Background Major incidents are characterized by a lack of resources compared to an overwhelming number of casualties, requiring a prioritization of medical treatment. Triage algorithms are an essential tool for prioritizing the urgency of treatment for patients, but the evidence to support one over another is very limited. We determined the influence of blood pressure limits on the diagnostic value of triage algorithms, considering if pulse should be palpated centrally or peripherally. Methods We used a database representing 500 consecutive HEMS patients. Each patient was allocated a triage category (T1/red, T2/yellow, T3/green) by a group of experienced doctors in disaster medicine, independent of any algorithm. mSTaRT, ASAV, Field Triage Score (FTS), Care Flight (CF), “Model Bavaria” and two Norwegian algorithms (Nor and TAS), all containing the question “Pulse palpable?”, were translated into Excel commands, calculating the triage category for each patient automatically. We used 5 blood pressure limits ranging from 130 to 60 mmHg to determine palpable pulse. The resulting triage categories were analyzed with respect to sensitivity, specificity and Youden Index (J) separately for trauma and non-trauma patients, and for all patients combined. Results For the entire population of patients within all triage algorithms the Youden Index (J) was highest for T1 (J between 0,14 and 0,62). Combining trauma and non-trauma patients, the highest J was obtained by ASAV (J = 0,62 at 60 mmHg). ASAV scored the highest within trauma patients (J = 0,87 at 60 mmHg), whereas Model Bavaria (J = 0,54 at 80 mmHg) reached highest amongst non-trauma patients. FTS performed worst for all patients (J = 0,14 at 60 mmHg), showing a lower score for trauma patients (J = 0,0 at 60 mmHg). Change of blood pressure limits resulted in different diagnostic values of all algorithms. Discussion We demonstrate that differing blood pressure limits have a remarkable impact on diagnostic values of triage algorithms. Further research is needed to determine the lowest blood pressure value that is possible to palpate at a peripheral artery compared to a central artery. Conclusion As a consequence, it might be important in which location pulses are palpated according to the algorithm at hand during triage of patients.

No description available.

Background: Historically, patients with cancer have been perceived as poor candidates for ICU admission. General ICU admission criteria lists cancer patients as low priority in ICU admission depriving them of the care they rightfully deserve. The purpose of this literary synthesis was to examine ICU admission criteria, risk factors, and outcomes of ICU admission in relation to hematological and solid tumor cancers and discuss ways that practitioners and nurses can educate patients with cancer and their families on appropriateness of ICU care. Methods: A total of 768 articles were found in a literature search including all literature from 2005 to 2016 from all countries using the databases CINAHL Plus, MEDLINE, PsycINFO, and Academic Search Premier. These were further narrowed down based on relevancy by topic or reading abstracts. A total of 13 articles utilizing the inclusion and exclusion criteria of the literature search were included in the final literature synthesis. Results: In addition to general ICU admission criteria several other criteria and scores can be helpful in admitting patients with cancer to the ICU including cancer specific criteria, mortality predictor tools, performance status, and ICU trials. Mortality predictors, in combination with other patient characteristics, demonstrated effectiveness to predict outcomes in patients with cancer. Survival rates in hematological and solid tumor cancers have improved from the past, and lower prognostic scores can predict who will have better outcomes. Conclusion: Cancer specific criteria, mortality predictor tools, performance status, and ICU trials in addition to general ICU criteria should be used for admission of cancer patients into ICU. Practitioners and nurses should become familiar with the newest outcomes in patients with cancer to make collaborative informed decisions about ICU admission.

