Neural Network-Based Prediction of Perceived Sleep Quality Through Wearable Device Data.

BACKGROUND: This study focuses on the development of a neural network model to predict perceived sleep quality using data from wearable devices. We collected various physiological metrics from 18 participants over four weeks, including heart rate, physical activity, and both device-measured and self-reported sleep quality. OBJECTIVES: The primary objective was to correlate wearable device data with subjective sleep quality perceptions. METHODS: Our approach used data processing, feature engineering, and optimizing a Multi-Layer Perceptron classifier. RESULTS: Despite comprehensive data analysis and model experimentation, the predictive accuracy for perceived sleep quality was moderate (59%), highlighting the complexities in accurately quantifying subjective sleep experiences through wearable data. Applying a tolerance of 1 grade (on a scale from 1-5), increased accuracy to 92%. DISCUSSION: More in-depth analysis is required to fully comprehend how wearables and artificial intelligence might assist in understanding sleep behavior.

Personalised prediction of maintenance dialysis initiation in patients with chronic kidney disease stages 3-5: a multicentre study using the machine learning approach.

BACKGROUND: Optimal timing for initiating maintenance dialysis in patients with chronic kidney disease (CKD) stages 3-5 is challenging. This study aimed to develop and validate a machine learning (ML) model for early personalised prediction of maintenance dialysis initiation within 1-year and 3-year timeframes among patients with CKD stages 3-5. METHODS: Retrospective electronic health record data from the Taipei Medical University clinical research database were used. Newly diagnosed patients with CKD stages 3-5 between 2008 and 2017 were identified. The observation period spanned from the diagnosis of CKD stages 3-5 until the maintenance dialysis initiation or a maximum follow-up of 3 years. Predictive models were developed using patient demographics, comorbidities, laboratory data and medications. The dataset was divided into training and testing sets to ensure robust model performance. Model evaluation metrics, including area under the curve (AUC), sensitivity, specificity, positive predictive value, negative predictive value and F1 score, were employed. RESULTS: A total of 6123 and 5279 patients were included for 1 year and 3 years of the model development. The artificial neural network demonstrated better performance in predicting maintenance dialysis initiation within 1 year and 3 years, with AUC values of 0.96 and 0.92, respectively. Important features such as baseline estimated glomerular filtration rate and albuminuria significantly contributed to the predictive model. CONCLUSION: This study demonstrates the efficacy of an ML approach in developing a highly predictive model for estimating the timing of maintenance dialysis initiation in patients with CKD stages 3-5. These findings have important implications for personalised treatment strategies, enabling improved clinical decision-making and potentially enhancing patient outcomes.

Seamless EMR data access: Integrated governance, digital health and the OMOP-CDM.

Objectives In this overview, we describe theObservational Medical Outcomes Partnership Common Data Model (OMOP-CDM), the established governance processes employed in EMR data repositories, and demonstrate how OMOP transformed data provides a lever for more efficient and secure access to electronic medical record (EMR) data by health service providers and researchers.Methods Through pseudonymisation and common data quality assessments, the OMOP-CDM provides a robust framework for converting complex EMR data into a standardised format. This allows for the creation of shared end-to-end analysis packages without the need for direct data exchange, thereby enhancing data security and privacy. By securely sharing de-identified and aggregated data and conducting analyses across multiple OMOP-converted databases, patient-level data is securely firewalled within its respective local site.Results By simplifying data management processes and governance, and through the promotion of interoperability, the OMOP-CDM supports a wide range of clinical, epidemiological, and translational research projects, as well as health service operational reporting.Discussion Adoption of the OMOP-CDM internationally and locally enables conversion of vast amounts of complex, and heterogeneous EMR data into a standardised structured data model, simplifies governance processes, and facilitates rapid repeatable cross-institution analysis through shared end-to-end analysis packages, without the sharing of data.Conclusion The adoption of the OMOP-CDM has the potential to transform health data analytics by providing a common platform for analysing EMR data across diverse healthcare settings.

