RESUMO
Convolutional Neural Networks (CNN) are a class of machine learning models predominately used in computer vision tasks and can achieve human-like performance through learning from experience. Their striking similarities to the structural and functional principles of the primate visual system allow for comparisons between these artificial networks and their biological counterparts, enabling exploration of how visual functions and neural representations may emerge in the real brain from a limited set of computational principles. After considering the basic features of CNNs, we discuss the opportunities and challenges of endorsing CNNs as in silico models of the primate visual system. Specifically, we highlight several emerging notions about the anatomical and physiological properties of the visual system that still need to be systematically integrated into current CNN models. These tenets include the implementation of parallel processing pathways from the early stages of retinal input and the reconsideration of several assumptions concerning the serial progression of information flow. We suggest design choices and architectural constraints that could facilitate a closer alignment with biology provide causal evidence of the predictive link between the artificial and biological visual systems. Adopting this principled perspective could potentially lead to new research questions and applications of CNNs beyond modeling object recognition.
RESUMO
Resumen Introducción: La mortalidad fetal intrauterina (MFIU) es un problema de salud pública por sus elevadas tasas a nivel mundial y en poblaciones de ingresos medios y bajos. Sin embargo, es un evento poco estudiado y carece de visibilidad en las políticas, planes y programas de salud pública. Objetivo: Analizar los determinantes sociodemográficos y clínicos asociados a la MFIU en Pasto-Colombia. Materiales y métodos: Estudio analítico observacional con 88 muertes fetales como casos y 88 nacidos vivos como controles, ocurridas en hospitales de tercer nivel en el municipio de Pasto-Colombia durante 2010-2012, para determinar la relación entre mortalidad fetal, condiciones clínicas (complicaciones del embarazo, edad gestacional, peso al nacer, controles prenatales, antecedentes obstétricos, tóxicos o aborto) y sociodemográficas (edad, etnia, ocupación, estado civil, estrato, zona de residencia, escolaridad, paridad, condición de desplazamiento, embarazo planeado). Resultados: Se identificó que el riesgo de mortalidad fetal es significativamente menor con el incremento de la edad gestacional (OR ajustado=0,76 IC95% 0,62; 0,93) y el peso al nacer (OR ajustado=0,99 IC95% 0,98; 0,99). Otras variables clínicas y sociodemográficas no se asociaron. Conclusión: Los resultados proveen evidencia para la planificación de planes de intervención que prioricen a mujeres cuyo feto tenga un peso inferior al normal y un riesgo de nacimiento prematuro.
Abstract Introduction: Intrauterine fetal mortality (IUFM) is a public health problem because of its high rates worldwide and in low-and middle-income populations. However, it is a little-studied event and lacks visibility in public health policies, plans and programs. Objective: To analyze the sociodemographic and clinical determinants associated with IUFM in Pasto-Colombia. Materials and methods: A study, that includes 88 fetal deaths as cases and 88 live births as controls occurred in third level hospitals in Pasto-Colombia during 2010 and 2012, was carried out to determine the relationship between fetal mortality, clinical conditions (complications of pregnancy, gestational age, birth weight, prenatal controls, pathological and toxic medical history, or abortion) and sociodemographic conditions (age, ethnicity, occupation, marital status, stratum, area of residence , schooling, parity, displacement condition, planned pregnancy). Results: It was identified that the risk of fetal mortality is significantly lower with the increase in gestational age (OR ajustado = 0.76 IC95% 0.62; 0.93) and birth weight (OR ajustado = 0.99 IC95% 0.98; 0.99). Other clinical and sociodemographic variables were not associated. Conclusion: The results provide evidence for planning intervention plans that prioritize women whose fetus has a lower-than-normal weight and a risk of premature birth.