<p>Тешка траума се може дефинисати на неколико различитих начина. Најчешће коришћена дефиниција укључује коришћење Injury Severity Score ( ISS скор). Ако је вредност ISS скор &gt;15, ради се о тешкој трауми. Траума је временски осетљиво стање, због тога је за збрињавање тешко повређених пацијената неопходна добра сарадња различитих нивоа здравствене заштите и здравствених стручњака различитих специјалности. У претходних неколико деценија, због сложенијег процеса лечења и великих трошкова, дошло је до потребе за што објективнијом проценом стања повређеног и исхода лечења. Циљ: Упоредити сензитивност и специфичност T-RTS скорa (Triage Revised Trauma Score), CRAMS скалe (Circulation, Respiration, Abdomen, Motor and Speech), МGAP скора (Mechanism, Glasgow coma scale, Age, and arterial Pressure) и GAP скора (Glasgow coma scale, Age, and arterial Pressure) примењених на прехоспиталном нивоу, проценити могућности прехоспитално примењених RTS, CRAMS, МGAP и GAP скорова у предикцији исхода лечења повређеног пацијента и утврдити значај појединачних фактора, одређиваних на прехоспиталном нивоу током иницијалног прегледa повређеног, за процену тежине повреде и предикцију исхода лечења повређеног. Истраживање је проспективног, опсервационог карактера. У истраживање су укључени пацијенти старији од 18 година, које су лекари Заводa за хитну медицинску помоћ Нови Сад (ЗЗХМП НС) и Службe хитне медицинске помоћи Дома здравља Бечеј (СХМП ДЗ Бечеј) збрињавали на терену након трауме, а потом их транспортовали у Ургентни центар Клиничког центра Војводине (УЦ КЦВ). На основу вредности ISS скора пацијенти су сврстани у једну од две групе: група А- пацијенти код којих је ISS скор након завршене дијагностике изнад 15- тешка траума и група Б -пацијенти код којих је након завршене дијагностике ISS скор &le;15- лака траума. У групи А је било 50, а у групи Б 257 пацијената. За граничне вредности скорова које означавају да траума није лака, највећу сензитивност у оцени тежине трауме је имао GAP скор 98,8%, а највећу специфичност MGAP скор 62%. У предвиђању исхода лечења, највећу сензитивност је имао RTS скор за предикцију 95,2%, а специфичност GAP скор и CRAMS скала 87,5%. MGAP скор, а пошто је у снажној корелацији са њим и GAP скор, мерени прехоспитално, су се издвојили као независни предиктор у оцени тежине трауме и предвиђању исхода лечења повређеног. Т- RTS скор и CRAMS скала су се издвојили као појединачни предиктори у оцени тежине трауме, али не и као независни предиктори. RTS скор за предикцију нема статистичку значајност у предвиђању исхода лечења повређеног, за разлику од CRAMS скале која има, али се није издвојила као независни предиктор исхода лечења. Осим наведених траума скорова, као независни предиктори у оцени тежине трауме издвојили су се: систолни крвни притисак, SaO2 у периферној крви мерена пулсном оксиметријом, повреда главе и врата и повреда грудног коша. За предвиђање исхода лечења повређеног само се SaO2 у периферној крви мерена пулсном оксиметријом издвојила као појединачни предиктор, али не и као независни предиктор исхода.</p> / <p>Teška trauma se može definisati na nekoliko različitih načina. Najčešće korišćena definicija uključuje korišćenje Injury Severity Score ( ISS skor). Ako je vrednost ISS skor &gt;15, radi se o teškoj traumi. Trauma je vremenski osetljivo stanje, zbog toga je za zbrinjavanje teško povređenih pacijenata neophodna dobra saradnja različitih nivoa zdravstvene zaštite i zdravstvenih stručnjaka različitih specijalnosti. U prethodnih nekoliko decenija, zbog složenijeg procesa lečenja i velikih troškova, došlo je do potrebe za što objektivnijom procenom stanja povređenog i ishoda lečenja. Cilj: Uporediti senzitivnost i specifičnost T-RTS skora (Triage Revised Trauma Score), CRAMS skale (Circulation, Respiration, Abdomen, Motor and Speech), MGAP skora (Mechanism, Glasgow coma scale, Age, and arterial Pressure) i GAP skora (Glasgow coma scale, Age, and arterial Pressure) primenjenih na prehospitalnom nivou, proceniti mogućnosti prehospitalno primenjenih RTS, CRAMS, MGAP i GAP skorova u predikciji ishoda lečenja povređenog pacijenta i utvrditi značaj pojedinačnih faktora, određivanih na prehospitalnom nivou tokom inicijalnog pregleda povređenog, za procenu težine povrede i predikciju ishoda lečenja povređenog. Istraživanje je