Development of a natural language processing model for deriving breast cancer quality indicators : A cross-sectional, multicenter study.

OBJECTIVES: Medico-administrative data are promising to automate the calculation of Healthcare Quality and Safety Indicators. Nevertheless, not all relevant indicators can be calculated with this data alone. Our feasibility study objective is to analyze 1) the availability of data sources; 2) the availability of each indicator elementary variables, and 3) to apply natural language processing to automatically retrieve such information. METHOD: We performed a multicenter cross-sectional observational feasibility study on the clinical data warehouse of Assistance Publique - Hôpitaux de Paris (AP-HP). We studied the management of breast cancer patients treated at AP-HP between January 2019 and June 2021, and the quality indicators published by the European Society of Breast Cancer Specialist, using claims data from the Programme de Médicalisation du Système d'Information (PMSI) and pathology reports. For each indicator, we calculated the number (%) of patients for whom all necessary data sources were available, and the number (%) of patients for whom all elementary variables were available in the sources, and for whom the related HQSI was computable. To extract useful data from the free text reports, we developed and validated dedicated rule-based algorithms, whose performance metrics were assessed with recall, precision, and f1-score. RESULTS: Out of 5785 female patients diagnosed with a breast cancer (60.9 years, IQR [50.0-71.9]), 5,147 (89.0%) had procedures related to breast cancer recorded in the PMSI, and 3732 (72.5%) had at least one surgery. Out of the 34 key indicators, 9 could be calculated with the PMSI alone, and 6 others became so using the data from pathology reports. Ten elementary variables were needed to calculate the 6 indicators combining the PMSI and pathology reports. The necessary sources were available for 58.8% to 94.6% of patients, depending on the indicators. The extraction algorithms developed had an average accuracy of 76.5% (min-max [32.7%-93.3%]), an average precision of 77.7% [10.0%-97.4%] and an average sensitivity of 71.6% [2.8% to 100.0%]. Once these algorithms applied, the variables needed to calculate the indicators were extracted for 2% to 88% of patients, depending on the indicators. DISCUSSION: The availability of medical reports in the electronic health records, of the elementary variables within the reports, and the performance of the extraction algorithms limit the population for which the indicators can be calculated. CONCLUSIONS: The automated calculation of quality indicators from electronic health records is a prospect that comes up against many practical obstacles.

Effect of Electronic Health Record Prompts on the Frequency of Blood Lead Tests.

INTRODUCTION/OBJECTIVES: Elevated blood lead levels can cause impaired cognition and behavioral problems in children. Screening is important for identifying children with elevated blood lead levels, but many children who qualify for screening do not get tested. We aimed to see if the addition of prompts in the electronic health record (EHR) would lead to differences in blood lead tests ordered for children with government insurance. METHODS: In May 2018, a prompt was added to our institutional EHR that reminded primary care practitioners to recommend lead testing for patients with government insurance. For this retrospective observational pre-post comparative study, we reviewed the rate of blood lead test orders and completed collection before and after the prompt was introduced. RESULTS: The number of blood lead tests ordered did not increase after prompts were introduced in the EHR; rather, the lead screening rates at 12-month well-child visits decreased from 63.6% to 53.8% (P = .008). The 24-month visit data did not change significantly for the number of lead tests ordered before and after the prompt was introduced in the EHR. The number of lead tests completed showed a significant decrease after the prompt was introduced for the 12-month visit (P < .001) but no significant change for the 24-month visit (P = .70). CONCLUSIONS: This study showed that the addition of prompts in the EHR was not associated with an increase in the number of blood lead level tests ordered. Further research is needed to determine factors that could affect lead screening rates.

Challenge of Detecting Personal Deviations and Trends in Sensor Based Activity Data.

