Browsing by Author "Capurro, Daniel"

Now showing 1 - 20 of 26
  • Thumbnail Image
    Item
    A conjoint analysis framework for evaluating user preferences in machine translation
    (2014) Kirchhoff, K.; Capurro, Daniel; Turner, A. M.
    Despite much research on machine translation (MT) evaluation, there is surprisingly little work that directly measures users' intuitive or emotional preferences regarding different types of MT errors. However, the elicitation and modeling of user preferences is an important prerequisite for research on user adaptation and customization of MT engines. In this paper we explore the use of conjoint analysis as a formal quantitative framework to assess users' relative preferences for different types of translation errors. We apply our approach to the analysis of MT output from translating public health documents from English into Spanish. Our results indicate that word order errors are clearly the most dispreferred error type, followed by word sense, morphological, and function word errors. The conjoint analysis-based model is able to predict user preferences more accurately than a baseline model that chooses the translation with the fewest errors overall. Additionally we analyze the effect of using a crowd-sourced respondent population versus a sample of domain experts and observe that main preference effects are remarkably stable across the two samples
  • Thumbnail Image
    Item
    A Survey on Deep Learning and Explainability for Automatic Report Generation from Medical Images
    (2022) Messina, Pablo; Pino, Pablo; Parra Santander, Denis; Soto Arriaza, Álvaro Marcelo; Besa, Cecilia; Uribe Arancibia, Sergio A.; Andía Kohnenkampf, Marcelo Edgardo; Tejos Núñez, Cristián Andrés; Prieto Vásquez, Claudia; Capurro, Daniel
  • Thumbnail Image
    Item
    Analysis of Emergency Room Episodes Duration Through Process Mining
    (2019) Rojas, Eric; Cifuentes Soto, Andrés Alonso; Burattin, A.; Muñoz Gama, Jorge; Sepúlveda, Marcos; Capurro, Daniel
    This study presents the proposal of a performance analysis method for ER Processes through Process Mining. This method helps to determine which activities, sub-processes, interactions and characteristics of episodes explain why the process has long episode duration, besides providing decision makers with additional information that will help to decrease waiting times, reduce patient congestion and increment quality of provided care. By applying the exposed method to a case study, it was discovered that when a loop is formed between the Examination and Treatment sub-processes, the episode duration lengthens. Moreover, the relationship between case severity and the number of repetitions of the Examination-Treatment loop was also studied. As the case severity increases, the number of repetitions increases as well.
  • Thumbnail Image
    Item
    Analyzing medical emergency processes with process mining: the stroke case
    (2019) Fernandez-Llatas, Carlos; Ibanez-Sanchez, Gema; Celda, Angeles; Mandingorra, Jesus; Aparici-Tortajada, Lucia; Martinez-Millana, Antonio; Munoz-Gama, Jorge; Sepúlveda, Marcos; Rojas, Eric; Gálvez, Víctor; Capurro, Daniel; Traver, Vicente
    Medical emergencies are one of the most critical processes that occurs in a hospital. The creation of adequate and timely triage protocols, can make the difference between the life and death of the patient. One of the most critical emergency care protocols is the stroke case. This disease demands an accurate and quick diagnosis for ensuring an immediate treatment in order to limit or even, avoid, the undesired cognitive decline. The aim of this paper is perform an analysis of how Process Mining techniques can support health professionals in the interactive analysis of emergency processes considering critical timing of Stroke, using a Question Driven methodology. To demonstrate the possibilities of Process Mining in the characterization of the emergency process, we have used a real log with 9046 emergency episodes from 2145 stroke patients that occurred from January of 2010 to June of 2017. Our results demonstrate how Process Mining technology can highlight the differences of the stroke patient flow in emergency, supporting professionals in the better understanding and improvement of quality of care.
