RESUMEN
The dynamism envisioned in future high-capacity gridless optical networks requires facing several challenges in distortion mitigation, such as the mitigation of interchannel interference (ICI) effects in any optical channel without information of their adjacent channels. Machine learning (ML)-based techniques have been proposed in recent works to estimate and mitigate different optical impairments with promising results. We propose and evaluate two training strategies for supervised learning algorithms with the aim to minimize ICI effects in a gridless 3×16-Gbaud 16-quadrature amplitude modulation (QAM) Nyquist-wavelength-division multiplexing (WDM) system. One strategy, called updating strategy, is based on symbol training sequence, and the other one, called characterization strategy, is based on an offline training using a previous system characterization. Artificial neural networks (ANN), support vector machine (SVM), K-nearest neighbors (KNN), and extreme learning machine (ELM) algorithms are explored for both training strategies. Experimental results showed a bit error rate (BER) improvement at low training lengths for both training strategies, for instance, gains up to â¼4dB in terms of optical signal-to-noise ratio were achieved in a back-to-back scenario. Besides, the KNN and ELM algorithms showed significant BER reduction in transmission over 250 km optical fiber. Additionally, we carried out a brief computational complexity analysis where ELM presented only 1.9% of ANN processing time. Hence, the use of ML-based techniques could enhance the optical gridless networks performance and consequently fulfill future traffic demands.
RESUMEN
En la actualidad una de cada seis parejas presenta problemas de fertilidad y en el 50% de los casos se debe al factor masculino. A la fecha, el análisis seminal es la única prueba que permite determinar el potencial fértil de un hombre. Entre otros parámetros, la viabilidad espermática es evaluada manualmente presentando una variabilidad debido a la subjetividad producida por la fatiga ocular del experto. El propósito de este trabajo fue desarrollar y validar experimentalmente una herramienta computacional flexible, programable y modular basada en el procesamiento digital de imágenes, para la identificación y clasificación de espermatozoides humanos en una muestra seminal. Las regiones fueron extraídas mediante la técnica de análisis discriminante de Fisher y su clasificación se realizó a través del análisis de agrupamiento y particularmente la técnica de K-medias. Los resultados obtenidos muestran 87,9% de exactitud en la identificación de los espermatozoides vivos y los espermatozoides muertos, 93,4% de efectividad para detectar espermatozoides vivos y 76% de efectividad para detectar los espermatozoides muertos, a partir de un conjunto de 110 imágenes obtenidas de 14 individuos, en comparación con el análisis manual acorde a los procedimientos establecidos por la Organización Mundial de la Salud. La herramienta computacional mostrada aquí contribuye al análisis objetivo de espermatozoides humanos, convirtiéndose en una alternativa a los costosos sistemas comerciales de análisis seminal asistido por computador.
Currently one out of six couples present fertility problems, with 50% of the cases being due to the male. Until now, seminal fluid analysis is the only test that evaluates a male's fertility potential. Among other parameters, sperm viability is manually assessed, which contributes to high data variability as a result of expert subjectivity and eye-fatigue. The purpose of the present study was to develop and experimentally validate a flexible, programmable and modular-based computational tool for digital image processing, identification and classification of human sperm in a semen sample. The regions were extracted using Fisher discriminant analysis and classification methods by cluster analysis and particularly the K-means technique. The results show 87.9% accuracy in identifying living and dead sperm, 93.4% effectiveness in detecting live sperm and 76% effectiveness in detecting dead sperm, from a set of 110 images obtained from 14 individuals, compared with manual analysis according to the procedures established by the World Health Organization. This computational tool contributes to the objective analysis of human sperm, becoming an alternative to expensive commercial systems for computer-assisted semen analysis.