prospektivnog, opservacionog karaktera. U istraživanje su uključeni pacijenti stariji od 18 godina, koje su lekari Zavoda za hitnu medicinsku pomoć Novi Sad (ZZHMP NS) i Službe hitne medicinske pomoći Doma zdravlja Bečej (SHMP DZ Bečej) zbrinjavali na terenu nakon traume, a potom ih transportovali u Urgentni centar Kliničkog centra Vojvodine (UC KCV). Na osnovu vrednosti ISS skora pacijenti su svrstani u jednu od dve grupe: grupa A- pacijenti kod kojih je ISS skor nakon završene dijagnostike iznad 15- teška trauma i grupa B -pacijenti kod kojih je nakon završene dijagnostike ISS skor &le;15- laka trauma. U grupi A je bilo 50, a u grupi B 257 pacijenata. Za granične vrednosti skorova koje označavaju da trauma nije laka, najveću senzitivnost u oceni težine traume je imao GAP skor 98,8%, a najveću specifičnost MGAP skor 62%. U predviđanju ishoda lečenja, najveću senzitivnost je imao RTS skor za predikciju 95,2%, a specifičnost GAP skor i CRAMS skala 87,5%. MGAP skor, a pošto je u snažnoj korelaciji sa njim i GAP skor, mereni prehospitalno, su se izdvojili kao nezavisni prediktor u oceni težine traume i predviđanju ishoda lečenja povređenog. T- RTS skor i CRAMS skala su se izdvojili kao pojedinačni prediktori u oceni težine traume, ali ne i kao nezavisni prediktori. RTS skor za predikciju nema statističku značajnost u predviđanju ishoda lečenja povređenog, za razliku od CRAMS skale koja ima, ali se nije izdvojila kao nezavisni prediktor ishoda lečenja. Osim navedenih trauma skorova, kao nezavisni prediktori u oceni težine traume izdvojili su se: sistolni krvni pritisak, SaO2 u perifernoj krvi merena pulsnom oksimetrijom, povreda glave i vrata i povreda grudnog koša. Za predviđanje ishoda lečenja povređenog samo se SaO2 u perifernoj krvi merena pulsnom oksimetrijom izdvojila kao pojedinačni prediktor, ali ne i kao nezavisni prediktor ishoda.</p> / <p>Severe trauma could be defined in several ways. The most commonly used definition includes Injury Severity Score (ISS) and severe trauma is determined if ISS &gt;15. Trauma management is a time sensitive issue and a coordination between different levels of health system and many specialists is vital in the treatment of severe trauma. In the last decades, a need for the objective evaluation of the severity of trauma and its outcome was perceived due to the complex management and treatment of trauma and its costs. Aim of the study: to compare the sensitivity and specificity between prehospital scores T-RTS (Revised Trauma Score), CRAMS (Circulation, Respiration, Abdomen, Motors, Speech), MGAP (Mechanism, Glasgow Coma Scale, Age, Arterial Pressure) and GAP (Glasgow Coma Scale, Age, Arterial Pressure), to assess the predictability of prehospital scores (RTS, CRAMS, MGAP and GAP) in the outcome of traumatized patients, to determine the significance of individual factors, initially determined during the prehospital evaluation, in evaluating the severity of trauma and the outcome of treatment. Patients enrolled into this prospective observational study were older than 18, prehospitally treated on the trauma site by the doctors of the Institute of the Emergency Medicine Novi Sad and Health Centre Bečej &ndash; Emergency Medical Service and afterward transported into the Emergency Centre Novi Sad. Based on ISS values, patients were divided into two groups: Group A &ndash; severe trauma (50 patients; ISS&gt;15) and Group B &ndash; mild trauma (257 patients; ISS&le;15). For the broder values of scores, determining the severity of trauma, GAP had the highest sensitivity (98%), while MGAP had the highest specificity (62%). RTS had the highest sensitivity in predicting the outcome (95.2%), while GAP and CRAMS had specificity of 87.5%. Prehospital MGAP score, in strong correlation with GAP, was singled out for its independent predictive value in determining the severity of trauma and its outcome. T-RTS and CRAMS stood out to be individual &ndash; but not independent &ndash; predictors in evaluating the severity of trauma. RTS was not statistically significant in predicting the outcome, in contrast with CRAMS. However, CRAMS was not singled out as an independent predictor of the outcome. In addition to the scores, independent predictors of the severity of trauma were: systolic blood pressure, arterial oxygen saturation (SaO2) by using the pulse oximeter, head, neck and thorax injuries. Only SaO2 proved to be a single &ndash; but not independent &ndash; predictor of the outcome.</p>

