Machine learning in the loop for tuberculosis diagnosis support.

The use of machine learning (ML) for diagnosis support has advanced in the field of health. In the present paper, the results of studying ML techniques in a tuberculosis diagnosis loop in a scenario of limited resources are presented. Data are analyzed using a tuberculosis (TB) therapy program at a health institution in a main city of a developing country using five ML models. Logistic regression, classification trees, random forest, support vector machines, and artificial neural networks are trained under physician supervision following physicians' typical daily work. The models are trained on seven main variables collected when patients arrive at the facility. Additionally, the variables applied to train the models are analyzed, and the models' advantages and limitations are discussed in the context of the automated ML techniques. The results show that artificial neural networks obtain the best results in terms of accuracy, sensitivity, and area under the receiver operating curve. These results represent an improvement over smear microscopy, which is commonly used techniques to detect TB for special cases. Findings demonstrate that ML in the TB diagnosis loop can be reinforced with available data to serve as an alternative diagnosis tool based on data processing in places where the health infrastructure is limited.

Time series forecasting for tuberculosis incidence employing neural network models.

Every effort aimed at stopping the expansion of Tuberculosis is important to national programs' struggle to combat this disease. Different computational tools have been proposed in order to design new strategies that allow managing potential patients and thus providing the correct treatment. In this work, artificial neural networks were used for time series forecasting, which were trained with information on reported cases obtained from the national vigilance institution in Colombia. Three neural models were proposed in order to determine the best one according to their forecasting performance. The first approach employed a nonlinear autoregressive model, the second proposal used a recurrent neural network, and the third proposal was based on radial basis functions. The results are presented in terms of the mean average percentage error, which indicates that the models based on traditional methods show better performance compared to connectionist ones. These models contribute to obtaining dynamic information about incidence, thus providing extra-help for health authorities to propose more strategies to control the disease's spread.

Guía de práctica clínica para el manejo de la neumonía adquirida en la comunidad / Clinical Practice Guideline for the management of community-acquired pneumonia

Resumen La neumonía sigue siendo una de las principales causas de consulta y de hospitalización a la que, además de su un alto impacto en términos de morbilidad y mortalidad, se suma la actual problemática de resistencia a los antimicrobianos, por lo que establecer directrices que permitan su adecuado diagnóstico y tratamiento es de gran importancia para obtener mejores desenlaces clínicos y promover un uso racional de antibióticos en estos pacientes. La presente guía de práctica clínica (GPC) contiene recomendaciones basadas en la evidencia para el diagnóstico y tratamiento de la neumonía adquirida en la comunidad en adultos, las cuales fueron realizadas mediante el proceso de adaptación de GPC basadas en la evidencia para el contexto colombiano.

Abstract Pneumonia continues to be one of the main causes of consultation and hospitalization to which, besides its high impact on morbidity and mortality, the current problem of antimicrobial resistance is added; thus, establishing guidelines that allow its adequate diagnosis and treatment is of great importance to obtain better clinical outcomes and promote a rational use of antibiotics in these patients. This clinical practice guideline (CPG) contains evidence-based recommendations for the diagnosis and treatment of community-acquired pneumonia in adult population; these recommendations were made by means of the process of adaptation of evidence-based CPGs for the Colombian context.

Preliminary Text Analysis from Medical Records for TB Diagnosis Support.

Tuberculosis is an infectious disease that is spread through the air from one person to another and is one of the top ten causes of death in the world according to the World Health Organization. From biomedical engineering, decision support systems based on artificial intelligence have shown advantages for healthcare personnel in tasks such as diagnosis and screening. A specific area of the artificial intelligence is the natural language processing, however, most of these approaches are based on available data. This paper shows the construction of a dataset based on medical records of subjects suspected of tuberculosis. In addition, an initial exploration of the contents of the constructed dataset and how this approach can be followed by a natural language processing to support tuberculosis diagnosis in data demanding scenarios are presented.Clinical Relevance- In some developing countries as Colombia, it is difficult to develop systems based on artificial intelligence due to the availability of data. This proposal holds a strategy to build a dataset to train machine learning models, and to obtain support diagnosis tools, employing natural language from the medical scenario from text written by health professionals in the medical record. In this way, trained models based on this information available can be employed in places where medical infrastructure is precarious.

Within patient genetic diversity of blaKPC harboring Klebsiella pneumoniae in a Colombian hospital and identification of a new NTEKPC platform.

