyla Waleska Simões, Samuel Cesconetto, Eduardo Daminelli Dalló, Maria Marlene de Souza Pires, Eros Comunello, Felipe Borges Tomaz, Eduardo Pícolo Xavier, Pedro Antonio da Rosa Brunel Alves, Luciane Bisognin Ceretta, Sandra Aparecida Manenti
1287 - 1287
The aim of this study was to use the Data Mining to analyze the profile of the use of contraceptive methods in a university population. We used a database about sexuality performed on a university population in southern Brazil. The results obtained by the generated rules are largely in line with the literature and epidemiology worldwide, showing significant points of vulnerability in the university population. Validation measures of the study, as such, accuracy, sensitivity, specificity, and area under the ROC curve were higher or at least similar as compared to recent studies using the same methodology.
Chemical-induced disease relations (CID) are crucial in various biomedical tasks. In the CID task of Biocreative V, no classifiers with multiple kernels have been developed. In this study, a multiple kernel learning–boosting (MKLB) method is proposed. Different kernel functions according to different types of features were constucted and boosted, each of which were learned with multiple kernels. Our multiple kernel learning–boosting (MKLB) method achieved a F1 score of 0.5068 without incorporating knowledge bases.
Volker S. Thiemann, Tingyan Xu, Rainer Röhrig, Raphael W. Majeed
1289 - 1289
We developed an automated toolchain to generate reports of i2b2 data. It is based on free open source software and runs on a Java Application Server. It is sucessfully used in an ED registry project. The solution is highly configurable and portable to other projects based on i2b2 or compatible factual data sources.
Yue Wang, Zasim Siddiqui, Anand Krishnan, Jay Patel, Thankam Thyvalikakath
1290 - 1290
With an increase in the geriatric population, dental care professionals are presented with older patients who are managing their comorbidities using multiple medications. In this study, we developed a system to extract medication information from electronic dental records (EDRs) and provided patient distribution by the number of medications.
We constructed a novel prognostic model using an innovative method of Bayesian Network (BN) to predict Non-Small Cell Lung Cancer survival status within 5 years after operation in the Asian population. The proposed BN model could present the relationship between prognostic factors and showed the highest performance among other machine learning (ML) algorithms.
Sunmoo Yoon, Faith Parsons, Kevin Sundquist, Jacob Julian, Joseph E. Schwartz, Matthew M. Burg, Karina W. Davidson, Keith M. Diaz
1292 - 1292
To visualize and compare three text analysis algorithms of sentiment (AFINN, Bing, Syuzhet), applied to 1549 ecologically assessed self-report stress notes obtained by smartphone, in order to gain insights about stress measurement and management.
Doctoral dissertations and master's theses of “hepatitis B” (HB) in CNKI database were retrieved to explore the hot spots on HB research field in China, we processed, visualized the data above, and analyzed those data by using co-word analysis informetrics method. Then, the figure for co-occurrence analysis of high-frequency keywords in theses on HB were plotted, which represented eight topics, could help health staff to understand hot topics in this field.
The American Medical Informatics Association (AMIA) established the Nursing Informatics History Project to recognize the pioneers of nursing informatics. Fundamental to the pioneers was dissemination of knowledge. The purpose of this review was to identify pioneers who have embraced social media as of 2016. It is suggested that the pioneers participate in the advancement of nursing informatics via social media.
A virtual reality (VR) based training simulator developed for nasogastric tube (NGT) placement is presented. It leverages the advantages of VR technology – safety, flexibility, interactivity and quantitative assessment – to speed up the learning curve of NGT placement. The simulator demonstrates the potential of VR for nurse education.
This poster describes the rationale and method for the development of a minimum data set as the first step toward an Australian national health information workforce census. Critical importance is attached to open and inclusive models, and the early and extensive engagement of key stakeholders, to provide a solid base for broader research into this workforce.
The quality of morbidity data in multiple routine inpatient records in a sample of South African hospitals is being assessed in terms of data accuracy and completeness. Extensive modification of available data collection tools was required to make it possible to collect the required data for the study.
The China Health and Retirement Longitudinal Study (CHARLS, 2013) data was used to investigate internet use and mobile phone ownership in older Chinese adults and examine digital divide and social economic status and mobile technology adoption and health outcomes associations. Results suggest a significant digital divide associated with not only individual characteristics, but also neighborhood resources. Future eHealth programs should consider the accessibility of mobile tools and develop culturally appropriate programs for different social groups.
Patient handovers are crucial to warrant continuity of care and patient safety. The handoverEHR is an instrument that allows users to depict the main features of a clinical case in cognitive maps. We were interested whether this tool had an effect on the task load. A cross-over study with 30 nursing students was therefore conducted. Mental demand showed a statistical trend to be lower in cognitive map handovers than in the other handover types.
We analyze the deterioration of clinical data quality due to anonymization. The result shows that data quality remained high with micro-aggregation and also verify the availability of noise addition to prevent illegal re-identification by matching another personal data.