INTRODUCTION: Physical activity and health are closely linked. Therefore, monitoring movement behavior is of great interest e.g., to monitor a patient's physical state. Nowadays it is easy to record movement with a smartphone. The aim of this work was to develop a concept to detect trends based on personalized movement behavior recorded with a smartphone. METHODS: A first prototype with a control chart was designed. Since this approach did not prove suitable for analyzing activity data for trends in practice, a second prototype was subsequently developed with a statistical trend test (Mann-Kendall test (MK test)). It was extended by the Yue-Wang correction approach to be able to deal effectively with serial correlation. Furthermore, the traditional trend modeling using Theil-Sen slope was extended by three additional models to be able to represent non-linear trend shapes. RESULTS: Movement behavior can be highly variable, which leads to wide control limits when using control charts. As the lower control limit was always in the negative range the use of a control chart was impossible for this use case. The evaluation results of the second prototype confirm the choice of a non-parametric test, as well as the decision for the Yue-Wang correction factor. Furthermore, it could be determined that the MK test is robust against outliers. The number of detected trends increases with increasing significance level. The MK test is also suitable for detecting step-like trends. CONCLUSION: Live trend detection is not straightforward with the MK test but can be simulated by overlapping time periods. In the future, trend modeling should be extended even further, as it plays a major role in the concept. The sensitivity of the test can be increased by means of various parameters.

Assessing the efficient use of the lightwave health information management system for health service delivery in Ghana.

BACKGROUND: In achieving the WHO's Universal Health Coverage and the Global Developmental Agenda: Sustainable Development Goal 3 and 9, the Ministry of Health launched a nationwide deployment of the lightwave health information management system (LHIMS) in the Central Region to facilitate health service delivery. This paper assessed the efficient use of the LHIMS among health professionals in the Central Region. METHODS: A non-interventional descriptive cross-sectional study design was employed for this research. The study used stratified and simple random sampling for selecting 1126 study respondents from 10 health facilities that use the LHIMS. The respondents included prescribers, nurses, midwives and auxiliary staff. Descriptive statistics (weighted mean) was computed to determine the average weighted score for all the indicators under efficiency. Also, bivariate (χ2) and multivariate (ordinal logistic regression) analyses were conducted to test the study's hypotheses. RESULTS: Findings revealed that the LHIMS enhanced efficient health service delivery. From the bivariate analysis, external factors; sex, educational qualification, work experience, profession type and computer literacy were associated with the efficient use of the LHIMS. However, training offered prior to the use of the LHIMS, and the duration of training had no association. At the multivariate level, only work experience and computer literacy significantly influenced the efficient use of the LHIMS. CONCLUSION: The implementation of LHIMS has the potential to significantly improve health service delivery. General computing skills should be offered to system users by the Ministry of Health to improve literacy in the use of computers. Active participation in the use of LHIMS by all relevant healthcare professionals should be encouraged.

Organizational and health professional readiness for the implementation of electronic medical record system: an implication for the current EMR implementation in northwest Ethiopia.

OBJECTIVE: The WHO developed a manual outlining the preliminary organizational and health professionals' readiness to implement electronic medical records (EMR). On the other hand, the readiness assessment in Ethiopia only includes the evaluation of health professionals, leaving out organisational readiness components. As a result, this research aimed to determine health professionals' and organizational readiness to implement EMR at a specialized teaching hospital. METHODS: An institutional-based cross-sectional study design was conducted among 423 health professionals and 54 managers. Self-administered and pretested questionnaires were used to collect data. Binary logistic regression analysis was used to identify factors associated with health professionals' readiness for EMR implementation. An OR with a 95% CI and p<0.05 was used to determine the strength of the association and the statistical significance, respectively. RESULTS: In this study, 53.7% management capacity, 33.3% finance and budget capacity, 42.6% operational capacity, 37.0% technology capability and 53.7% organisational alignment among the five dimensions evaluated to assess an organisation's readiness to implement an EMR system. Of 411 health professionals in this study, 173 (42.1%) with (95 CI 37.3% to 46.8%) were ready to implement an EMR system at the hospital. Sex (AOR 2.69, 95% CI 1.73 to 4.18), basic computer training (AOR 1.59, 95% CI 1.02 to 2.46), knowledge of EMR (AOR 1.88, 95% CI 1.19 to 2.97) and attitudes towards EMR (AOR 1.65, 95% CI 1.05 to 2.59) were significantly associated with health professionals' readiness towards EMR system implementation. CONCLUSIONS: Findings showed that most dimensions of organizational readiness for EMR implementation were below 50%. This study also revealed a lower level of EMR implementation readiness among health professionals compared with previous research studies' results. To improve organisational readiness to implement an electronic medical record system, a focus on management capability, financial and budget capability, operational capability, technical capability and organisational alignment was crucial. Likewise, having basic computer training, giving special attention to female health professionals and improving health professionals' knowledge of and attitudes towards EMR could help improve the readiness level of health professionals for implementing an EMR system.

