A stacked autoencoder-based aid system for severity degree classification of knee ligament rupture.

BACKGROUND: Knee ligament rupture is one of the most common injuries, but the diagnosis of its severity tends to require the use of complex methods and analyses that are not always available to patients. AIM: The objective of this research is the investigation and development of a diagnostic aid system to analyze and determine patterns that characterize the presence of the injury and its degree of severity. METHODS: Implement a novel proposal of a framework based on stacked auto-encoder (SAE) for ground reaction force (GRF) signals analysis, coming from the GaitRec database. Analysis of the raw data is used to determine the main features that allow us to diagnose the presence of a knee ligament rupture and classify its severity as high, mid or mild. RESULTS: The process is divided into two stages to determine the presence of the lesion and, if necessary, evaluate variations in features to classify the degree of severity as high, mid, and mild. The framework presents an accuracy of 87 % and a F1-Score of 90 % for detecting ligament rupture and an accuracy of 86.5 % and a F1-Score of 87 % for classifying severity. CONCLUSION: This new methodology aims to demonstrate the potential of SAE in physiotherapy applications as an evaluation and diagnostic tool, identifying irregularities associated with ligament rupture and its degree of severity, thus providing updated information to the specialist during the rehabilitation process.

Umbral del ciclo de amplificación del gen de la toxina B de Clostridioides difficile como predictor de gravedad / Clostridioides difficile toxin B gen amplification cycle threshold as a predictor of severity

INTRODUCCIÓN: La infección por Clostridioides dfficile (ICD) es la principal causa de diarrea nosocomial, generalmente asociada al consumo de antimicrobianos. Esta infección puede causar desde diarrea no complicada hasta colitis pseudomembranosa o megacolon tóxico. Estudios recientes han intentado relacionar el valor el ciclo umbral (Ct) de la RT-PCR con la mortalidad, como un método rápido, sencillo, objetivo y eficaz. OBJETIVO: Evaluar el Ct como predictor de mala evolución en pacientes con y sin criterio clínico de dicha gravedad. PACIENTES Y MÉTODOS: Realizamos un estudio retrospectivo entre enero 2015 y diciembre 2018, incluyendo todos los pacientes del área de referencia del Hospital Universitario de Canarias en Tenerife (396.483 habitantes) en pacientes con criterios clínicos de gravedad (de acuerdo a la Guía para la Práctica Clínica de la enfermedad por C. dfficile de la Sociedad de Epidemiología del Cuidado de la Salud de América (SHEA) y la Sociedad de Enfermedades Infecciosas de Norteamérica (IDSA) y pacientes sin criterios clínicos de gravedad evaluando el Ct como predictor de mala evolución. RESULTADOS: Se diagnosticó un total de 202 episodios de ICD. El 77,7% (n = 157) presentó criterios clínicos de gravedad. La presencia de colitis ulcerosa (p < 0,001), fiebre (p < 0,001), leucocitosis (p < 0,001), neutrofilia (p < 0,001), creatininemia (p = 0,005) se presentaron como factores de riesgo para el desarrollo de ICD grave. El sexo femenino, la institucionalización, el ingreso previo y el exitus se describieron con mayor frecuencia en el grupo con ICD-G, no encontrando diferencias significativas. No encontramos diferencias respecto a los días de estancia previa, o de estancia post-ICD, aunque en este último, la media fue mayor en el caso de los pacientes con ICD-G. No se encontraron diferencias significativas en cuanto al Ct en ambos grupos; siendo sólo un punto menor en pacientes con criterio de gravedad (Ct = 26,1) que sin criterios de gravedad (Ct = 27,4) (p = 0,326).