Resistance to carbapenems in Klebsiella pneumoniae has been mostly related with the worldwide dissemination of KPC, largely due to the pandemic clones belonging to the complex clonal (CC) 258. To unravel blaKPC post-endemic clinical impact, here we describe clinical characteristics of 68 patients from a high complexity hospital, and the molecular and genetic characteristics of their 139 blaKPC-K. pneumoniae (KPC-Kp) isolates. Of the 26 patients that presented relapses or reinfections, 16 had changes in the resistance profiles of the isolates recovered from the recurrent episodes. In respect to the genetic diversity of KPC-Kp isolates, PFGE revealed 45 different clonal complexes (CC). MLST for 12 representative clones showed ST258 was present in the most frequent CC (23.0%), however, remaining 11 representative clones belonged to non-CC258 STs (77.0%). Interestingly, 16 patients presented within-patient genetic diversity of KPC-Kp clones. In one of these, three unrelated KPC-Kp clones (ST258, ST504, and ST846) and a blaKPC-K. variicola isolate (ST182) were identified. For this patient, complete genome sequence of one representative isolate of each clone was determined. In K. pneumoniae isolates blaKPC was mobilized by two Tn3-like unrelated platforms: Tn4401b (ST258) and Tn6454 (ST504 and ST846), a new NTEKPC-IIe transposon for first time characterized also determined in the K. variicola isolate of this study. Genome analysis showed these transposons were harbored in different unrelated but previously reported plasmids and in the chromosome of a K. pneumoniae (for Tn4401b). In conclusion, in the blaKPC post-endemic dissemination in Colombia, different KPC-Kp clones (mostly non-CC258) have emerged due to integration of the single blaKPC gene in new genetic platforms. This work also shows the intra-patient resistant and genetic diversity of KPC-Kp isolates. This circulation dynamic could impact the effectiveness of long-term treatments.

Recomendaciones de diagnóstico y tratamiento del ataque cerebrovascular isquémico agudo durante la contingencia por covid-19 / Recommendations for diagnosis and treatment of the acute ischemic stroke during COVID-19 contingency

RESUMEN INTRODUCCIÓN. El tiempo es un factor importante en el tratamiento del ACV agudo, con la pandemia de covid-19 se puede ver afectada la adherencia a las guías. Se han reportado tasas de ACV en presencia de covid-19 alrededor del 5%, además un aumento en el tiempo de consulta de los pacientes en otras regiones del mundo y una mayor ocupación de camas de UCI para patologías respiratorias que usualmente se destinan a pacientes con ACV. Este documento recoge las recomendaciones del comité de Enfermedad Cerebrovascular de la Asociación Colombiana de Neurología basados en la literatura y votadas por consenso. OBJETIVOS. Proporcionar una guía rápida adaptada a los diferentes escenarios de atención de ACV en Colombia, divididos en centro listo, primario y avanzado incluyendo una ruta de atención para paciente con sospecha de covid-19 denominada Código ACV protegido, que incluye el uso de elementos de protección personal para pacientes y personal de salud. PUNTOS PRINCIPALES. Dentro de las recomendaciones principales está el cuestionario de tamizaje al ingreso a urgencias, uso de tapabocas para el paciente, considerar todo Código ACV sospechoso de covid-19, toma de PCR para covid-19 en todos los ACV y llevar a zona buffer. En sospecha de oclusión proximal incluir tomo-grafía de tórax a la evaluación de imágenes. CONCLUSIONES. Establecer pautas de atención durante la pandemia covid-19 permite optimizar protección para pacientes y personal de salud, disminuir retrasos y tratar de evitar que pacientes no reciban el tratamiento adecuado en los diferentes niveles de atención del sistema de salud.

SUMMARY INTRODUCTION. Time is an important factor in the treatment of acute stroke, with the COVID-19 pandemic adherence to guidelines can be affected. Stroke rates in the presence of COVID-19 have been reported around 5%, in addition to an increase in the consultation time of patients in other regions of the world and a greater occupation of ICU beds for respiratory pathologies that are usually intended for patients with stroke. This document contains the recommendations of the Cerebrovascular Disease Committee of the Colombian Neurology Association based on the literature and voted by consensus. OBJECTIVES. Provide a quick guide adapted to the different stroke care scenarios in Colombia, divided into ready, primary and advanced centers, including a care route for a patient with suspected COVID-19 called the protected stroke code, which includes the use of protection elements for the patient and the healthcare team. MAIN FACTS. Among the main recommendations are the screening questionnaire on admission to the emergency department, the use of face masks for the patient, considering all suspected stroke code as COVID-19, taking the PCR for COVID-19 in all strokes and taking them to the buffer zone. In suspected large vessel occlusion, include chest tomography to image evaluation. CONCLUSIONS. Establishing guidelines of care during the COVID-19 pandemic allows optimizing protection for patients and health care personnel, reducing delays and trying to avoid patients not receiving adequate treatment at different levels of attention in the health system.

Factores asociados al abandono de terapia antirretroviral de alta efectividad en pacientes con VIH SIDA en un hospital de tercer nivel / Factors associated with the abandonment of highly effective antiretroviral therapy in patients with HIV in a third-level hospital

Resumen Introducción: la baja adherencia en el tratamiento antirretroviral de alta efectividad TARGA se asocia a peores desenlaces en salud; además de las comorbilidades infecciosas, muchos pacientes con VIH/SIDA presentan asociado psicopatologia, farmacodependencia y problemas socioeconómicos. Existe alta frecuencia de abandono a la terapia en nuestro medio. Objetivo: identificar las variables asociadas al abandono de terapia TARGA en pacientes con VIH de un hospital de tercer nivel en Colombia. Métodos y resultados estudio descriptivo corte transversal: se incluyeron 51 pacientes pertenecientes al programa durante un año; 12 pacientes (23%) tuvieron adherencia al tratamiento, 32 (62%) tenían antecedente de consumo de sustancias psicoactivas y 33 (64%) tenían asociada alguna patología psiquiátrica. Los factores que se asociaron a la no adherencia fueron la psicopatología y la pobre red de apoyo. Conclusión: En la población a estudio se documenta alta tasa de abandono a la terapia TARGA, probablemente asociado a enfermedad mental y pobre red social. (Acta Med Colomb 2018; 43: 31-36).