Submitted by Augusto João Américo de Sousa (augustojas@yahoo.com.br) on 2018-08-31T00:51:15Z No. of bitstreams: 1 DissertaçãoFinalMestradoAugusto_Teletriagem2017.pdf: 1598011 bytes, checksum: 1ada674249fd062716dbb942fa347a79 (MD5) / Approved for entry into archive by ÁUREA CORRÊA DA FONSECA CORRÊA DA FONSECA (aurea.fonseca@fgv.br) on 2018-09-03T20:28:35Z (GMT) No. of bitstreams: 1 DissertaçãoFinalMestradoAugusto_Teletriagem2017.pdf: 1598011 bytes, checksum: 1ada674249fd062716dbb942fa347a79 (MD5) / Made available in DSpace on 2018-09-11T17:55:38Z (GMT). No. of bitstreams: 1 DissertaçãoFinalMestradoAugusto_Teletriagem2017.pdf: 1598011 bytes, checksum: 1ada674249fd062716dbb942fa347a79 (MD5) Previous issue date: 2017-11-30 / Objetivo – Esta dissertação teve como principal objetivo estudar a existência ou não de benefícios aos usuários na adoção da teletriagem nos sistemas de saúde. Metodologia – Como método optou-se por um estudo exploratório através de pesquisa bibliográfica buscando dados na literatura de experiências internacionais que pudessem fundamentar este estudo. Comparou-se os resultados de serviços de teletriagem implantados em 5 países. Aplicando a metodologia de Kim e Mauborgne buscou-se identificar o tipo de inovação introduzida e seu impacto nos custos e benefícios advindos da inovação. Resultados – Os resultados da análise dos dados mostram que os serviços de teletriagem das experiências internacionais conseguiram reduzir a utilização dos serviços de urgência e emergência e levaram os usuários a priorizar a atenção primária nos sistemas de saúde. Limitações – A principal limitação da pesquisa é que se baseia em experiências internacionais. Houve limitação de tempo e recursos para realização de testes laboratoriais, experimentais e para desenvolvimento de projetos pilotos que pudessem ser aplicado numa pesquisa de campo. Aplicabilidade do trabalho – Os resultados do trabalho permitiram recomendar sua aplicação em estudos com testes e pesquisas de campo, experimentos e simulações realísticas, em serviços de telessaúde no Brasil. Contribuições para a sociedade - O estudo sugere que os serviços de teletriagem teriam potencial de gerar benefícios para os usuários do Sistema Único de Saúde gerando educação em saúde, orientação para o autocuidado e fortalecimento das políticas de atenção primária em saúde. Originalidade – Pelo nosso conhecimento, este é o primeiro estudo que relaciona com teletriagem para os sistemas universais de saúde no Brasil. / Purpuse - This thesis of masters had as main objective to study the existence or not of benefits to the users in the adoption of the teletriage in the healthcare systems. Design/Methodology - As method, we opted for an exploratory study through bibliographical research seeking data in the literature of international experiences that could base this study. The results of teletriage services implemented in 5 countries were compared. Applying the Kim and Mauborgne methodology sought to identify the type of innovation introduced and its impact on the costs and benefits of innovation. Findings - The results of the data analysis show that the telephone triage services of the international experiences have managed to decrease the use of emergency/urgent care services and have led users to prioritize primary care in health systems. Research limitations - The main limitation of the research is that it is based on international experiences. There was limited time and resources for laboratory and experimental tests and pilot project development that could be applied in fieldwork. Practical implications - The results of the study allowed to recommend its application in studies with tests and fieldwork, experiments and realistic simulations, in telehealth services in Brazil. Social implications - The study suggests that teletriage services would have the potential to produce benefits for the users of the Brazilian Public Healthcare System, generating health education, orientation for self-care and strengthening of primary health care policies. Originality - To our knowledge, this is the first study that relates to telegraphy for universal health systems in Brazil.

