To describe novel embryo features capable of predicting implantation potential as input data for an artificial neural network (ANN) model.
Retrospective cohort study.
University-affiliated private ...IVF center.
This study included 637 patients from the oocyte donation program who underwent single-blastocyst transfer during two consecutive years.
None.
The research was divided into two phases. Phase 1 consisted of the description and analysis of the following embryo features in implanted and nonimplanted embryos: distance and speed of pronuclear migration, blastocyst expanded diameter, inner cell mass area, and trophectoderm cell cycle length. Phase 2 consisted of the development of an ANN algorithm for implantation prediction. Results were obtained for four models fed with different input data. The predictive power was measured with the use of the area under the receiver operating characteristic curve (AUC).
Out of the five novel described parameters, blastocyst expanded diameter and trophectoderm cell cycle length had statistically different values in implanted and nonimplanted embryos. After the ANN models were trained and validated using fivefold cross-validation, they were capable of predicting implantation on testing data with AUCs of 0.64 for ANN1 (conventional morphokinetics), 0.73 for ANN2 (novel morphodynamics), 0.77 for ANN3 (conventional morphokinetics + novel morphodynamics), and 0.68 for ANN4 (discriminatory variables from statistical test).
The novel proposed embryo features affect the implantation potential, and their combination with conventional morphokinetic parameters is effective as input data for a predictive model based on artificial intelligence.
Nuevos y convencionales parámetros embrionarios como datos de entrada para redes neuronales artificiales: un modelo de inteligencia artificial aplicado para la predicción del potencial de implantación.
Describir nuevas características de embriones capaces de predecir el potencial de implantación como datos de entrada para un modelo de red neuronal artificial (ANN).
Estudio de cohorte retrospectivo.
Centro de FIV privado afiliado a la universidad.
Este estudio incluyó a 637 pacientes del programa de donación de ovocitos que se sometieron a transferencia de un solo blastocisto durante dos años consecutivos.
Ninguna.
La investigación se dividió en dos fases. La fase 1 consistió en la descripción y análisis de las siguientes características embrionarias en embriones implantados y no implantados: distancia y velocidad de migración pronuclear, diámetro del blastocisto expandido, área de masa celular interna y duración del ciclo celular del trofoectodermo. La fase 2 consistió en el desarrollo de un algoritmo ANN para la predicción de la implantación. Se obtuvieron resultados para cuatro modelos alimentados con diferentes datos de entrada. El poder predictivo se midió con el uso del área bajo la curva característica operativa del receptor (AUC).
De los cinco nuevos parámetros descritos, el diámetro expandido del blastocisto y la duración del ciclo celular del trofoectodermo tenían valores estadísticamente diferentes en los embriones implantados y no implantados. Después de que los modelos ANN fueron entrenados y validados mediante validación cruzada cinco veces, estos fueron capaces de predecir la implantación en los datos de prueba con AUC de 0,64 para ANN1 (morfocinética convencional), 0,73 para ANN2 (morfodinámica novedosa), 0,77 para ANN3 (morfocinética convencional þ morfodinámica novedosa) y 0,68 para ANN4 (variables discriminatorias de prueba estadística).