BACKGROUND: Clostridioides dfficile infection (CDI) is the main cause of nosocomial diarrhea, generally associated with the use of antibiotics. This infection can cause uncomplicated diarrhea to pseudomembranous colitis or toxic megacolon. Recent studies have attempted to relate the threshold cycle (Ct) value of RT-PCR with mortality, as a fast, simple, objective and efficient method. AIM: To evaluate Ct as a predictor of poor outcome in patients with C. dfficile disease with/without clinical signs of severity. METHODS: We carried out a retrospective study between January 2015 and December 2018, including all patients in the reference area of the Hospital Universitario de Canarias in Tenerife (396,483 inhabitants) in patients with clinical criteria of severity and patients without clinical severity criteria (according to the guide for the clinical practice of CDI of the Society of Healthcare Epidemiology of America (SHEA) and the Infectious Diseases Society of North America (IDSA). RESULTS: A total of 202 CDI episodes were diagnosed. 77.7% (n = 157) presented clinical severity criteria. The presence of ulcerative colitis (p < 0.001), fever (p < 0.001), leukocytosis (p < 0.001), neutrophilia (p < 0.001), creatininemia (p = 0.005) were presented as risk factors for the development of severe CDI (S-CDI). Female sex, institutionalization, previous admission and death were described more frequently in the group with S-CDI, not finding significant differences. We found no differences with respect to the days of previous stay, or of post-CDI stay, although in the latter, the mean was higher in the case of S-CDI patients. No significant differences were found in terms of Ct in both groups; being only one point lower in patients with severity criteria (Ct = 26.1) than without severity criteria (Ct = 27.4) (p = 0.326). CONCLUSION: Based on the results of our study, it has not been possible to systematically implement the Ct value as a predictor of severity to the clinical report, and it is not possible to extrapolate this predictive variable from S-CDI and standardize the Ct value as a predictor of severity. Conclusion: Basándonos en los resultados de nuestro estudio, no ha sido posible la implementación sistemática del valor Ct como predictor de gravedad al informe clínico, no siendo posible extrapolar esta variable predictora de enfermedad por C difficile-G y estandarizar el valor Ct como factor predictor de gravedad.

Distinct systemic immune networks define severe vs. mild COVID-19 in hematologic and solid cancer patients.

Introduction: The COVID-19 pandemic, caused by the coronavirus SARS-CoV-2, has impacted health across all sectors of society. A cytokine-release syndrome, combined with an inefficient response of innate immune cells to directly combat the virus, characterizes the severe form of COVID-19. While immune factors involved in the development of severe COVID-19 in the general population are becoming clearer, identification of the immune mechanisms behind severe disease in oncologic patients remains uncertain. Methods: Here we evaluated the systemic immune response through the analysis of soluble blood immune factors and anti-SARS-CoV-2 antibodies within the early days of a positive SARS-CoV-2 diagnostic in oncologic patients. Results: Individuals with hematologic malignancies that went on to die from COVID-19 displayed at diagnosis severe leukopenia, low antibody production against SARS-CoV-2 proteins, and elevated production of innate immune cell recruitment and activation factors. These patients also displayed correlation networks in which IL-2, IL-13, TNF-alpha, IFN-gamma, and FGF2 were the focal points. Hematologic cancer patients that showed highly networked and coordinated anti-SARS-CoV-2 antibody production, with central importance of IL-4, IL-5, IL-12A, IL-15, and IL-17A, presented only mild COVID-19. Conversely, solid tumor patients that had elevated levels of inflammatory cytokines IL-6, CXCL8, and lost the coordinate production of anti-virus antibodies developed severe COVID-19 and died. Patients that displayed positive correlation networks between anti-virus antibodies, and a regulatory axis involving IL-10 and inflammatory cytokines recovered from the disease. We also provided evidence that CXCL8 is a strong predictor of death for oncologic patients and could be an indicator of poor prognosis within days of the positive diagnostic of SARS-CoV-2 infection. Conclusion: Our findings defined distinct systemic immune profiles associated with COVID-19 clinical outcome of patients with cancer and COVID-19. These systemic immune networks shed light on potential immune mechanisms involved in disease outcome, as well as identify potential clinically useful biomarkers.

