Examinando por Autor "Querales, Marvin"

Mostrando 1 - 3 de 3
  • Miniatura
    Ítem
    A self-identification Neuro-Fuzzy inference framework for modeling rainfall-runoff in a Chilean watershed
    (Elsevier, 2021) Morales, Yerel; Querales, Marvin; Rosas, Harvey; Allende-Cid, Hector; Salas, Rodrigo
    Modeling the relationship between rainfall and runoff is an important issue in hydrology, but it is a complicated task because both the high levels of complexity in which both processes are embedded and the associated uncertainty, affect the forecasting. Neuro-fuzzy models have emerged as a useful approach, given the ability of neural networks to optimize parameters in a fuzzy system. In this work a Self-Identification Neuro-Fuzzy Inference Model (SINFIM) for modeling the relationship between rainfall and runoff on a Chilean watershed is proposed to reduce the uncertainty of selecting both the rainfall and runoff lags and the number of membership functions required in a fuzzy system. The data comes from the Diguillín river located in Ñuble region and average daily runoff and average daily rainfall recorded from years 2000 to 2018, according to the Chilean directorate of water resources (DGA). In addition, we worked with the Colorado River basin, located in the Maule region, to validate the method developed. The experimental results showed a good adjustment using the last 3 years as validation set, further improvement was achieved using only the last year was used as validation test, obtaining 84% of and Kling Gupta Efficiency, higher than other forecasting models such as Adaptive Neuro-Fuzzy Inference System (ANFIS), Artificial neural networks (ANN), and Long Short-Term Memory (LSTM) approach. In addition, Nash-Sutcliffe efficiency and percent BIAS indicate the method is a promising model. On the other hand, even better results were obtained in the validation basin, whose adjustment was 94% and an efficiency of 97%. Therefore, the proposed model is a solid alternative to forecast the runoff in a given watershed, obtaining good performance measurements, managing to predict both the low and peak runoff values from rainfall events, avoiding the requirement to determine a priori the lags of time series and the number of fuzzy rules.
  • Miniatura
    Ítem
    An Efficient Multi-Level Convolutional Neural Network Approach for White Blood Cells Classification
    (MDPI, 2022) Cheuque, César; Querales, Marvin; León|, Roberto|Salas, Rodrigo; Torres, Torres
    The evaluation of white blood cells is essential to assess the quality of the human immune system; however, the assessment of the blood smear depends on the pathologist’s expertise. Most machine learning tools make a one-level classification for white blood cell classification. This work presents a two-stage hybrid multi-level scheme that efficiently classifies four cell groups: lymphocytes and monocytes (mononuclear) and segmented neutrophils and eosinophils (polymorphonuclear). At the first level, a Faster R-CNN network is applied for the identification of the region of interest of white blood cells, together with the separation of mononuclear cells from polymorphonuclear cells. Once separated, two parallel convolutional neural networks with the MobileNet structure are used to recognize the subclasses in the second level. The results obtained using Monte Carlo cross-validation show that the proposed model has a performance metric of around 98.4% (accuracy, recall, precision, and F1-score). The proposed model represents a good alternative for computer-aided diagnosis (CAD) tools for supporting the pathologist in the clinical laboratory in assessing white blood cells from blood smear images.
  • Miniatura
    Ítem
    Image Quality Assessment to Emulate Experts’ Perception in Lumbar MRI Using Machine Learning
    (MDPI, 2021) Chabert, Steren; Castro, Juan Sebastian; Muñoz, Leonardo; Cox, Pablo; Riveros, Rodrigo; Vielma, Juan; Huerta, Gamaliel; Querales, Marvin; Saavedra, Carolina; Veloz, Alejandro; Salas, Rodrigo
    Medical image quality is crucial to obtaining reliable diagnostics. Most quality controls rely on routine tests using phantoms, which do not reflect closely the reality of images obtained on patients and do not reflect directly the quality perceived by radiologists. The purpose of this work is to develop a method that classifies the image quality perceived by radiologists in MR images. The focus was set on lumbar images as they are widely used with different challenges. Three neuroradiologists evaluated the image quality of a dataset that included T1-weighting images in axial and sagittal orientation, and sagittal T2-weighting. In parallel, we introduced the computational assessment using a wide range of features extracted from the images, then fed them into a classifier system. A total of 95 exams were used, from our local hospital and a public database, and part of the images was manipulated to broaden the distribution quality of the dataset. Good recall of 82% and an area under curve (AUC) of 77% were obtained on average in testing condition, using a Support Vector Machine. Even though the actual implementation still relies on user interaction to extract features, the results are promising with respect to a potential implementation for monitoring image quality online with the acquisition process.