Artificial neural networks in medical diagnosis

Detecting central fixation by means of artificial neural networks in a pediatric vision screener using retinal birefringence scanning

Background

Reliable detection of central fixation and eye alignment is essential in the diagnosis of amblyopia (“lazy eye”), which can lead to blindness. Our lab has developed and reported earlier a pediatric vision screener that performs scanning of the retina around the fovea and analyzes changes in the polarization state of light as the scan progresses. Depending on the direction of gaze and the instrument design, the screener produces several signal frequencies that can be utilized in the detection of central fixation. The objective of this study was to compare artificial neural networks with classical statistical methods, with respect to their ability to detect central fixation reliably.

Methods

A classical feedforward, pattern recognition, two-layer neural network architecture was used, consisting of one hidden layer and one output layer. The network has four inputs, representing normalized spectral powers at four signal frequencies generated during retinal birefringence scanning. The hidden layer contains four neurons. The output suggests presence or absence of central fixation. Backpropagation was used to train the network, using the gradient descent algorithm and the cross-entropy error as the performance function. The network was trained, validated and tested on a set of controlled calibration data obtained from 600 measurements from ten eyes in a previous study, and was additionally tested on a clinical set of 78 eyes, independently diagnosed by an ophthalmologist.

Results

In the first part of this study, a neural network was designed around the calibration set. With a proper architecture and training, the network provided performance that was comparable to classical statistical methods, allowing perfect separation between the central and paracentral fixation data, with both the sensitivity and the specificity of the instrument being 100%. In the second part of the study, the neural network was applied to the clinical data. It allowed reliable separation between normal subjects and affected subjects, its accuracy again matching that of the statistical methods.

Conclusion

With a proper choice of a neural network architecture and a good, uncontaminated training data set, the artificial neural network can be an efficient classification tool for detecting central fixation based on retinal birefringence scanning.

Machine-learning algorithms define pathogen-specific local immune fingerprints in peritoneal dialysis patients with bacterial infections

The immune system has evolved to sense invading pathogens, control infection, and restore tissue integrity. Despite symptomatic variability in patients, unequivocal evidence that an individual's immune system distinguishes between different organisms and mounts an appropriate response is lacking. We here used a systematic approach to characterize responses to microbiologically well-defined infection in a total of 83 peritoneal dialysis patients on the day of presentation with acute peritonitis. A broad range of cellular and soluble parameters was determined in peritoneal effluents, covering the majority of local immune cells, inflammatory and regulatory cytokines and chemokines as well as tissue damage–related factors. Our analyses, utilizing machine-learning algorithms, demonstrate that different groups of bacteria induce qualitatively distinct local immune fingerprints, with specific biomarker signatures associated with Gram-negative and Gram-positive organisms, and with culture-negative episodes of unclear etiology. Even more, within the Gram-positive group, unique immune biomarker combinations identified streptococcal and non-streptococcal species including coagulase-negative Staphylococcus spp. These findings have diagnostic and prognostic implications by informing patient management and treatment choice at the point of care. Thus, our data establish the power of non-linear mathematical models to analyze complex biomedical datasets and highlight key pathways involved in pathogen-specific immune responses.

Handling limited datasets with neural networks in medical applications: A small-data approach

MOTIVATION

Single-centre studies in medical domain are often characterised by limited samples due to the complexity and high costs of patient data collection. Machine learning methods for regression modelling of small datasets (less than 10 observations per predictor variable) remain scarce. Our work bridges this gap by developing a novel framework for application of artificial neural networks (NNs) for regression tasks involving small medical datasets.

METHODS

In order to address the sporadic fluctuations and validation issues that appear in regression NNs trained on small datasets, the method of multiple runs and surrogate data analysis were proposed in this work. The approach was compared to the state-of-the-art ensemble NNs; the effect of dataset size on NN performance was also investigated.

RESULTS

The proposed framework was applied for the prediction of compressive strength (CS) of femoral trabecular bone in patients suffering from severe osteoarthritis. The NN model was able to estimate the CS of osteoarthritic trabecular bone from its structural and biological properties with a standard error of 0.85MPa. When evaluated on independent test samples, the NN achieved accuracy of 98.3%, outperforming an ensemble NN model by 11%. We reproduce this result on CS data of another porous solid (concrete) and demonstrate that the proposed framework allows for an NN modelled with as few as 56 samples to generalise on 300 independent test samples with 86.5% accuracy, which is comparable to the performance of an NN developed with 18 times larger dataset (1030 samples).