A Bibliometric Analysis and Benchmark of Machine Learning and AutoML in Crash Severity Prediction: The Case Study of Three Colombian Cities.

Traffic accidents are of worldwide concern, as they are one of the leading causes of death globally. One policy designed to cope with them is the design and deployment of road safety systems. These aim to predict crashes based on historical records, provided by new Internet of Things (IoT) technologies, to enhance traffic flow management and promote safer roads. Increasing data availability has helped machine learning (ML) to address the prediction of crashes and their severity. The literature reports numerous contributions regarding survey papers, experimental comparisons of various techniques, and the design of new methods at the point where crash severity prediction (CSP) and ML converge. Despite such progress, and as far as we know, there are no comprehensive research articles that theoretically and practically approach the model selection problem (MSP) in CSP. Thus, this paper introduces a bibliometric analysis and experimental benchmark of ML and automated machine learning (AutoML) as a suitable approach to automatically address the MSP in CSP. Firstly, 2318 bibliographic references were consulted to identify relevant authors, trending topics, keywords evolution, and the most common ML methods used in related-case studies, which revealed an opportunity for the use AutoML in the transportation field. Then, we compared AutoML (AutoGluon, Auto-sklearn, TPOT) and ML (CatBoost, Decision Tree, Extra Trees, Gradient Boosting, Gaussian Naive Bayes, Light Gradient Boosting Machine, Random Forest) methods in three case studies using open data portals belonging to the cities of Medellín, Bogotá, and Bucaramanga in Colombia. Our experimentation reveals that AutoGluon and CatBoost are competitive and robust ML approaches to deal with various CSP problems. In addition, we concluded that general-purpose AutoML effectively supports the MSP in CSP without developing domain-focused AutoML methods for this supervised learning problem. Finally, based on the results obtained, we introduce challenges and research opportunities that the community should explore to enhance the contributions that ML and AutoML can bring to CSP and other transportation areas.

Índice de gravedad en neumonía comunitaria hospitalizada / Severit predictiion index in hospitalized communitary acquired pneumonia

Fine's severity prediction index (SPI), was retrospectively analyzed in community acquired pneumonia (CAP), in patients at Concepción Regional Hospital, from June to August 2000. We studied 57 CAP patients: 23 as low risk and 34 as high risk patients. In comparison to low risk patients the main features of high risk patients were: older age (p < 0.00001), higher comorbility (p < 0.004), longer hospitalization (p < 0.0007) and higher mortality (p < 0.018). Mortality in low risk patients was similar to Fine's study: 4.3 versus 3.5%. In high risk patients mortality was 26% versus 38%. Main complications in our series were mechanical ventilation (43.8%), PaO2/FiO2 < 250 mmHg (43.8%), and hepatic coma (38.5%). As a conclusion, we recommend the use of SPI in CAP at Emergency Services in order to better evaluate mortality, optimizing criteria of hospital admission.

Se evaluó el Índice de Gravedad de Fine (IG) en neumonías de la comunidad (NAC) hospitalizadas en el Hospital Regional de Concepción en trabajo retrospectivo con 57 casos entre Junio y Agosto del año 2000. Se estudiaron 23 pacientes catalogados de bajo riesgo y 34 de alto riesgo. Características de alto riesgo fueron, mayor edad (p < 0,00001), mayor comorbilidad (p < 0,0004), estadía prolongada (p < 0,00007) y mayor mortalidad (p < 0,018). La mortalidad de bajo riesgo fue similar a la de Fine, 4,3% versus 3,5%, siendo menor en el grupo de alto riesgo, 26% versus 38%. Factores de mayor complicación en NAC fueron, ventilación mecánica (43,8%), PaO2/FiO2 < 250 mmHg (43,8%) y coma hepático (35,8%). Conclusión: es aconsejable el uso del IG en NAC a nivel de Servicios de Urgencia, para evaluar los riesgos de mortalidad, optimizando así los criterios de ingreso al hospital.