Las nuevas características embrionarias propuestas afectan al potencial de implantación y su combinación con parámetros morfocinéticos convencionales es eficaz como datos de entrada para un modelo predictivo basado en inteligencia artificial.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZAGLJ, ZRSKP
This paper presents an investigation on the thermal conductivity of nanofluids using experimental data, neural networks, and correlation for modeling thermal conductivity. The thermal conductivity of ...Mg(OH)2 nanoparticles with mean diameter of 10nm dispersed in ethylene glycol was determined by using a KD2-pro thermal analyzer. Based on the experimental data at different solid volume fractions and temperatures, an experimental correlation is proposed in terms of volume fraction and temperature. Then, the model of relative thermal conductivity as a function of volume fraction and temperature was developed via neural network based on the measured data. A network with two hidden layers and 5 neurons in each layer has the lowest error and highest fitting coefficient. By comparing the performance of the neural network model and the correlation derived from empirical data, it was revealed that the neural network can more accurately predict the Mg(OH)2–EG nanofluids' thermal conductivity.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UL, UM, UPCLJ, UPUK
In this paper, a comprehensive review of the artificial neural network (ANN) based model predictive control (MPC) system design is carried out followed by a case study in which ANN models of a ...residential house located in Ontario, Canada are developed and calibrated with the data measured from site. A new algorithm called best network after multiple iterations (BNMI) is introduced to help in determining the appropriate ANN architecture. The prediction performance of the developed models using BNMI algorithm was significantly better (between 6% and 59% better goodness of fit for various models) when compared to a previous study carried out by the authors which used the default single iteration ANN training algorithm of MATLAB®. The ANN models were further used to design the supervisory MPC for the residential HVAC system. The MPC generated the dynamic temperature set-point profiles of the zone air and buffer tank water which resulted in the operating cost reduction of the equipment without violating the thermal comfort constraints. When compared to the fixed set-point (FSP), MPC was able to save operating cost between 6% and 73% depending on the season.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UL, UM, UPCLJ, UPUK, ZRSKP
This paper aims to review Artificial neural networks, Multi-Layer Perceptron Neural network (MLP) and Convolutional Neural network (CNN) employed to detect breast malignancies for early diagnosis of ...breast cancer based on their accuracy in order to identify which method is better for the diagnosis of breast cell malignancies. Deep comparison of functioning of each network and its designing is performed and then analysis is done based on the accuracy of diagnosis and classification of breast malignancy by the network to decide which network outperforms the other. CNN is found to give slightly higher accuracy than MLP for diagnosis and detection of breast cancer. There still is the need to carefully analyse and perform a thorough research that uses both these methods on the same data set under same conditions in order identify the architecture that gives better accuracy.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZAGLJ, ZRSKP
A mechanics‐informed artificial neural network approach for learning constitutive laws governing complex, nonlinear, elastic materials from strain–stress data is proposed. The approach features a ...robust and accurate method for training a regression‐based model capable of capturing highly nonlinear strain–stress mappings, while preserving some fundamental principles of solid mechanics. In this sense, it is a structure‐preserving approach for constructing a data‐driven model featuring both the form‐agnostic advantage of purely phenomenological data‐driven regressions and the physical soundness of mechanistic models. The proposed methodology enforces desirable mathematical properties on the network architecture to guarantee the satisfaction of physical constraints such as objectivity, consistency (preservation of rigid body modes), dynamic stability, and material stability, which are important for successfully exploiting the resulting model in numerical simulations. Indeed, embedding such notions in a learning approach reduces a model's sensitivity to noise and promotes its robustness to inputs outside the training domain. The merits of the proposed learning approach are highlighted using several finite element analysis examples. Its potential for ensuring the computational tractability of multi‐scale applications is demonstrated with the acceleration of the nonlinear, dynamic, multi‐scale, fluid‐structure simulation of the supersonic inflation dynamics of a parachute system with a canopy made of a woven fabric.
Full text
Available for:
BFBNIB, FZAB, GIS, IJS, KILJ, NLZOH, NUK, OILJ, SAZU, SBCE, SBMB, UL, UM, UPUK
We propose a novel context-dependent (CD) model for large-vocabulary speech recognition (LVSR) that leverages recent advances in using deep belief networks for phone recognition. We describe a ...pre-trained deep neural network hidden Markov model (DNN-HMM) hybrid architecture that trains the DNN to produce a distribution over senones (tied triphone states) as its output. The deep belief network pre-training algorithm is a robust and often helpful way to initialize deep neural networks generatively that can aid in optimization and reduce generalization error. We illustrate the key components of our model, describe the procedure for applying CD-DNN-HMMs to LVSR, and analyze the effects of various modeling choices on performance. Experiments on a challenging business search dataset demonstrate that CD-DNN-HMMs can significantly outperform the conventional context-dependent Gaussian mixture model (GMM)-HMMs, with an absolute sentence accuracy improvement of 5.8% and 9.2% (or relative error reduction of 16.0% and 23.2%) over the CD-GMM-HMMs trained using the minimum phone error rate (MPE) and maximum-likelihood (ML) criteria, respectively.
