Deep Learning Based Classification of True/False Arrhythmia Alarms in the Intensive Care Unit

Objective.Surveillance of nosocomial bloodstream infection (BSI) is recommended, but time-consuming. We explored strategies for automated surveillance.Methods.Cohort study. We prospectively processed microbiological and administrative patient data with computerized algorithms to identify contaminated blood cultures, community-acquired BSI, and hospital-acquired BSI and used algorithms to classify the latter on the basis of whether it was a catheter-associated infection. We compared the automatic classification with an assessment (71% prospective) of clinical data.Setting.An 850-bed university hospital.Participants.All adult patients admitted to general surgery, internal medicine, a medical intensive care unit, or a surgical intensive care unit over 3 years.Results.The results of the automated surveillance were 95% concordant with those of classical surveillance based on the assessment of clinical data in distinguishing contamination, community-acquired BSI, and hospital-acquired BSI in a random sample of 100 cases of bacteremia. The two methods were 74% concordant in classifying 351 consecutive episodes of nosocomial BSI with respect to whether the BSI was catheter-associated. Prolonged episodes of BSI, mostly fungemia, that were counted multiple times and incorrect classification of BSI clinically imputable to catheter infection accounted for 81% of the misclassifications in automated surveillance. By counting episodes of fungemia only once per hospital stay and by considering all cases of coagulase-negative staphylococcal BSI to be catheter-related, we improved concordance with clinical assessment to 82%. With these adjustments, automated surveillance for detection of catheter-related BSI had a sensitivity of 78% and a specificity of 93%; for detection of other types of nosocomial BSI, the sensitivity was 98% and the specificity was 69%.Conclusion.Automated strategies are convenient alternatives to manual surveillance of nosocomial BSI.

Download Full-text

Laryngomalacia in Neonates Versus Older Infants: HCUP-KID Perspective

Clinical Pediatrics ◽

10.1177/0009922820908917 ◽

2020 ◽

Vol 59 (7) ◽

pp. 679-685

Author(s):

Michele M. Carr ◽

Jad Ramadan ◽

Emma Bauer

Keyword(s):

Intensive Care Unit ◽

Intensive Care ◽

Mortality Rate ◽

Neonatal Intensive Care ◽

International Classification Of Diseases ◽

Classification Of Diseases ◽

Cost Utilization ◽

Inpatient Database ◽

Total Charges

This study evaluated the hospital course for neonates and older infants with a diagnosis of laryngomalacia (LM). Data came from the 2016 Kids’ Inpatient Database of the Healthcare Cost Utilization Project. A total of 6537 children aged <1 year with a diagnosis of LM (International Classification of Diseases, 10th Revision, code Q31.5) were identified: 2212 neonates and 4325 non-neonates. Neonates had a higher mortality rate, 1.31% versus 0.72% in older infants, had more diagnoses (median 9 vs 7) and procedures (mean 85.24 vs 21.83), longer length of stay (median 10 vs 4 days), and higher total charges (median US$65 722 vs US$25 582). A total of 23.3% of neonates born during the admission and diagnosed with LM had undergone laryngoscopy. Second airway lesions were present in 12.33% of neonates and 15.77% of older infants. It appears that neonates are being discharged with a diagnosis of LM without laryngoscopy. Neonatal intensive care unit and newborn nursery policies should require visualization of the larynx prior to diagnosis of LM.

Download Full-text

A Deep Learning Model for Early Prediction of Sepsis from Intensive Care Unit Records

Communications in Computer and Information Science - Neural Information Processing ◽

10.1007/978-3-030-63820-7_90 ◽

2020 ◽

pp. 791-798

Author(s):

Rui Zhao ◽

Tao Wan ◽

Deyu Li ◽

Zhengbo Zhang ◽

Zengchang Qin

Keyword(s):

Intensive Care Unit ◽

Intensive Care ◽

Deep Learning ◽

Learning Model ◽

Early Prediction ◽

Deep Learning Model

Download Full-text

New Atlanta Classification of acute pancreatitis in intensive care unit: Complications and prognosis

European Journal of Internal Medicine ◽

10.1016/j.ejim.2016.01.007 ◽

2016 ◽

Vol 30 ◽

pp. 82-87 ◽

Cited By ~ 10

Author(s):

María-Consuelo Pintado ◽

María Trascasa ◽

Cristina Arenillas ◽

Yaiza Ortiz de Zárate ◽

Ana Pardo ◽

...

Keyword(s):

Intensive Care Unit ◽

Acute Pancreatitis ◽

Intensive Care

Download Full-text

Construção do banco de termos da linguagem especial de enfermagem de uma unidade de terapia intensiva com base no modelo de sete eixos da cipe® versão 1.0

Revista de Enfermagem UFPE on line ◽

10.5205/reuol.149-181-1-rv.0303200908 ◽

2009 ◽

Vol 3 (3) ◽

pp. 496

Author(s):

Elizabeth Vasconcelos Trigueiro ◽

Ana Claudia Torres de Medeiros ◽

Maria Miriam Lima da Nóbrega ◽

Telma Ribeiro Garcia

Keyword(s):