  • Thumbnail Image
    Item
    Association between demographic, clinical characteristics and severe complications by SARS-CoV-2 infection in a community-based healthcare network in Chile
    (2024) Leniz Martelli, Javiera; Hernández Jaña, Sam Steven; Soto Durán, Mauricio Javier; Soto, Mauricio; Arenas, Eduardo; Margozzini Maira, Paula Andrea; Suárez, Francisco; Capurro, Daniel; Rojas Villar, María Paulina; Bambs Sandoval, Claudia Elena
    Background Most of the evidence on risk factors for COVID-19 complications comes from North America or Europe with very little research from Latin-America. We aimed to evaluate the association between sociodemographic, clinical factors and the risk of COVID-19 complications among adults in Chile, the fifth Latin-American country with more COVID-19 reported cases since de beginning of the Pandemic. Methods A retrospective population-based cohort study using data from electronic health records from a large Primary Care Network, linked to national hospital, immunization, Covid-19 PCR surveillance, mortality and birth records. We included people 18+ years old enrolled in the Primary Care Network between 1st January 2020 and 31st December 2021. Using Multivariate Cox proportional hazard models, we evaluate the association between sociodemographic, clinical characteristics with three COVID-19 complications: (1) a hospital admission, (2) an ICU admission, and (3) death due to a COVID-19 infection that occurred between the 1st January 2020 and the 31st December 2021. Results 44,674 people were included. The mean age was 44.30 (sd 17.31), 55.6% were female, 15.9% had a type of healthcare insurance for people from the lowest category of income, 11.6% and 9.4% had a record of hypertension or diabetes mellitus diagnosis. Among the 44,674 people, 455 (1.02%) had a hospital admission due to a COVID-19 infection and 216(0.48%) of them also had an ICU admission. Among the 44,674 people,148(0.33%) died due to COVID-19 infection. Older age and male sex were consistently associated with a higher risk of the three COVID-19 complications. Hypertension and diabetes were associated with a higher risk of a hospital admission and death, but not with an ICU admissions due to COVID-19 infection. Having two or more COVID-19 vaccine doses compared with no doses was associated with a lower risk of any hospital admission (HR 0.81; 95% CI 0.77–0.84), an ICU admission (HR 0.60; 95% CI 0.57–0.63) and death (HR 0.50; 95% CI 0.46–0.54). Pregnant or puerperal women were more likely to be admitted to hospital (HR 2.89; 95% CI 1.41–5.89) or ICU (HR 3.04; 95% CI 1.01–9.14). Conclusions Sociodemographic and clinical factors associated with COVID-19 complications such as age, sex and pre-existing conditions were comparable to those reported in similar studies from higher-income countries, and can be used to predict severity in COVID-19 patients.
  • Thumbnail Image
    Item
    Availability of Structured and Unstructured Clinical Data for Comparative Effectiveness Research and Quality Improvement: A Multi-Site Assessment
    (2014) Capurro, Daniel; Yetisgen, Meliha; Van Eaton, Erik; Black, Robert; Tarczy Hornoch, Peter
  • Thumbnail Image
    Item
    Correlation between spontaneous preterm birth and mid-trimester maternal serum estriol
    (2014) Olsen, R. N.; Dunsmoor-Su; Capurro, Daniel; Mcmahon, K.; Gravett, M. G.
  • Thumbnail Image
    Item
    Development of mobile technologies for the prevention of cervical cancer in Santiago, Chile study protocol : a randomized controlled trial
    (2017) Martínez Gutiérrez, Javiera; Soto Subiabre, Mauricio Andrés; Capurro, Daniel; Puschel Illanes, Klaus; Momany, McKenzie C.; Ciampi, Francis; Thompson, Beti
    Abstract Background In Chile, more than 500 women die every year from cervical cancer, and a majority of Chilean women are not up-to-date with their Papanicolau (Pap) test. Mobile health has great potential in many health areas, particularly in health promotion and prevention. There are no randomized controlled trials in Latin America assessing its use in cervical cancer screening. The ‘Development of Mobile Technologies for the Prevention of Cervical Cancer in Santiago, Chile’ study aims to determine the efficacy of a text-message intervention on Pap test adherence among Chilean women in the metropolitan region of Santiago. Methods/design This study is a parallel randomized-controlled trial of 400 Chilean women aged 25–64 who are non-adherent with current recommendations for Pap test screening. Participants will be randomly assigned to (1) a control arm (usual care) or (2) an intervention arm, where text and voice messages containing information and encouragement to undergo screening will be sent to the women. The primary endpoint is completion of a Pap test within 6 months of baseline assessment, as determined by medical record review at community-based clinics. Medical record reviewers will be blinded to randomization arms. The secondary endpoint is an evaluation of the implementation and usability of the text message intervention as a strategy to improve screening adherence. Discussion This intervention using mobile technology intends to raise cervical cancer screening adherence and compliance among a Chilean population of low and middle-low socioeconomic status. If successful, this strategy may reduce