Breast cancer screening with digital breast tomosynthesis: comparison of different reading strategies implementing artificial intelligence.

OBJECTIVES: Digital breast tomosynthesis (DBT) can detect more cancers than the current standard breast screening method, digital mammography (DM); however, it can substantially increase the reading workload and thus hinder implementation in screening. Artificial intelligence (AI) might be a solution. The aim of this study was to retrospectively test different ways of using AI in a screening workflow. METHODS: An AI system was used to analyse 14,772 double-read single-view DBT examinations from a screening trial with paired DM double reading. Three scenarios were studied: if AI can identify normal cases that can be excluded from human reading; if AI can replace the second reader; if AI can replace both readers. The number of detected cancers and false positives was compared with DM or DBT double reading. RESULTS: By excluding normal cases and only reading 50.5% (7460/14,772) of all examinations, 95% (121/127) of the DBT double reading detected cancers could be detected. Compared to DM screening, 27% (26/95) more cancers could be detected (p < 0.001) while keeping recall rates at the same level. With AI replacing the second reader, 95% (120/127) of the DBT double reading detected cancers could be detected-26% (25/95) more than DM screening (p < 0.001)-while increasing recall rates by 53%. AI alone with DBT has a sensitivity similar to DM double reading (p = 0.689). CONCLUSION: AI can open up possibilities for implementing DBT screening and detecting more cancers with the total reading workload unchanged. Considering the potential legal and psychological implications, replacing the second reader with AI would probably be most the feasible approach. KEY POINTS: • Breast cancer screening with digital breast tomosynthesis and artificial intelligence can detect more cancers than mammography screening without increasing screen-reading workload. • Artificial intelligence can either exclude low-risk cases from double reading or replace the second reader. • Retrospective study based on paired mammography and digital breast tomosynthesis screening data.

Understanding barriers of receiving short message service appointment reminders across African regions: a systematic review.

OBJECTIVE: Patients frequently miss their medical appointments. Therefore, short message service (SMS) has been used as a strategy for medical and healthcare service appointment reminders. This systematic review aimed to identify barriers to SMS appointment reminders across African regions. METHODS: PubMed, Google Scholar, Semantic Scholar and Web of Science were used for searching, and hand searching was done. Original studies written in English, conducted in Africa, and published since 1 December 2018, were included. The standard quality assessment checklist was used for the quality appraisal of the included studies. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses flowchart diagram was used for study selection and screening, and any disagreements were resolved via discussions. RESULTS: A total of 955 articles were searched, 521 studies were removed due to duplication and 105 studies were assessed for eligibility. Consequently, nine studies met the inclusion criteria. Five out of nine included studies were done by randomised control trials. The barriers that hampered patients, mothers and other parental figures of children when they were notified via SMS of medical and health services were identified. Among the 11 identified barriers, illiteracy, issues of confidentiality, familiarised text messages, inadequate information communication technology infrastructure, being a rural resident and loss of mobile phones occurred in at least two studies. CONCLUSIONS: SMS is an effective and widely accepted appointment reminder tool. However, it is hampered by numerous barriers. Hence, we gathered summarised information about users' barriers to SMS-based appointment reminders. Therefore, stakeholders should address existing identified barriers for better Mhealth interventions. PROSPERO REGISTRATION NUMBER: CRD42022296559.