In order to improve the performance and reliability of the three-phase PWM rectifier in electric vehicle (EV) charging system, it is important to focus on the identification of the DC bus ...capacitance. The existing methods have the disadvantages such as a low identification accuracy, a high hardware dependency, and high costs. Therefore, the development of new identification methodologies could be the way out of the aforementioned disadvantages, in terms of advanced algorithms. In this paper, a DC bus capacitance identification method based on back propagation (BP) artificial neural network (ANN) algorithm is proposed. The capacitance can be accurately identified in the diode rectification stage, by training the neural network to extract the potential laws in the data, which only includes the Phase-a voltage and current, as well as the DC bus voltage ripple. The regression response is 0.99991. Finally, the experimental results show the accuracy and effectiveness of the proposed method. A simple and effective BP ANN is designed, and it can identify the capacitance at different grid voltages and DC bus loads. It has a high identification accuracy that the maximum identification error is less than 4.5%.
Optoelectronic synaptic devices, which combine the functions of photosensitivity and information processing, are essential for the development of artificial visual perception systems. Nevertheless, ...improving the paired pulse facilitation (PPF) index of optoelectronic synaptic devices, which is an urgent problem in the construction of high‐precision artificial visual perception systems, has received less attention so far. Herein, a light‐stimulated synaptic transistor (LSST) device with an ultra‐high PPF index (≈196%) is presented by introducing an ultra‐thin carrier regulator layer hexagonal boron nitride (h‐BN) into a classic graphene‐based hybrid transistor frame (graphene/CsPbBr3 quantum dots). Crucially, analysis of the rate‐limiting effect of h‐BN on photogenerated carriers reveals the mechanism behind the LSST ultra‐high PPF index. Furthermore, a two‐layer artificial neural network connected by LSST devices demonstrate ≈91.5% recognition accuracy of handwritten digits. This work provides an effective method for constructing artificial visual perception systems using a hybrid transistor frame in the future.
Optoelectronic synaptic devices with high paired pulse facilitation (PPF) index are essential for constructing high‐precision artificial visual perception systems. However, improving the PPF index of optoelectronic synaptic devices has received little attention. In this paper, a light‐stimulated synaptic transistor with an ultra‐high PPF index by introducing hexagonal boron nitride into a classic graphene‐based hybrid transistor framework is proposed.
Full text
Available for:
BFBNIB, FZAB, GIS, IJS, KILJ, NLZOH, NUK, OILJ, SAZU, SBCE, SBMB, UL, UM, UPUK
Summary
The uncertainty associated with modeling and performance prediction of solar photovoltaic systems could be easily and efficiently solved by artificial intelligence techniques. During the past ...decade of 2009 to 2019, artificial neural network (ANN), fuzzy logic (FL), genetic algorithm (GA) and their hybrid models are found potential artificial intelligence tools for performance prediction and modeling of solar photovoltaic systems. In addition, during this decade there is no extensive review on applicability of ANN, FL, GA and their hybrid models for performance prediction and modeling of solar photovoltaic systems. Therefore, this article focuses on extensive review on design, modeling, maximum power point tracking, fault detection and output power/efficiency prediction of solar photovoltaic systems using artificial intelligence techniques of the ANN, FL, GA and their hybrid models. In addition, the selected articles on the solar radiation prediction using ANN, FL, GA and their hybrid models are also summarized. Total of 122 articles are reviewed and summarized in the present review for the period of 2009 to 2019 with 90 articles in the field of {ANN, FL, GA and their hybrid models} + solar photovoltaic systems and 32 articles in the field of {ANN, FL, GA and their hybrid models} + solar radiation. The review shows the suitability and reliability of ANN, FL, GA and hybrid models for accurate prediction of the solar radiation and the performance characteristics of solar photovoltaic systems. In addition, this review presents the guidance for the researchers and engineers in the field of solar photovoltaic systems to select the suitable prediction tool for enhancement of the performance characteristics of the solar photovoltaic systems and the utilization of the available solar radiation.
Summary of 155 articles on artificial neural network, fuzzy logic, genetic algorithm and their hybrid models integrated with solar radiation and solar photovoltaic systems.
Applicability and suitability of standalone and hybrid artificial intelligence techniques for prediction of solar radiation and performance of solar photovoltaic systems is presented.
A guidance for deciding the appropriate artificial intelligence techniques for accurate prediction, existing research gaps and their remedies and future scope are presented.