La masse de la baie est le résultat de l’effet intégré de plusieurs facteurs. La recherche a été dessinée afin d’étudier l’effet simultané des facteurs majeurs influençant la masse et la composition de la baie, de les hiérarchiser selon leur degré d’impact à des échelles différentes, de séparer leur effet direct et indirect sur la composition du raisin et de comparer le profil de vins élaborés à partir de petites et grosses baies. L’étude a été conduite sur deux sites expérimentaux, localisés dans les régions de Saint-Emilion (France) et Alcamo (Italie), pendant les années 2014 et 2015. Sur le premier site, les vignes sont plantées sur deux types de sols, tandis que sur le deuxième, deux traitements hydriques étaient appliqués. A l’échelle intra-parcellaire, l’état hydrique de la vigne représente le facteur le plus important, tandis que l’effet du nombre de pépins par baie n’est pas significatif. Des résultats opposés sont obtenus lorsque les relations sont étudiées à l’échelle de la grappe et de la plante. A large échelle, les facteurs impactent directement et indirectement la composition du raisin et les petites baies produisent des moûts et des vins plus concentrés. A l’inverse, à l’échelle de la grappe et de la plante, la masse de la baie n’influence pas la composition du raisin. Seule la concentration en anthocyanes est significativement liée à la masse à toutes les échelles. Cette relation est particulièrement évidente sous conditions hydriques limitantes. Un déficit hydrique augmente le ratio pellicule/pulpe, indépendamment de la masse de la baie. Petites et grosses baies d’une parcelle ayant une condition hydrique homogène, tendent à avoir un profil similaire. / Final berry mass is the result of the integrated effect of several factors. They also influence berry composition. The present work was designed to study the simultaneous effect of major factors influencing berry mass and composition, to hierarchize their impact at different scales, to distinguish their direct and indirect effect on berry composition and to compare the profile of wines made from large and small berries. The study was carried out simultaneously on two vineyards located in the Saint Emil ion (France) and Alcamo (Sicily) areas, during 2014 and 2015. On the first site, vines were planted on two soil types, while on the second site two different irrigation treatments were applied. Depending on the scale, some factors homogeneously impacted the berry mass and composition. At the intra-parcel scale, vine water status represented the most impacting factor, while berry seed number did not have significant effect. Opposite results were obtained when the investigation was carried out at the intra-bunch and intra-plant scales. At large scale, factors impacted directly and indirectly berry compounds and grape juices and wines produced from smaller berries were more concentrated. Neither at intra-bunch, nor at intra-plant scales, berry size effect on juice composition was significant. Only anthocyanin concentration was related to berry size at all scales. This fact was particularly obvious in berries produced under limited water conditions. Water deficit increased the skin to flesh ratio, independently of berry size. This means that small and large berries, produced from a single parcel with homogenous water uptake conditions, tend to have similar enological profiles.

[ES] El triaje de los incidentes de urgencias y emergencias extrahospitalarias representa un reto difícil, debido a las limitaciones temporales y a la incertidumbre. Además, errores en este proceso pueden tener graves consecuencias para los pacientes. Por lo tanto, cualquier herramienta o estrategia novedosa que mejore estos procesos ofrece un valor sustancial en términos de atención al paciente y gestión global de los incidentes. La hipótesis en la que se basa esta tesis es que el Aprendizaje Automático, concretamente el Aprendizaje Profundo, puede mejorar estos procesos proporcionando estimaciones de la gravedad de los incidentes, mediante el análisis de millones de datos derivados de llamadas de emergencia de la Comunitat Valenciana (España) que abarcan desde 2009 hasta 2019. Por tanto, esta tesis profundiza en el diseño y desarrollo de modelos basados en Aprendizaje Profundo Multitarea que aprovechan los datos multimodales asociados a eventos de urgencias y emergencias extrahospitalarias. Nuestro objetivo principal era predecir si el incidente suponía una situación de riesgo vital, la demora admisible de la respuesta y si era competencia del sistema de emergencias o de atención primaria. Utilizando datos disponibles entre 2009 y 2012, se observaron mejoras sustanciales en las métricas macro F1, con ganancias del 12.5% para la clasificación de riesgo vital, del 17.5% para la demora en la respuesta y del 5.1% para la clasificación por jurisdicción, en comparación con el protocolo interno de triaje de la Comunidad