Parámetros para la evaluación visual e instrumental del color dental en estudios in-vitro. Revisión de la literatura / Parameters for visual and instrumental assessment of tooth colour in in-vitro studies. Literature review

Objetivo: establecer los parámetros para la evaluación visual e instrumental del color dental en estudios in-vitro a partir de la literatura científca publicada entre 2015 y 2021. Métodos: se realizó la búsqueda en las bases de datos: PubMed, Web of Science, Science Direct, Scopus, Scielo y Lilacs; también en el motor de búsqueda Google Académico y las bibliotecas de las editoriales Wiley y Springer. Las palabras clave utilizadas fueron tooth, color, in-vitro, color perception, shade matching, thresholds, appearance, surrounding, "CIELAB" y "CIEDE2000". Teniendo en cuenta los criterios de elegibilidad, se seleccionaron los estudios de acuerdo al título, resumen y texto completo. Resultados: la búsqueda arrojó un total de 37 publicaciones que se agruparon en tres tópicos: 1. toma de color visual: condiciones ambientales, observadores y nivelación; 2. toma de color instrumental: instrumentos; y 3. procesamiento de datos: cálculo de la diferencia de color y umbrales de perceptibilidad (PT) y aceptabilidad (AT). Conclusiones: los aspectos más importantes en la evaluación visual son la iluminación, el ambiente para registro (sitio, entorno y fondo alrededor de la muestra), las condiciones geométricas de visualización, los observadores y el uso de guías. En la evaluación instrumental es relevante elegir el aparato apropiado de acuerdo con su precisión y reproducibilidad, como los espectroradiómetros y los espectrofotómetros de uso clínico. Se presenta el procesamiento de datos para establecer las variaciones de cada coordenada, las diferencias de color (ΔE): CIELAB y CIEDE2000, los umbrales y los lineamientos.

Objective: To establish the parameters for the visual and instrumental evaluation of tooth color in in-vitro studies based on the scientifc literature published between 2015 and 2021. Methods: The search was carried out in the databases of PubMed, Web of Science, Science Direct, Scopus, Scielo, Lilacs; search engine Google Scholar and publishers' library of Wiley and Scielo, using the keywords "tooth", "color", "in vitro", "color perception", "shade matching", "thresholds", "appearance", "surrounding", "CIELAB", and "CIEDE2000". The literature was selected according to the title, abstract and full text taking into the eligibility criteria. Results: It yielded a total of 37 publications, which were grouped into three topics: 1. visual color acquisition: environmental conditions for color acquisition, observers and levelling. 2. instrumental color sampling: instruments. 3. Data processing: Calculation of color diference and perception thresholds (PT) and acceptability thresholds (AT). Conclusions: The most important aspects in the visual assessment are lighting, the environment for color registration (site, environment and background around the sample), the geometric conditions of visualization, the observers and the use of guides. Regarding the instrumental assessment of color, the appropriate devices must be chosen according to its precision and reproducibility, being the spectrophotometers and spectroradiometers the most precise ones. It is presented how the data processing is carried out to establish the variations of each coordinate, the color diferences (ΔE): CIELAB and CIEDE2000, thresholds and guidelines.

[Automation and application of robotics in the pathology laboratory]. / Automatisierung und Einsatz von Robotern im Pathologielabor.

Over the last 20 years, numerous technical innovations have been introduced to the histopathology laboratory, providing tools for improved standardization and occupational safety. Digital tracking serves as a backbone accompanying the workflow from labeling cassettes and slides to the final steps of preparation of whole slide images and archiving blocks and sections. Multifunctional devices eliminated time consuming manual work prone to mistakes and loss of materials. At present, collaborative robots take over manual work that was considered to be exclusive to humans. The advent of these new technologies is expected to ameliorate the increasing staffing shortage in the laboratory and on the side of histopathologists as well.

