RESUMO
The new coronavirus disease (COVID-19), pneumonia, tuberculosis, and breast cancer have one thing in common: these diseases can be diagnosed using radiological studies such as X-rays images. With radiological studies and technology, computer-aided diagnosis (CAD) results in a very useful technique to analyze and detect abnormalities using the images generated by X-ray machines. Some deep-learning techniques such as a convolutional neural network (CNN) can help physicians to obtain an effective pre-diagnosis. However, popular CNNs are enormous models and need a huge amount of data to obtain good results. In this paper, we introduce NanoChest-net, which is a small but effective CNN model that can be used to classify among different diseases using images from radiological studies. NanoChest-net proves to be effective in classifying among different diseases such as tuberculosis, pneumonia, and COVID-19. In two of the five datasets used in the experiments, NanoChest-net obtained the best results, while on the remaining datasets our model proved to be as good as baseline models from the state of the art such as the ResNet50, Xception, and DenseNet121. In addition, NanoChest-net is useful to classify radiological studies on the same level as state-of-the-art algorithms with the advantage that it does not require a large number of operations.
RESUMO
Una etapa importante y fundamental en el reconocimiento de patrones sobre imágenes es la determinación del conjunto de características que mejor pueda describir la misma. En este artículo se presenta una etapa adicional entre la caracterización de la imagen y su posterior clasificación o recuperación de imágenes similares a una imagen dada, conocido como análisis de relevancia. Este permite reducir la dimensionalidad del conjunto inicial de características a un nuevo conjunto de menor dimensión que conserva la tasa de acierto de la recuperación. Las imágenes analizadas correspondieron a nódulos pulmonares de placas radiológicas de tórax disponibles en una base de datos de acceso libre disponible a través de la sociedad japonesa de tecnología radiológica. Se analizaron algoritmos de selección de características basados en filtros que incluyeron los métodos FOCUS, RELIEEF-F y Branch & Bound (B&B). Estos algoritmos fueron modificados e implementados en C++. En el caso de RELIEF-F se logró obtener un ahorro del 34% de características sin afectar la tasa de recuperación cuando se empleaba el 100% de las características originales. Asimismo, el algoritmo implementado presentó un desempeño superior al algoritmo original disponible en la herramienta de código abierto Weka. Asimismo se implementó una estrategia de ponderación de pesos aplicada a las características identificadas cuando se utilizaron los algoritmos RELIEF-F, FOCUS y B&B simultáneamente. Dicha estrategia permitió ponderar cada característica de acuerdo a su participación en los conjuntos mínimos de características relevantes y determinar la consistencia de los mismos. La estrategia de pesos permitió un ahorro del 48% de características necesarias para la recuperación, aunque la tasa de recuperación fue disminuida de 77% a 76%.
An important and fundamental stage in the image pattern recognition is the determination of the characteristics set that best describes the image. This paper describes a further step between the image characterization and its posterior classification or image retrieval similar to a given image, known as relevance analysis. It allows reducing the dimensionality of an initial set of features to a new set with fewer dimensions that preserves the hit rate of the retrieval. The analyzed images corresponded to lung nodules of radiological plaques of thorax, available through the open access library available through the Japanese society of radiological technology. To achieve these results, characteristic selection algorithms based on different filters such as FOCUS, RELIEEF-F, and BRANCH & BOUND (B&B) were analyzed. In the case of RELIEF-F it was possible to save as much as 34% of the initial characteristics set without affecting the retrieval rate compared to when the 100% of characteristics were used. Further, the implemented algorithm achieved a superior performance to that of the original algorithm included in the validated Weka software. Likewise, a strategy consisting in weights averaging was implemented that was applied to identified characteristics when the algorithms RELIEF-F, FOCUS and B&B were used simultaneously. Such weighting scheme, allowed the averaging of each characteristic according to its contribution in the minimal set of relevant features, allowing to determinate their consistency. The weighting strategy allowed a 48% reduction in the characteristics, although the retrieval hit rate slightly decreased from 77% to 76%.
RESUMO
Los catéteres venosos centrales son un recurso terapéutico fundamental debido al advenimiento de unidades de pacientes críticos, donde se utilizan para control de variable hemodinámicas, administración de drogas y nutrición parenteral; a la masificación de la hemodiálisis, siendo una herramienta transitoria previo a la obtención de accesos definitivos y para pacientes oncológicos, con el objetivo de administración de quimioterapia. Sin embargo, el uso de catéteres venosos centrales no es inocuo. En la literatura se reportan diversas complicaciones. Dentro de éstas, las mecánicas son frecuentes, presentándose el 5-19 por ciento de los pacientes. Las complicaciones más prevalerte son la punción arterial, hemotórax y neumotórax. El objetivo de esta revisión es actualizar conceptos básicos relacionados a las complicaciones mecánicas asociadas a la instalación de catéteres venosos centrales y mostrar imágenes obtenidas en el servicio de imagenología de nuestro hospital, correspondientes a ocho pacientes que presentaron alguna de éstas.
Central venous catheters are important therapeutic resources nowadays, because of the development of critical care units, where these devices are aimed to monitorize hemodynamic parameters and to administrate drugs and parenteral nutrition to patients; massification of hemodyalisis, where central venous catheters are used as transitory accesses before the definitive ones are obtained and in the oncologycal setting, to administrate chemotherapeutic drugs to patients. Nevertheless, the use of such devices is not harmless. Several complications are reported in current literature. Mechanical complications are frequent and they are found in 5-19 percent of patients in whom central venous catheters are installed. According to their prevalence, the most important are arterial puncture, hemothorax and pneumothorax. This review is aimed to update basic concepts regarding to mechanical complications associated to the use of central venous catheters and to show radiological images from eight patients who presented to the imagenology department of our hospital with such complications.