China
Africa
Explore chapters and articles related to this topic
Advances in Artificial Intelligence Applied to Heart Failure
Published in Kayvan Najarian, Delaram Kahrobaei, Enrique Domínguez, Reza Soroushmehr, Artificial Intelligence in Healthcare and Medicine, 2022
Jose M. García-Pinilla, Francisco Lopez Valverde
A Bayesian network is a probabilistic model that represents a series of random variables and their conditional independencies through a directed acyclic graph (Friedman et al., 1997). A Bayesian network can represent, for example, the probabilistic relationships between diseases and symptoms. Given certain symptoms, the network can be used to calculate the probabilities that certain diseases are present in an organism. There are efficient algorithms that infer and learn using this type of representation.
Machine Learning Algorithms Used in Medical Field with a Case Study
Published in K. Gayathri Devi, Kishore Balasubramanian, Le Anh Ngoc, Machine Learning and Deep Learning Techniques for Medical Science, 2022
Bayes’ Theorem is applied in Bayesian methods for solving the classification and regression type problems. Mainly used Bayesian algorithms are:Gaussian Naive BayesNaive BayesAveraged One-Dependence Estimators (AODE)Bayesian Belief Network (BBN)Bayesian Network (BN)Multinomial Naive Bayes
Disease Prediction and Drug Development
Published in Arvind Kumar Bansal, Javed Iqbal Khan, S. Kaisar Alam, Introduction to Computational Health Informatics, 2019
Arvind Kumar Bansal, Javed Iqbal Khan, S. Kaisar Alam
An alternative approach to build pathways is based on maximizing the joint-probability distribution of Bayesian networks that uses mutual information and entropy. As described in Section 3.7.1, Bayesian network is a directed acyclic graph involving N variables where nodes are the variables, and edges show the probability of occurrences of an event given the set of parent conditions. The joint probability distribution in a Bayesian network is given by Equation 10.8.
Using neural networks to predict high-risk flight environments from accident and incident data
Published in International Journal of Occupational Safety and Ergonomics, 2022
Aviation risk-analysis research is dominated by methods dependent on SME knowledge, often incorporated into Bayesian models. However, expert knowledge of highly complex systems is limited since threat–consequence relationships can be ambiguous [9–11] and unknown or unexpected relationships cannot be accounted for [11]. Furthermore, while Bayesian methods may result in high-performing models if the prior probability is based on population statistics, in practice it is typically calculated from finite samples and/or SME knowledge. Therefore, the applicability of Bayesian networks is limited [12]. Fuzzy logic has also been applied to the analysis of flight environment factors to assess risk (e.g., [13,14]). However, in concluding his discussion of fuzzy logic methods, Hadjimichael [13, p.6516] stated that ‘a more robust method of determining the “most causal” risk factors is necessary. This is a complex issue, as finding a meaningful and useful definition of “most causal” is a significant research challenge’.
Apathetic symptoms and white matter integrity after traumatic brain injury
Published in Brain Injury, 2021
B Navarro-Main, AM Castaño-León, A Hilario, A Lagares, G Rubio, JA Periañez, M Rios-Lago
In order to detect multivariate relationships among initial clinical measures, standard outcome scores and apathy groups of items a particular type of Probabilistic Graphical Models (PGM) named Bayesian Networks (BN) (30,31) a machine learning approach was applied. In general, BN are a visual metaphor of the relationships among variables using directed acyclic graphs (DAGs). BN encodes probabilistic dependency relationships among a set of variables by a DAG, where nodes correspond to variables and arcs represent dependence. We used a score‐based hill‐climbing algorithm to infer network structure and applied bootstrap resampling to learn a set of 500 network structures, then an average network was created where the most significant edges were kept. This procedure is more robust and gives a better model than choosing a single high-scoring network. Arc strength represents how frequent an edge appears in those resampling networks; it is included in the plot by means of edges width. Bayesian network learning algorithms were developed using the R package bnlearn (32).
The shape of low-concentration dose–response functions for benzene: implications for human health risk assessment
Published in Critical Reviews in Toxicology, 2021
Louis A. Cox, Hans B. Ketelslegers, R. Jeffrey Lewis
Biomarkers are potentially useful for estimating unobserved benzene exposures (or improving estimates of concentrations measured with error) and for predicting health risks caused by exposures. The predictive value of a biomarker depends on how well it can be measured and on by how much conditioning on its measured value reduces uncertainty (e.g. the average entropy of the conditional probability distribution) of the dependent variable being predicted. For example, Figure 4 shows the structure of a Bayesian network (BN) model for quantifying conditional probabilities of some variables (e.g. various metabolites, markers and AML (the Leukemia node at the lower right)), given observed or assumed values of others. The conditional probability distribution of each variable (node in the network) depends on the values of the variables that point into it. This BN, from Hack et al. (2010), was constructed manually based on a detailed literature review of candidate markers and outcomes and verifiable, usefully accurate predictive relationships among them, given limitations in measurement techniques.