Development of a customised programme to standardise comorbidity diagnosis codes in a large-scale database.

OBJECTIVES: The transition from ICD-9 to ICD-10 coding creates a data standardisation challenge for large-scale longitudinal research. We sought to develop a programme that automated this standardisation process. METHODS: A programme was developed to standardise ICD-9 and ICD-10 terminology into one system. Code was improved to reduce runtime, and two iterations were tested on a joint ICD-9/ICD-10 database of 15.8 million patients. RESULTS: Both programmes successfully standardised diagnostic terminology in the database. While the original programme updated 100 000 cells in 12.5 hours, the improved programme translated 3.1 million cells in 38 min. DISCUSSION: While both programmes successfully translated ICD-related data into a standardised format, the original programme suffered from excessive runtimes. Code improvement with hash tables and parallelisation exponentially reduced these runtimes. CONCLUSION: Databases with ICD-9 and ICD-10 codes require terminology standardisation for analysis. By sharing our programme's implementation, we hope to assist other researchers in standardising their own databases.

[Digital health applications and data protection]. / Digitale Gesundheitsanwendungen und Datenschutz.

Digital health applications represent a new form of care. The basis for the approval of digital health applications is the Digital Healthcare Act. In order to be included in the directory, the digital health applications must undergo an extensive evaluation process by the Federal Institute for Drugs and Medical Devices. The focus is on proving added value for care, but also on the technical aspects. This strictly differentiates the digital health applications from the health apps. Cutting-edge apps enable a simple output of collected data to make doctor-patient interactions efficient. Appropriate remuneration and education could increase the acceptance by the medical profession and thus accelerate implementation; however, such instruments and incentives are not currently provided for in the system.

[EASYSCAN: French pilot study for securing drug administration by barcode reading]. / EASYSCAN : étude pilote française pour la sécurisation de l'administration des médicaments par lecture code-barres.

OBJECTIVES: Medication errors are common at the time of administration. To prevent them, technologies allowing consistency check by bar code technology at bedside have been developed. Our study focuses on the evaluation of a BarCode Medication Administration (BCMA) called EASYSCAN with Electronic Medication Administration Record (e-MAR) to verify both patient's identity and medication to be administrated. METHODS: A prospective observational study was conducted during seven weeks in a French medicine ward. The performance of the system was evaluated by the success rate of BCMA and by the average time for administration with and without EASYSCAN. A satisfaction questionnaire about BCMA was proposed to nurses. RESULTS: We observed 182 administrations including 87 (48%) with EASYSCAN. The verification of the patient's identity was successful in 77% of administrations and 65% of the drugs were scanned successfully. The main causes of check failures were the lack of datamatrix on the drug (81%), error messages (14%) and the lack of system functionality (5%). The average time for administration per patient was significantly increased: 4.68min/patient with versus 2.87min/patient without EASYSCAN. CONCLUSIONS: The study shows the EASYSCAN's performance in its first version. Material and software evolutions and an increase of nurses'pratices will be necessary to continue the experimentation of this system still unpublished in France.

Making Clinical Data Acquisition Standards Harmonization (CDASH) Electronic Case Report Forms Available on the REDCap Shared Data Instrument Library.

Introduction: A guiding principle behind the development and deployment of the REDCap data management platform has always included attention to workflow design that allows easy implementation of best practices for clinical and translational researchers. CDISC standards such as CDASH have helped the clinical research community improve the efficiency, actionability, and quality of their clinical trials data, but have had limited uptake among the academic institutions. Objective: To create a scalable methodology to convert CDISC CDASH eCRF instrument metadata into REDCap data dictionaries for the purpose of simplifying adoption and use of CDASH instruments by research teams across the REDCap Consortium. Implementation: We have used our replicable methods to translate metadata from 34 CDASH Foundational eCRFs and 20 CDASH Crohn's Disease eCRFs into REDCap eCRF metadata and have made these instruments available in the REDCap Shared Data Instrument Library for widespread sharing and uptake across the REDCap Consortium. Users can import the standardized eCRFs directly into their REDCap projects for immediate use in clinical trial data collection. Conclusion: Disseminating CDISC standards through the REDCap community will increase the accessibility of these standards for academic medical centers. Having academic clinical researchers using CDISC standards may lead to more research datasets that interoperate with pharmaceutical sponsored trials, and more discoveries from secondary use of clinical research data.