Valenciana. Sin embargo, los sistemas, los protocolos de triaje y las prácticas operativas evolucionan de forma natural con el tiempo. Los modelos que mostraron un rendimiento excelente con el conjunto de datos inicial de 2009 a 2012 no demostraron la misma eficacia cuando se evaluaron con datos posteriores que abarcaban de 2014 a 2019. Estos últimos habían sufrido modificaciones en comparación con los anteriores, que dieron lugar a variaciones en las distribuciones de probabilidad, caracterizadas e investigadas meticulosamente en esta tesis. Continuando con nuestra investigación, nos centramos en la incorporación de técnicas de Aprendizaje Continuo Profundo en nuestros desarrollos. Gracias a ello, pudimos mitigar sustancialmente los efectos adversos consecuencia de los cambios distribucionales sobre el rendimiento. Los resultados indican que, si bien las fluctuaciones de rendimiento no se eliminan por completo, pueden mantenerse dentro de un rango manejable. En particular, con respecto a la métrica F1, cuando las variaciones distribucionales son ligeras o moderadas, el comportamiento se mantiene estable, sin variar más de un 2.5%. Además, nuestra tesis demuestra la viabilidad de construir herramientas auxiliares que permitan a los operadores interactuar con estos complejos modelos. En consecuencia, sin interrumpir el flujo de trabajo de los profesionales, se hace posible proporcionar retroalimentación mediante predicciones de probabilidad para cada clase de etiqueta de gravedad y tomar las medidas pertinentes. Por último, los resultados de esta tesis tienen implicaciones directas en la gestión de las urgencias y emergencias extrahospitalarias en la Comunidad Valenciana, al integrarse el modelo final resultante en los centros de atención de llamadas. Este modelo utilizará los datos proporcionados por los operadores telefónicos para calcular automáticamente las predicciones de gravedad, que luego se compararán con las generadas por el protocolo de triaje interno. Cualquier disparidad entre estas predicciones desencadenará la derivación del incidente a un coordinador médico, que supervisará su tratamiento. Por lo tanto, nuestra tesis, además de realizar importantes contribuciones al campo de la Investigación en Aprendizaje Automático Biomédico, también conlleva implicaciones sustanciales para mejorar la gestión de las urgencias y emergencias extrahospitalarias en el contexto de la Comunidad Valenciana. / [CA] El triatge dels incidents d'urgències i emergències extrahospitalàries representa un repte difícil, a causa de les limitacions temporals i de la incertesa. A més, els errors en aquest procés poden tindre greus conseqüències per als pacients. Per tant, qualsevol eina o estratègia innovadora que millore aquests processos ofereix un valor substancial en termes d'atenció al pacient i gestió global dels incidents. La hipòtesi en què es basa aquesta tesi és que l'Aprenentatge Automàtic, concretament l'Aprenentatge Profund, pot millorar significativament aquests processos proporcionant estimacions de la gravetat dels incidents, mitjançant l'anàlisi de milions de dades derivades de trucades d'emergència de la Comunitat Valenciana (Espanya) que abasten des de 2009 fins a 2019. Per tant, aquesta tesi aprofundeix en el disseny i desenvolupament de models basats en Aprenentatge Profund Multitasca que aprofiten dades multimodals d'incidents mèdics d'urgències i emergències extrahospitalàries. El nostre objectiu principal era predir si l'incident suposava una situació de risc vital, la demora admissible de la resposta i si era competència del sistema d'emergències o d'atenció primària. Utilitzant dades disponibles entre 2009 i 2012, es van observar millores substancials en les mètriques macro F1, amb guanys del 12.5% per a la classificació de risc vital, del 17.5% per a la demora en la resposta i del 5.1% per a la classificació per jurisdicció, en comparació amb el protocol intern de triatge de la Comunitat Valenciana. Tanmateix, els protocols de triatge i les pràctiques operatives evolucionen de forma natural amb el temps. Els models que van mostrar un rendiment excel·lent amb el conjunt de dades inicial de 2009 a 2012 no van demostrar la mateixa eficàcia quan es van avaluar amb dades posteriors que abastaven de 2014 a 2019. Aquestes últimes havien sofert modificacions en comparació amb les anteriors, que van donar lloc a variacions en les distribucions de probabilitat, caracteritzades i investigades minuciosament en aquesta tesi. Continuant amb la nostra investigació, ens vam centrar en la incorporació de tècniques d'Aprenentatge Continu als nostres desenvolupaments. Gràcies a això, vam poder mitigar substancialment els efectes adversos sobre el rendiment conseqüència dels canvis distribucionals. Els resultats indiquen que, si bé les fluctuacions de rendiment no s'eliminen completament al llarg del temps, poden mantenir-se dins d'un rang manejable. En particular, respecte