Neural Computation Methods in the Determination of Fungicides

Araujo, Carmen Paz Suárez; Báez, Patricio García; Trujillo, Yaride Hernández

doi:10.5772/13229

Cited by 3 publications

(11 citation statements)

References 55 publications

(65 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Rumelhart [ 94 ] proposed eight elements to define a general framework of an ANN: a set of processing units with local memory; a state of activation of each of them; an output function for each of these units; a pattern of connectivity among the units; a propagation rule or network function that defines how the activity patterns propagate and is based on the connectivity pattern; an activation rule for combining the state of a neuron with the inputs that arrive to it and produce a new activation of that unit; a learning rule that indicates how the connectivity patterns change based on experience; and representation of the environment in which the ANN works (a local and a global information setting exist) [ 95 ].…”

Section: Methodsmentioning

confidence: 99%

“…Connectivity topology indicates the shape in which the artificial neurons are interconnected, frequently grouped in layers [ 95 ]. Generally, the basic structure of an ANN consists of three layers: one that receives the input data (hence called “input layer”), a “hidden layer” where the processing of the input data is done, and the “output layer” that provides the output of the ANN.…”

Section: Methodsmentioning

confidence: 99%

“…Not only the direction of the connections (that is, the information flow direction) but also the connectivity density may be defined: fully or partially connected [ 95 ].…”

Section: Methodsmentioning

confidence: 99%

“…The network neurodynamic describes how the processing units process the local information [ 95 ]. It can be expressed by time-continuous or discrete functions, the latter case requiring to know when to calculate the outputs of the processing units [ 45 ].…”

Section: Methodsmentioning

confidence: 99%

“…This way, the parameters of the network (generally, the weights of the connections) change iteratively (often after an input sample is used for learning) and, finally, get optimized so that the error made by the network is minimized, principally trying to reduce the difference between the expected output and the one obtained by the network. Depending on the level of tutoring for each input data, three different paradigms exist [ 95 ]: supervised (desired output is also provided to the ANN), reinforcement (only the correctness is indicated), and unsupervised (the ANN self-organizes based on the structures and redundancy found in the input data).…”

Section: Methodsmentioning

confidence: 99%

See 4 more Smart Citations

Neural Computation-Based Methods for the Early Diagnosis and Prognosis of Alzheimer’s Disease Not Using Neuroimaging Biomarkers: A Systematic Review

Cabrera-León,

Báez,

Fernández-López

et al. 2024

JAD

View full text Add to dashboard Cite

Background: The growing number of older adults in recent decades has led to more prevalent geriatric diseases, such as strokes and dementia. Therefore, Alzheimer’s disease (AD), as the most common type of dementia, has become more frequent too. Background: Objective: The goals of this work are to present state-of-the-art studies focused on the automatic diagnosis and prognosis of AD and its early stages, mainly mild cognitive impairment, and predicting how the research on this topic may change in the future. Methods: Articles found in the existing literature needed to fulfill several selection criteria. Among others, their classification methods were based on artificial neural networks (ANNs), including deep learning, and data not from brain signals or neuroimaging techniques were used. Considering our selection criteria, 42 articles published in the last decade were finally selected. Results: The most medically significant results are shown. Similar quantities of articles based on shallow and deep ANNs were found. Recurrent neural networks and transformers were common with speech or in longitudinal studies. Convolutional neural networks (CNNs) were popular with gait or combined with others in modular approaches. Above one third of the cross-sectional studies utilized multimodal data. Non-public datasets were frequently used in cross-sectional studies, whereas the opposite in longitudinal ones. The most popular databases were indicated, which will be helpful for future researchers in this field. Conclusions: The introduction of CNNs in the last decade and their superb results with neuroimaging data did not negatively affect the usage of other modalities. In fact, new ones emerged.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

“…Not only the direction of the connections (that is, the information flow direction) but also the connectivity density may be defined: fully or partially connected [ 95 ].…”

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

See 3 more Smart Citations