Potencialidades e limites do processamento de dados em pesquisas sobre a produção científica / Potential and limits of data processing in research on scientific production / Potencial y límites del procesamiento de datos en investigaciones sobre la producción científica

Este trabalho versa sobre potencialidades e limites relacionados à utilização de processamentos de dados para auxiliar na produção e sistematização de conhecimento científico. Objetiva, através de um exercício experimental envolvendo a utilização de algoritmo, discutir a viabilidade do uso de técnicas de coleta automatizada para levantamento e produção de dados utilizáveis no âmbito das pesquisas científicas. Como demonstração, busca reproduzir de maneira automatizada processos relacionados à coleta de dados de pesquisa anteriormente publicada neste periódico, descrevendo metodologicamente como foram organizados e desenvolvidos a extração e o tratamento desses dados. Como resultado, constata que o processamento automatizado pode ser uma alternativa produtiva e eficiente para auxiliar nas sistematizações e análises sobre o acumulado crescente de publicações no campo científico, podendo abrir novos caminhos metodológicos de pesquisa na Educação Física, especialmente considerando o volume de dados passível de coleta e análise em redes sociais, fóruns e outras plataformas na web. (AU)

This paper deals with potentials and limits related to the use of data processing to assist in the production and systematization of scientific knowledge. It aims, through an experimental exercise involving the use of an algorithm, to discuss the feasibility of using automated collection techniques for surveying and producing data that can be used in scientific research. As a demonstration, it seeks to automatically reproduce processes related to the collection of research data previously published in this journal, describing methodologically how the extraction and treatment of these data was organized and developed. As a result, it finds that automated processing can be a productive and efficient alternative to assist in the systematization and analysis of the growing accumulation of publications in the scientific field, which may open new methodological paths for research in Physical Education, especially considering the volume of data subject to collection and analysis on social networks, forums and other web platforms. (AU)

Este trabajo aborda las potencialidades y los límites relacionados con el uso del procesamiento de datos para ayudar en la producción y sistematización del conocimiento científico. Su objetivo, a través de un ejercicio experimental que implica el uso de un algoritmo, es discutir la viabilidad del uso de técnicas de recolección automatizada para la obtención y producción de datos que se puedan utilizar en el ámbito de las investigaciones científicas. A modo de demostración, se busca reproducir de manera automatizada procesos relacionados con la recolección de datos de una investigación previamente publicada en esta revista, describiendo metodológicamente cómo se organizó y desarrolló la extracción y el tratamiento de esos datos. Como resultado, se constata que el procesamiento automatizado puede ser una alternativa productiva y eficiente para ayudar en la sistematización y análisis de la creciente acumulación de publicaciones en el campo científico, lo que puede abrir nuevos caminos metodológicos para la investigación en Educación Física, especialmente considerando el volumen de datos que se pueden recolectar y analizar en redes sociales, foros y otras plataformas web. (AU)

Diagnosing post-traumatic stress disorder using electronic medical record data.