a la mètrica F1, quan les variacions distribucionals són lleugeres o moderades, el comportament es manté estable, sense variar més d'un 2.5%. A més, la nostra tesi demostra la viabilitat de construir eines auxiliars que permeten als operadors interactuar amb aquests models complexos. En conseqüència, sense interrompre el flux de treball dels professionals, es fa possible proporcionar retroalimentació mitjançant prediccions de probabilitat per a cada classe d'etiqueta de gravetat i prendre les mesures pertinents. Finalment, els resultats d'aquesta tesi tenen implicacions directes en la gestió de les urgències i emergències extrahospitalàries a la Comunitat Valenciana, al integrar-se el model final resultant als centres d'atenció de telefonades. Aquest model utilitzarà les dades proporcionades pels operadors telefònics per calcular automàticament les prediccions de gravetat, que després es compararan amb les generades pel protocol de triatge intern. Qualsevol disparitat entre aquestes prediccions desencadenarà la derivació de l'incident a un coordinador mèdic, que supervisarà el seu tractament. Per tant, és evident que la nostra tesi, a més de realitzar importants contribucions al camp de la Investigació en Aprenentatge Automàtic Biomèdic, també comporta implicacions substancials per a millorar la gestió de les urgències i emergències extrahospitalàries en el context de la Comunitat Valenciana. / [EN] Triage for out-of-hospital emergency incidents represents a tough challenge, primarily due to time constraints and uncertainty. Furthermore, errors in this process can have severe consequences for patients. Therefore, any novel tool or strategy that enhances these processes can offer substantial value in terms of patient care and overall management of out-of-hospital emergency medical incidents. The hypothesis upon which this thesis is based is that Machine Learning, specifically Deep Learning, can improve these processes by providing estimations of the severity of incidents, by analyzing millions of data derived from emergency calls from the Valencian Region (Spain) spanning from 2009 to 2019. Hence, this thesis delves into designing and developing Deep Multitask Learning models that leverage multimodal out-of-hospital emergency medical data. Our primary objective was to predict whether the incident posed a life-threatening situation, the admissible response delay, and whether it fell under the jurisdiction of the emergency system or primary care. Using data available from 2009 to 2012, the results obtained were promising. We observed substantial improvements in macro F1-scores, with gains of 12.5% for life-threatening classification, 17.5% for response delay, and 5.1% for jurisdiction classification, compared to the in-house triage protocol of the Valencian Region. However, systems, dispatch protocols, and operational practices naturally evolve over time. Models that exhibited excellent performance with the initial dataset from 2009 to 2012 did not demonstrate the same efficacy when evaluated on data spanning from 2014 to 2019. This later dataset had undergone modifications compared to the earlier one, which led to dataset shifts, which we have meticulously characterized and investigated in this thesis. Continuing our research, we incorporated Deep Continual Learning techniques in our developments. As a result, we could substantially mitigate the adverse performance effects consequence of dataset shifts. The results indicate that, while performance fluctuations are not completely eliminated, they can be kept within a manageable range. In particular, with respect to the F1-score, when distributional variations fall within the light to moderate range, the performance remains stable, not varying by more than 2.5%. Furthermore, our thesis demonstrates the feasibility of building auxiliary tools that enable dispatchers to interact with these complex deep models. Consequently, without disrupting professionals' workflow, it becomes possible to provide feedback through probability predictions for each severity label class and take appropriate actions based on these predictions. Finally, the outcomes of this thesis hold direct implications for the management of out-of-hospital emergency medical incidents in the Valencian Region. The final model resulting from our research is slated for integration into the emergency medical dispatch centers of the Valencian Region. This model will utilize data provided by dispatchers to automatically compute severity predictions, which will then be compared with those generated by the in-house triage protocol. Any disparities between these predictions will trigger the referral of the incident to a physician coordinator, who will oversee its handling. Therefore, it is evident that our thesis, in addition to making significant contributions to the field of Biomedical Machine Learning Research, also carries substantial implications for enhancing the management of out-of-hospital emergencies in the context of the Valencian Region. / Ferri Borredà, P. (2024). Deep Continual Multimodal Multitask Models for Out-of-Hospital Emergency Medical Call Incidents Triage Support in the Presence of Dataset Shifts [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/203192