Intensive Care Unit ◽

Intensive Care ◽

Continuity Of Care ◽

Clinical Area ◽

Exploratory Investigation ◽

Terapia Intensiva

ABSTRACTObjective: to build the Bank of Terms of Language of Special Nursing for Intensive Care Unit of the HULW/UFPB, based on Model of Seven Axes from ICNP® Version 1.0. Method: descriptive exploratory investigation, developed in the stages following: assessment of collection of special nursing terms take out nursing registrations of Intensive Care Unit of the HULW/UFPB; establishment of the tree of special nursing terms, for axes of ICNP® Version 1.0, and the classification of terms respects as constants and not constants set according to the axes of the ICNP® Version 1.0. Results: the terms identified were classified as constants and not constants in ICNP® Version 1.0. After the review was began the establishment of trees in this clinical area e and finally there was the construction of the bank consisted of 213 terms constants and 78 non-constants terms. Conclusion: it was concludes that the objective of the study was achieved and it is expected that their use contributes to the nursing unit’s use their own vocabulary in the practice, valuing the systematic registration of nursing actions and favoring the continuity of care. Descriptors: nursing; terminology; language; intensive care.RESUMOObjetivo: construir o Banco de Termos da Linguagem Especial de Enfermagem para a Unidade de Terapia Intensiva do HULW/UFPB, fundamentado no Modelo de Sete Eixos da CIPE® Versão 1.0. Método: pesquisa exploratória descritiva, desenvolvida nas seguintes etapas: avaliação do conjunto de termos da linguagem especial de enfermagem, extraídos de registros de enfermagem da Unidade de Terapia Intensiva do HULW/UFPB; estabelecimento da árvore de termos da linguagem especial de enfermagem, por eixo da CIPE® Versão 1.0, e a classificação dos termos considerados como constantes e não constantes de acordo com os eixos da CIPE® Versão 1.0. Resultados: os termos identificados foram classificados como constantes e não constantes na CIPE® Versão 1.0. Após a análise iniciou-se o estabelecimento das árvores de domínio desta clínica e por fim se deu a construção do banco constituído por 213 termos constantes e 78 termos não constantes. Conclusão: conclui-se que o objetivo do estudo foi alcançado e espera-se que a sua utilização contribua para que a equipe de enfermagem da unidade utilize seu próprio vocabulário na prática profissional, valorizando o registro sistematizado das ações de enfermagem e favorecendo a continuidade da assistência. Descritores: enfermagem; terminologia; linguagem; cuidados intensivos.RESUMENObjetivo: construir el Banco de Términos del Lenguaje Especial de la Enfermería para la Unidad de los Cuidados Intensivos de HULW/UFPB, basado en el Modelo de Siete Ejes CIPE® Versión 1.0. Método: investigación descriptiva exploratoria, desarrollada en los siguientes pasos: evaluación del conjunto de términos del lenguaje especial de la enfermería, sacados de los registros de enfermería de la Unidad de Cuidados Intensivos de HULW/UFPB, establecer los términos del lenguaje especial de enfermería, del eje CIPE® Versión 1.0 y la clasificación de los términos considerados como constantes y no constantes de acuerdo con los ejes de la CIPE® Versión 1.0. Resultados: los términos identificados fueron clasificados como constantes y no constantes en la CIPE® Versión 1.0. Después del análisis empezó el establecimiento de árboles en esta área clínica y por último, ocurrió la construcción del banco constando de 213 términos constantes y 78 no constantes. Conclusión: se concluye que el objetivo del estudio fue alcanzado y esperase que su utilización va a contribuir para que el equipo de enfermería de la unidad utilice su propio vocabulario en la práctica profesional, valorizando el registro sistematizado de las acciones de enfermería y la promoción de la continuidad de la atención. Descriptores: enfermería, terminología, lenguaje, cuidados intensivos.

Download Full-text

Contribution of Heritable Disorders to Mortality in the Pediatric Intensive Care Unit

PEDIATRICS ◽

10.1542/peds.95.5.678 ◽

1995 ◽

Vol 95 (5) ◽

pp. 678-681

Author(s):

Christopher Cunniff ◽

Janet L. Carmack ◽

Russell S. Kirby ◽

Debra H. Fiser

Keyword(s):

Intensive Care Unit ◽

Intensive Care ◽

Medical Record ◽

Cause Of Death ◽

Pediatric Intensive Care ◽

Vital Statistics ◽

Heritable Disorder ◽

Underlying Cause Of Death ◽

Record Review

Objectives. To determine the percentage of patients dying in the pediatric intensive care unit (PICU) who have heritable disorders and to compare vital statistics classification of underlying cause of death with underlying heritable disorder identified from medical record review. Design. Retrospective medical record review. Setting. The PICU of a university-affiliated hospital. Methods. Medical records were reviewed for all deaths occurring in the PICA over a 5-year period. Further review, including hospital course, clinical findings, and the presence or absence of a genetic evaluation, was accomplished for those patients found to have a chromosome abnormality, recognized syndrome, single major malformation, or unrecognized syndrome. Underlying cause of death classification obtained from the Center for Health Statistics, Arkansas Department of Health was reviewed to determine the frequency with which the underlying heritable disorder was recorded. Results. Fifty-one of 268 (19%) deaths during the study period were in patients with heritable disorders. Of these 51 patients, eight (16%) had chromosome abnormalities, 17 (33%) had a recognized syndrome, 15 (29%) had a single primary defect in development, and 11 (22%) had an unrecognized syndrome. Genetic evaluation was carried out on 45% of patients, with the frequency of evaluation differing between categories of patients with heritable conditions. When underlying cause of death from vital statistics classification was reviewed, 21 of 51 (41%) records did not include the underlying heritable disorder. Conclusions. Heritable disorders are a frequent cause of mortality in the PICU. Vital statistics classification of underlying cause of death in this population often fails to identify heritable disorders, leading to an underascertainment of these conditions in mortality statistics. Improved cause of death classification procedures will be necessary to target public health interventions to etiology-specific populations.

Download Full-text