Abstract Introduction: the low adherence in the highly effective antiretroviral treatment HAART is associated with worse health outcomes. In addition to infectious comorbidities, many patients with HIV / AIDS have associated psychopathology, drug dependence and socioeconomic problems. There is a high frequency of abandonment of therapy in our environment. Objective: to identify the variables associated with the abandonment of HAART therapy in patients with HIV from a third level hospital in Colombia. Methods and Results: descriptive cross-sectional study: 51 patients belonging to the program during one year were included; 12 patients (23%) had adherence to treatment, 32 (62%) had a history of psychoactive substance use and 33 (64%) had some psychiatric pathology. The factors that were associated with non-adherence were psychopathology and the poor support network. Conclusion: high rate of abandonment of HAART therapy is documented in the study population, probably associated with mental illness and poor social network. (Acta Med Colomb 2018; 43: 31-36).

Descripción de los pacientes con fibrilación auricular no valvular que ingresan al servicio de urgencias / Description of patients with non-valvular atrial fibrillation admitted to the emergency room

Introducción y objetivos: La fibrilación auricular es la arritmia cardiaca más frecuente y una de las complicaciones más temidas es la embolia sistémica y/o el accidente cerebrovascular (ACV). En la actualidad se cuenta con diversas terapias antitrombóticas para prevención del primero, sin embargo, hay temor por complicaciones relacionadas con la anticoagulación, y aunque existen recomendaciones de diferentes guías de prácticas clínicas, continúa el uso subóptimo de la misma. El objetivo del estudio es describir datos clínicos, diagnósticos y terapéuticos de pacientes con fibrilación auricular que ingresan al servicio de urgencias, además, se realizó una correlación entre el riesgo de embolia sistémica y de sangrado con el tipo de terapia recibida. Métodos: Estudio descriptivo retrospectivo tipo serie de casos de pacientes con diagnóstico de la fibrilación auricular no valvular durante un año. Se reportan las frecuencias de comorbilidades, diagnósticos de ingreso y tipo de terapia específica. Resultados: Se incluyeron 98 pacientes, la media de edad fue 72 años, la falla cardiaca y la hipertensión arterial fueron las principales comorbilidades; el 92% tenían indicación de anticoagulación por la escala CHA2DS2VASc, 54% de estos pacientes no la venían recibiendo a pesar de que se calculó en la mayoría bajo riesgo de sangrado. El 5% de los pacientes ingresaron con diagnóstico de ACV isquémico y el 5% presentaban sobreanticogulación. Conclusión: Las características de nuestra población con la fibrilación auricular es similar a la descrita en la literatura, pero existe baja adherencia a las recomendaciones terapéuticas en anticoagulación.

Introduction and motivation: Atrial fibrillation is the most common cardiac arrhythmia. One of its most feared complications is the systemic embolism and/or the cerebrovascular accident (CVA) or stroke. Nowadays several antithrombotic therapies are available for the former; however, there are fears regarding complications related to anticoagulation, and even though recommendations of different clinical practice guidelines exist, blood-thinning drugs are still being underused. The goal of this study is to describe clinical, diagnostic and therapeutic data of patients admitted in the emergency room with atrial fibrillation; in addition, a correlation was made between the risk of systemic embolism and bleeding depending on the received treatment. Methods: Retrospective, descriptive study of case series of patients diagnosed with non-valvular atrial fibrillation during a year. Comorbidity frequencies, admission diagnoses and specific therapy types are reported. Results: 98 patients were included, average age was 72 years old, cardiac failure and high blood pressure were the main comorbidities; 92% showed an indication for anticoagulation therapy according to the CHA2DS2-VASc scale, 54% of those patients had not been receiving this therapy despite a low risk for bleeding was established for most of them. 5% of the patients received an admission diagnosis of ischemic stroke and 5% showed overanticoagulation. Conclusion: The characteristics of our sample with atrial fibrillation are similar to those described in the literature, but there is a low adherence to the treatment recommendations for anticoagulation therapy.

Proteinosis alveolar pulmonar: reporte de caso y revisión de la literatura / Alveolar pulmonary proteinosis: case report and literature review

Se describe el caso de una mujer joven con proteinosis alveolar primaria, con cuadro clínico poco frecuente dado la corta duración de la sintomatología, sin factores de riesgo para la entidad, con su evolución clínica y algunas complicaciones relacionadas con el tratamiento. Se realiza además una breve revisión de la literatura sobre la entidad.