the incidence of cervical cancer. Trial registration Clinicaltrials.gov NCT02376023 Registered 2/17/2015. First participant enrolled Feb 22nd 2016.Abstract Background In Chile, more than 500 women die every year from cervical cancer, and a majority of Chilean women are not up-to-date with their Papanicolau (Pap) test. Mobile health has great potential in many health areas, particularly in health promotion and prevention. There are no randomized controlled trials in Latin America assessing its use in cervical cancer screening. The ‘Development of Mobile Technologies for the Prevention of Cervical Cancer in Santiago, Chile’ study aims to determine the efficacy of a text-message intervention on Pap test adherence among Chilean women in the metropolitan region of Santiago. Methods/design This study is a parallel randomized-controlled trial of 400 Chilean women aged 25–64 who are non-adherent with current recommendations for Pap test screening. Participants will be randomly assigned to (1) a control arm (usual care) or (2) an intervention arm, where text and voice messages containing information and encouragement to undergo screening will be sent to the women. The primary endpoint is completion of a Pap test within 6 months of baseline assessment, as determined by medical record review at community-based clinics. Medical record reviewers will be blinded to randomization arms. The secondary endpoint is an evaluation of the implementation and usability of the text message intervention as a strategy to improve screening adherence. Discussion This intervention using mobile technology intends to raise cervical cancer screening adherence and compliance among a Chilean population of low and middle-low socioeconomic status. If successful, this strategy may reduce the incidence of cervical cancer. Trial registration Clinicaltrials.gov NCT02376023 Registered 2/17/2015. First participant enrolled Feb 22nd 2016.Abstract Background In Chile, more than 500 women die every year from cervical cancer, and a majority of Chilean women are not up-to-date with their Papanicolau (Pap) test. Mobile health has great potential in many health areas, particularly in health promotion and prevention. There are no randomized controlled trials in Latin America assessing its use in cervical cancer screening. The ‘Development of Mobile Technologies for the Prevention of Cervical Cancer in Santiago, Chile’ study aims to determine the efficacy of a text-message intervention on Pap test adherence among Chilean women in the metropolitan region of Santiago. Methods/design This study is a parallel randomized-controlled trial of 400 Chilean women aged 25–64 who are non-adherent with current recommendations for Pap test screening. Participants will be randomly assigned to (1) a control arm (usual care) or (2) an intervention arm, where text and voice messages containing information and encouragement to undergo screening will be sent to the women. The primary endpoint is completion of a Pap test within 6 months of baseline assessment, as determined by medical record review at community-based clinics. Medical record reviewers will be blinded to randomization arms. The secondary endpoint is an evaluation of the implementation and usability of the text message intervention as a strategy to improve screening adherence. Discussion This intervention using mobile technology intends to raise cervical cancer screening adherence and compliance among a Chilean population of low and middle-low socioeconomic status. If successful, this strategy may reduce the incidence of cervical cancer. Trial registration Clinicaltrials.gov NCT02376023 Registered 2/17/2015. First participant enrolled Feb 22nd 2016.
  • Thumbnail Image
    Item
    Discovering role interaction models in the Emergency Room using Process Mining
    (2018) Alvarez, Camilo; Rojas, Eric; Arias, Michael; Munoz-Gama, Jorge; Sepulveda, Marcos; Herskovic, Valeria; Capurro, Daniel
  • Thumbnail Image
    Item
    Epistemonikos: a comprehensive database of systematic reviews for health decision-making
    (2020) Rada G., Gabriel; Bravo Soto, Gonzalo A.; Bravo Jeria, Rocío; Capurro, Daniel; Morel Marambio, Macarena; Neumann Burotto, Gonzalo Ignacio; Ortiz Muñoz, Luis; Peña Durán, José Esteban; Rivera Mercado, Solange; Soto Subiabre, Mauricio Andrés
    Abstract Background Systematic reviews allow health decisions to be informed by the best available research evidence. However, their number is proliferating quickly, and many skills are required to identify all the relevant reviews for a specific question. Methods and findings We screen 10 bibliographic databases on a daily or weekly basis, to identify systematic reviews relevant for health decision-making. Using a machine-based approach developed for this project we select reviews, which are then validated by a network of more than 1000 collaborators. After screening over 1,400,000 records we have identified more than 300,000 systematic reviews, which are now stored in a single place and accessible through an easy-to-use search engine. This makes Epistemonikos the largest database of its kind. Conclusions Using a systematic approach, recruiting a broad network of collaborators and implementing automated methods, we developed a one-stop shop for systematic reviews relevant for health decision making.