Full text
Available for:
FZAB, GIS, IJS, KILJ, NLZOH, NUK, OILJ, SAZU, SBCE, SBMB, UL, UM, UPUK
Artificial intelligence has been applied in wildfire science and management since the 1990s, with early applications including neural networks and expert systems. Since then, the field has rapidly ...progressed congruently with the wide adoption of machine learning (ML) methods in the environmental sciences. Here, we present a scoping review of ML applications in wildfire science and management. Our overall objective is to improve awareness of ML methods among wildfire researchers and managers, as well as illustrate the diverse and challenging range of problems in wildfire science available to ML data scientists. To that end, we first present an overview of popular ML approaches used in wildfire science to date and then review the use of ML in wildfire science as broadly categorized into six problem domains, including (i) fuels characterization, fire detection, and mapping; (ii) fire weather and climate change; (iii) fire occurrence, susceptibility, and risk; (iv) fire behavior prediction; (v) fire effects; and (vi) fire management. Furthermore, we discuss the advantages and limitations of various ML approaches relating to data size, computational requirements, generalizability, and interpretability, as well as identify opportunities for future advances in the science and management of wildfires within a data science context. In total, to the end of 2019, we identified 300 relevant publications in which the most frequently used ML methods across problem domains included random forests, MaxEnt, artificial neural networks, decision trees, support vector machines, and genetic algorithms. As such, there exists opportunities to apply more current ML methods—including deep learning and agent-based learning—in the wildfire sciences, especially in instances involving very large multivariate datasets. We must recognize, however, that despite the ability of ML models to learn on their own, expertise in wildfire science is necessary to ensure realistic modelling of fire processes across multiple scales, while the complexity of some ML methods such as deep learning requires a dedicated and sophisticated knowledge of their application. Finally, we stress that the wildfire research and management communities play an active role in providing relevant, high-quality, and freely available wildfire data for use by practitioners of ML methods.
L’intelligence artificielle a ete utilisee en science et en gestion des feux de foret depuis les annees 1990, les premieres applications comprenant les reseaux neuronaux et les systemes experts. Depuis lors, le domaine a rapidement progresse parallelement a l’adoption des methodes d’apprentissage machine (AM) en sciences de l’environnement. Les auteurs presentent ici une synthese du cadrage des applications de l’AM en science et en gestion des feux de foret. Leur objectif global consiste a ameliorer la notoriete des methodes d’AM aupres des chercheurs et des gestionnaires des feux de foret, de meme qu’a illustrer l’etendue vaste et complexe des problemes en science des feux de foret dont disposent les scientifiques specialistes de donnees en AM. A cette fin, ils presentent d’abord un survol des approches populaires en AM utilisees en science des feux de foret a ce jour et font ensuite la synthese de l’utilisation de l’AM en science des feux de foret, selon six grands domaines de problemes dont (i) la caracterisation des carburants, la detection et la cartographie de l’incendie; (ii) la temperature de l’incendie et les changements climatiques; (iii) les circonstances, la susceptibilite et le risque d’incendie; (iv) la prediction du comportement de l’incendie; (v) les effets de l’incendie; et (vi) la gestion de l’incendie. Par ailleurs, les auteurs discutent des avantages et des limites de differentes approches d’AM en lien avec la taille des donnees, les exigences de calcul, le potentiel de generalisation et d’interpretation et identifient egalement les possibilites d’avancees futures en science et gestion des feux de foret dans le contexte de la science des donnees. Ils ont identifie au total 300 publications pertinentes jusqu’a la fin de 2019 qui comprennent les methodes d’AM les plus frequemment utilisees a travers les domaines de problemes, dont les forets aleatoires, MaxEnt, les reseaux de neurones artificiels, les arbres de decision, les separateurs a vaste marge et les algorithmes genetiques. Il existe ainsi des possibilites d’appliquer davantage de methodes actuelles d’AM—y compris l’apprentissage profond et l’apprentissage base sur l’agent—en sciences des feux de foret, particulierement dans les cas impliquant de tres grands ensembles de donnees multivariees. Ilsreconnaissent cependant que, malgre la capacite des methodes en AM d’apprendre par elles-memes, l’expertise en science des feux de foret est necessaire pour s’assurer d’une modelisation realiste des processus des incendies a differentes echelles, alors que la complexite de certaines methodes en AM telles que l’apprentissage profond, requiert une connaissance approfondie et specifique de leur application. Finalement, ils soulignent que les communautes qui se consacrent a la recherche et a la gestion des feux de foret jouent un role actif en fournissant des donnees pertinentes, de haute qualite et en libre acces a l’usage des praticiens des methodes en AM.
Full text
Available for:
BFBNIB, DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, UILJ, UKNU, UL, UM, UPUK