CONCLUSION

The significance of this work is two-fold: the practical application allows for non-destructive prediction of bone fracture risk, while the novel methodology extends beyond the task considered in this study and provides a general framework for application of regression NNs to medical problems characterised by limited dataset sizes.

A novel method for the discrimination of Hawthorn and its processed products using an intelligent sensory system and artificial neural networks

Hawthorn (CFS) has commonly been applied as an important traditional Chinese medicine and food for thousands of years. The raw material of CFS is commonly processed by stir-frying to obtain yellow (CFY), dark brown (CFD), and carbon dark (CFC) colored products, which are used for different clinical uses. In this study, an intelligent sensory system (ISS) was used to obtain the color, gas, and flavor samples data, which were further employed to develop a novel and accurate method for the identification of CFS and its processed products using principal component analysis. Moreover, this research developed a model of an artificial neural network, which could be used to predict the total organic acid, total flavonoids, citric acid, hyperin, and 5-hydroxymethyl furfural via determination of the color, odor, and taste of a sample. In conclusion, the ISS and the artificial neural network are useful tools for rapid, accurate, and effective discrimination of CFS and its processed products.

Optimizing the Learning Process of Feedforward Neural Networks Using Lightning Search Algorithm

Evolutionary Neural Networks are proven to be beneficial in solving challenging datasets mainly due to the high local optima avoidance. Stochastic operators in such techniques reduce the probability of stagnation in local solutions and assist them to supersede conventional training algorithms such as Back Propagation (BP) and Levenberg-Marquardt (LM). According to the No-Free-Lunch (NFL), however, there is no optimization technique for solving all optimization problems. This means that a Neural Network trained by a new algorithm has the potential to solve a new set of problems or outperform the current techniques in solving existing problems. This motivates our attempts to investigate the efficiency of the recently proposed Evolutionary Algorithm called Lightning Search Algorithm (LSA) in training Neural Network for the first time in the literature. The LSA-based trainer is benchmarked on 16 popular medical diagnosis problems and compared to BP, LM, and 6 other evolutionary trainers. The quantitative and qualitative results show that the LSA algorithm is able to show not only better local solutions avoidance but also faster convergence speed compared to the other algorithms employed. In addition, the statistical test conducted proves that the LSA-based trainer is significantly superior in comparison with the current algorithms on the majority of datasets.

Boosting backpropagation algorithm by stimulus-sampling: Application in computer-aided medical diagnosis

Neural networks (NNs), in general, and multi-layer perceptron (MLP), in particular, represent one of the most efficient classifiers among the machine learning (ML) algorithms. Inspired by the stimulus-sampling paradigm, it is plausible to assume that the association of stimuli with the neurons in the output layer of a MLP can increase its performance. The stimulus-sampling process is assumed memoryless (Markovian), in the sense that the choice of a particular stimulus at a certain step, conditioned by the whole prior evolution of the learning process, depends only on the network's answer at the previous step. This paper proposes a novel learning technique, by enhancing the standard backpropagation algorithm performance with the aid of a stimulus-sampling procedure applied to the output neurons. The network uses the observable behavior that varies throughout the training process by stimulating the correct answers through corresponding rewards/penalties assigned to the output neurons. The proposed model has been applied in computer-aided medical diagnosis using five real-life breast cancer, colon cancer, diabetes, thyroid, and fetal heartbeat databases. The statistical comparison to well-established ML algorithms proved beyond doubt its efficiency and robustness.

Use of predictive algorithms in-home monitoring of chronic obstructive pulmonary disease and asthma: A systematic review

