Recaídas del trastorno de espectro de la neuromielitis óptica e influencia estacional en una cohorte de un país ecuatorial / Neuromyelitis optica spectrum disorders relapses and seasonal influence in an equatorial country cohort

Introducción La evidencia sobre la distribución estacional de las recaídas del trastorno del espectro de la neuromielitis óptica (NMOSD), especialmente en países tropicales, es limitada y diversa. Objetivo Evaluar la influencia de las variaciones estacionales en las recaídas del NMOSD en un país localizado sobre la línea ecuatorial. Pacientes y métodos Se llevó a cabo un estudio ecológico, con información retrospectiva de una cohorte de pacientes con NMOSD atendida entre enero de 2003 y diciembre de 2020 en Medellín, Colombia. Se recolectaron datos demográficos y clínicos de los pacientes, así como información sobre variables estacionales y climáticas. Se calculó la frecuencia de recaídas por estación, mes y año, y se realizó una regresión binomial negativa para evaluar la asociación entre el número de recaídas, y las variables estacionales y climáticas. Resultados Se incluyó a 113 pacientes, de los cuales el 89,38% eran mujeres, con una edad media en el momento del diagnóstico de NMOSD de 44,97 (±13,98) años y una mediana de tres recaídas (rango intercuartílico: 1-2). Se registraron 237 recaídas, la mayoría en pacientes seropositivos para anticuerpos antiacuaporina 4 (87,76%) y con mielitis longitudinal extensa como la presentación clínica más común (53,59%). Las recaídas se presentaron con mayor frecuencia durante la segunda temporada lluviosa (28,69%; n = 68), y en los meses de noviembre y diciembre. Sin embargo, en la regresión binomial negativa no se observó una asociación significativa entre el número de recaídas y las variables climáticas y estacionales, los meses y los años. Conclusión Las variables climáticas y los patrones estacionales no muestran una asociación significativa con cambios en el número de recaídas del NMOSD en pacientes residentes en un país localizado sobre la línea ecuatorial. (AU)

INTRODUCTION Information about seasonal distribution of Neuromyelitis optica spectrum disorders (NMOSD) attacks, particularly in tropical countries, has rarely been described and the reported data are diverse. OBJECTIVE. To evaluate influence of seasonal variation in NMOSD relapses in an equatorial country. PATIENTS AND METHODS Exploratory observational, retrospective ecological study in a cohort of patients with NMOSD followed from January 2008 to December 2019. Data of demographic, clinical information, characteristics of relapses and seasonal temporal variation were recorded. Also, the annual, monthly and intra-annual seasonal variation of relapses was quantified. A negative binomial regression was used to estimate the associations between the number of relapses and climatic and temporal variables. RESULTS One hundred thirteen patients were included, most of them were female (89.38%), with a mean age at NMOSD diagnosis was 44.97 (±13.98) and the median of relapses per patient were 2 relapses (IQR 1-3). The patients presented 237 relapses, most of these in AQP4 seropositive patients (87.76%) and longitudinal extensive myelitis was the most frequent type of relapse (53.59%). According to the temporal variation, relapses were more common in the second rainy season (28.69%) during November and December. However, there weren’t significant differences in the number of relapses between seasons and climatic variables in the multivariable model. CONCLUSION. The number of NMOSD relapses in this equatorial country cohort did not exhibit any significant associations with climatic variations, including changes in rainy or dry seasons. (AU)

Training data enhancements for improving colonic polyp detection using deep convolutional neural networks.

BACKGROUND: Over the last years, the most relevant results in the context of polyp detection were achieved through deep learning techniques. However, the most common obstacles in this field are the small datasets with a reduced number of samples and the lack of data variability. This paper describes a method to reduce this limitation and improve polyp detection results using publicly available colonoscopic datasets. METHODS: To address this issue, we increased the number and variety of images from the original dataset. Our method consists on adding polyps to the dataset images. The developed algorithm performs a rigorous selection of the best region within the image to receive the polyp. This procedure preserves the realistic features of the images while creating more diverse samples for training purposes. Our method allows copying existing polyps to new non-polypoid target regions. We also develop a strategy to generate new and more varied polyps through generative adversarial neural networks. Hence, the developed approach enriches the training data, creating automatically new samples with their appropriate labels. RESULTS: We applied the proposed data enhancement over a colonic polyp dataset. Thus, we can assess the effectiveness of our approach through a Faster R-CNN detection model. Performance results show improvements over the polyp detections while reducing the false-negative rate. The experimental results also show better recall metrics in comparison with both the original training set and other studies in the literature. CONCLUSION: We demonstrate that our proposed method has the potential to increase the data variability and number of samples in a reduced polyp dataset, improving the polyp detection rate and recall values. These results open new possibilities for advancing the study and implementation of new methods to improve computer-assisted medical image analysis.