This study proposes a predictive model that uses structured data and unstructured narrative notes from Electronic Medical Records to accurately identify patients diagnosed with Post-Traumatic Stress Disorder (PTSD). We utilize data from primary care clinicians participating in the Manitoba Primary Care Research Network (MaPCReN) representing 154,118 patients. A reference sample of 195 patients that had their PTSD diagnosis confirmed using a manual chart review of structured data and narrative notes, and PTSD negative patients is used as the gold standard data for model training, validation and testing. We assess structured and unstructured data from eight tables in the MaPCReN namely, patient demographics, disease case, examinations, medication, billing records, health condition, risk factors, and encounter notes. Feature engineering is applied to convert data into proper representation for predictive modeling. We explore serial and parallel mixed data models that are trained on both structured and unstructured data to identify PTSD. Model performances were calculated based on a highly skewed hold-out test dataset. The serial model that uses both structured and text data as input, yielded the highest values in sensitivity (0.77), F-measure (0.76), and AUC (0.88) and the parallel model that uses both structured and text data as the input obtained the highest positive predicted value (PPV) (0.75). Diseases such as PTSD are difficult to diagnose. Information recorded in the chart note over multiple visits of the patients with the primary care physicians has higher predictive power than structured data and combining these two data types can increase the predictive capabilities of machine learning models in diagnosing PTSD. While the deep-learning model outperformed the traditional ensemble model in processing text data, the ensemble classifier obtained better results in ingesting a combination of features obtained from both data types in the serial mixed model. The study demonstrated that unstructured encounter notes enhance a model's ability to identify patients diagnosed with PTSD. These findings can enhance quality improvement, research, and disease surveillance related to PTSD in primary care populations.

Subcategorizing EHR diagnosis codes to improve clinical application of machine learning models.

BACKGROUND: Electronic health record (EHR) data is commonly used for secondary purposes such as research and clinical decision support. However, reuse of EHR data presents several challenges including but not limited to identifying all diagnoses associated with a patient's clinical encounter. The purpose of this study was to assess the feasibility of developing a schema to identify and subclassify all structured diagnosis codes for a patient encounter. METHODS: To develop a subclassification schema we used EHR data from an interhospital transport data repository that contained complete hospital encounter level data. Eight discrete data sources containing structured diagnosis codes were identified. Diagnosis codes were normalized using the Unified Medical Language System and additional EHR data were combined with standardized terminologies to create and validate the subcategories. We then employed random forest to assess the usefulness of the new subcategorized diagnoses to predict post-interhospital transfer mortality by building 2 models, one using standard diagnosis codes, and one using the new subcategorized diagnosis codes. RESULTS: Six subcategories of diagnoses were identified and validated. The subcategories included: primary or admitting diagnoses (10%), past medical, surgical or social history (9%), problem list (20%), comorbidity (24%), discharge diagnoses (6%), and unmapped diagnoses (31%). The subcategorized model outperformed the standard model, achieving a training AUROC of 0.97 versus 0.95 and testing model AUROC of 0.81 versus 0.46. DISCUSSION: Our work demonstrates that merging structured diagnosis codes with additional EHR data and secondary data sources provides additional information to understand the role of diagnosis throughout a clinical encounter and improves predictive model performance. Further work is necessary to assess if subcategorizing produces benefits in interpreting the results of prognostic models and/or operationalizing the results in clinical decision support applications.

First Steps to Evaluate an NLP Tool's Medication Extraction Accuracy from Discharge Letters.

INTRODUCTION: The aim of this study is to evaluate the use of a natural language processing (NLP) software to extract medication statements from unstructured medical discharge letters. METHODS: Ten randomly selected discharge letters were extracted from the data warehouse of the University Hospital Erlangen (UHE) and manually annotated to create a gold standard. The AHD NLP tool, provided by MIRACUM's industry partner was used to annotate these discharge letters. Annotations by the NLP tool where then compared to the gold standard on two levels: phrase precision (whether or not the whole medication statement has been identified correctly) and token precision (whether or not the medication name has been identified correctly within correctly discovered medication phrases). RESULTS: The NLP tool detected medication related phrases with an overall F-measure of 0.852. The medication name has been identified correctly with an overall F-measure of 0.936. DISCUSSION: This proof-of-concept study is a first step towards an automated scalable evaluation system for MIRACUM's industry partner's NLP tool by using a gold standard. Medication phrases and names have been correctly identified in most cases by the NLP system. Future effort needs to be put into extending and validating the gold standard.