Objetivo: este estudo teve por objetivo verificar a eficácia de um procedimento para a identificação da deficiência auditiva em crianças pré escolares de uma comunidade de baixa renda por pessoas não especialistas. Casuística e métodos: funcionários da creche e da unidade de saúde foram treinados para aplicarem um questionário para triagem auditiva nos pais de 224 crianças de 3 a 6 anos de idade, que foram submetidas à triagem audiométrica e imitanciométrica. Resultados: verificou-se alta concordância na coleta de dados por meio do questionário entre a pesquisadora e os não especialistas. O questionário apresentou melhor sensibilidade e pior especificidade. Os não especialistas relataram aprendizado com a pesquisa e conscientização sobre a importância da audição. Conclusão: é possível treinar agentes comunitários a identificarem a deficiência auditiva em suas comunidades com instrumentos de baixo custo. / Aim: this study aimed to access the efficacy of a low cost instrument used by non professionals to identify hearing loss in pre-school children in a low income community. Methods: employees of a nursery school and a health unit were trained to use a hearing screening questionnaire with parents of 224 children aged 3 to 6 years. The children underwent audiological screening. Results: the results showed that the non - specialists were capable of reproducing the evaluation of the professional regarding the use of the questionnaire. The questionnaire presented better sensitivity than specificity. The non - specialists reported learning and awareness of the importance of hearing in children. Conclusion: the questionnaire offers a low cost option for hearing screening, specially if it is administered by non professionals resources

Objetivo: este estudo teve por objetivo verificar a eficácia de um procedimento para a identificação da deficiência auditiva em crianças pré escolares de uma comunidade de baixa renda por pessoas não especialistas. Casuística e métodos: funcionários da creche e da unidade de saúde foram treinados para aplicarem um questionário para triagem auditiva nos pais de 224 crianças de 3 a 6 anos de idade, que foram submetidas à triagem audiométrica e imitanciométrica. Resultados: verificou-se alta concordância na coleta de dados por meio do questionário entre a pesquisadora e os não especialistas. O questionário apresentou melhor sensibilidade e pior especificidade. Os não especialistas relataram aprendizado com a pesquisa e conscientização sobre a importância da audição. Conclusão: é possível treinar agentes comunitários a identificarem a deficiência auditiva em suas comunidades com instrumentos de baixo custo. / Aim: this study aimed to access the efficacy of a low cost instrument used by non professionals to identify hearing loss in pre-school children in a low income community. Methods: employees of a nursery school and a health unit were trained to use a hearing screening questionnaire with parents of 224 children aged 3 to 6 years. The children underwent audiological screening. Results: the results showed that the non - specialists were capable of reproducing the evaluation of the professional regarding the use of the questionnaire. The questionnaire presented better sensitivity than specificity. The non - specialists reported learning and awareness of the importance of hearing in children. Conclusion: the questionnaire offers a low cost option for hearing screening, specially if it is administered by non professionals resources

In software development projects, bug triage consists mainly of assigning bug reports to software developers or teams (depending on the project). The partial or total automation of this task would have a positive economic impact on many software projects. This thesis introduces a systematic four-step method to find some of the best configurations of several machine learning algorithms intending to solve the automatic bug assignment problem. These four steps are respectively used to select a combination of pre-processing techniques, a bug report representation, a potential feature selection technique and to tune several classifiers. The aforementioned method has been applied on three software projects: 66 066 bug reports of a proprietary project, 24 450 bug reports of Eclipse JDT and 30 358 bug reports of Mozilla Firefox. 619 configurations have been applied and compared on each of these three projects. In production, using the approach introduced in this work on the bug reports of the proprietary project would have increased the accuracy by up to 16.64 percentage points.