  • Thumbnail Image
    Item
    Implementation science : Implications for intervention research in hospice and palliative care
    (2014) Demiris, George; Parker Oliver, Debra; Capurro, Daniel; Wittenberg-Lyles, Elaine
  • Thumbnail Image
    Item
    Importancia de la intención de tratar y el seguimiento en la validez interna de un estudio clínico randomizado
    (2004) Capurro, Daniel; Gabrielli, Luigi; Letelier Saavedra, Luz María
  • Thumbnail Image
    Item
    Informática biomédica
    (SOC MEDICA SANTIAGO, 2011) Capurro, Daniel; Soto, Mauricio; Vivent, Macarena; Lopetegui, Marcelo; Herskovic, Jorge R.
    Biomedical Informatics is a new discipline that arose from the need to incorporate information technologies to the generation, storage, distribution and analysis of information in the domain of biomedical sciences. This discipline comprises basic biomedical informatics, and public health informatics. The development of the discipline in Chile has been modest and most projects have originated from the interest of individual people or institutions, without a systematic and coordinated national development. Considering the unique features of health care system of our country, research in the area of biomedical informatics is becoming an imperative. (Rev Med Chile 2011; 139: 1611-1616).
  • Thumbnail Image
    Item
    Mapping the patient’s journey in healthcare through process mining
    (2020) Arias, M.; Rojas, Eric; Aguirre, S.; Cornejo, F.; Muñoz Gama, Jorge; Sepúlveda Fernández, Marcos Ernesto; Capurro, Daniel
  • Thumbnail Image
    Item
    Messages for your health : mobile use and cancer prevention for underserved Latinas in Santiago, Chile
    (2016) Martínez Gutiérrez, Javiera; Soto Subiabre, Mauricio Andrés; Momany, Mckenzie; Ciampi, Francis; Capurro, Daniel; Cea, Emilia; Puschel Illanes, Klaus
  • Thumbnail Image
    Item
    Multidisciplinary Collaboration in the Treatment of Patients With Type 2 Diabetes in Primary Care : Analysis Using Process Mining
    (2018) Conca, T.; Saint-Pierre Cortés, Cecilia; Herskovic, Valeria; Sepúlveda Fernández, Marcos Ernesto; Capurro, Daniel; Prieto Ugarte, Florencia; Fernandez Llatas, C.
  • Thumbnail Image
    Item
    PALIA-ER: Bringing question-driven process mining closer to the emergency room
    (2017) Rojas, Eric; Fernández-Llatas, Carlos; Traver, Vicente; Muñoz-Gama, Vicente; Sepúlveda, Marcos; Herskovic, Valeria; Capurro, Daniel
    This paper presents PALIA-ER, a web-based tool for question-driven process mining in Emergency Room. PALIA-ER uses Palia discovery algorithm and includes model simplification and filtering features specially domain-specific for ER. Most PALIA-ER functionalities can be easily applied to other interdisciplinary contexts such as other healthcare units, education, or logistics.
  • Thumbnail Image
    Item
  • Thumbnail Image
    Item
    Performance Analysis of Emergency Room Episodes Through Process Mining
    (2019) Rojas, E.; Cifuentes Soto, Andrés Alonso; Burattin, A.; Muñoz Gama, Jorge; Sepúlveda Fernández, Marcos Ernesto; Capurro, Daniel
  • Thumbnail Image
    Item
    Preferences of Underserved Chilean Women on a Mobile Technology Intervention for Cervical Cancer Screening : Qualitative Study
    (2018) Soto Subiabre, Mauricio Andrés; Martínez Gutiérrez, Javiera; Momany, McKenzie; Capurro, Daniel; Ciampi Spode, Francis; Cea, Emilia; Mergudich, Tania; Puschel Illanes, Klaus