Major reported factors associated with the limited effectiveness of home telemonitoring interventions in chronic respiratory conditions include the lack of useful early predictors, poor patient compliance and the poor performance of conventional algorithms for detecting deteriorations. This article provides a systematic review of existing algorithms and the factors associated with their performance in detecting exacerbations and supporting clinical decisions in patients with chronic obstructive pulmonary disease (COPD) or asthma. An electronic literature search in Medline, Scopus, Web of Science and Cochrane library was conducted to identify relevant articles published between 2005 and July 2015. A total of 20 studies (16 COPD, 4 asthma) that included research about the use of algorithms in telemonitoring interventions in asthma and COPD were selected. Differences on the applied definition of exacerbation, telemonitoring duration, acquired physiological signals and symptoms, type of technology deployed and algorithms used were found. Predictive models with good clinically reliability have yet to be defined, and are an important goal for the future development of telehealth in chronic respiratory conditions. New predictive models incorporating both symptoms and physiological signals are being tested in telemonitoring interventions with positive outcomes. However, the underpinning algorithms behind these models need be validated in larger samples of patients, for longer periods of time and with well-established protocols. In addition, further research is needed to identify novel predictors that enable the early detection of deteriorations, especially in COPD. Only then will telemonitoring achieve the aim of preventing hospital admissions, contributing to the reduction of health resource utilization and improving the quality of life of patients.

Feed-forward neural network model for hunger and satiety related VAS score prediction

BACKGROUND

An artificial neural network approach was chosen to model the outcome of the complex signaling pathways in the gastro-intestinal tract and other peripheral organs that eventually produce the satiety feeling in the brain upon feeding.

METHODS

A multilayer feed-forward neural network was trained with sets of experimental data relating concentration-time courses of plasma satiety hormones to Visual Analog Scales (VAS) scores. The network successfully predicted VAS responses from sets of satiety hormone data obtained in experiments using different food compositions.

RESULTS

The correlation coefficients for the predicted VAS responses for test sets having i) a full set of three satiety hormones, ii) a set of only two satiety hormones, and iii) a set of only one satiety hormone were 0.96, 0.96, and 0.89, respectively. The predicted VAS responses discriminated the satiety effects of high satiating food types from less satiating food types both in orally fed and ileal infused forms.

CONCLUSIONS

From this application of artificial neural networks, one may conclude that neural network models are very suitable to describe situations where behavior is complex and incompletely understood. However, training data sets that fit the experimental conditions need to be available.

A General Fuzzy Cerebellar Model Neural Network Multidimensional Classifier Using Intuitionistic Fuzzy Sets for Medical Identification

The diversity of medical factors makes the analysis and judgment of uncertainty one of the challenges of medical diagnosis. A well-designed classification and judgment system for medical uncertainty can increase the rate of correct medical diagnosis. In this paper, a new multidimensional classifier is proposed by using an intelligent algorithm, which is the general fuzzy cerebellar model neural network (GFCMNN). To obtain more information about uncertainty, an intuitionistic fuzzy linguistic term is employed to describe medical features. The solution of classification is obtained by a similarity measurement. The advantages of the novel classifier proposed here are drawn out by comparing the same medical example under the methods of intuitionistic fuzzy sets (IFSs) and intuitionistic fuzzy cross-entropy (IFCE) with different score functions. Cross verification experiments are also taken to further test the classification ability of the GFCMNN multidimensional classifier. All of these experimental results show the effectiveness of the proposed GFCMNN multidimensional classifier and point out that it can assist in supporting for correct medical diagnoses associated with multiple categories.

Centered Kernel Alignment Enhancing Neural Network Pretraining for MRI-Based Dementia Diagnosis

Dementia is a growing problem that affects elderly people worldwide. More accurate evaluation of dementia diagnosis can help during the medical examination. Several methods for computer-aided dementia diagnosis have been proposed using resonance imaging scans to discriminate between patients with Alzheimer's disease (AD) or mild cognitive impairment (MCI) and healthy controls (NC). Nonetheless, the computer-aided diagnosis is especially challenging because of the heterogeneous and intermediate nature of MCI. We address the automated dementia diagnosis by introducing a novel supervised pretraining approach that takes advantage of the artificial neural network (ANN) for complex classification tasks. The proposal initializes an ANN based on linear projections to achieve more discriminating spaces. Such projections are estimated by maximizing the centered kernel alignment criterion that assesses the affinity between the resonance imaging data kernel matrix and the label target matrix. As a result, the performed linear embedding allows accounting for features that contribute the most to the MCI class discrimination. We compare the supervised pretraining approach to two unsupervised initialization methods (autoencoders and Principal Component Analysis) and against the best four performing classification methods of the 2014 CADDementia challenge. As a result, our proposal outperforms all the baselines (7% of classification accuracy and area under the receiver-operating-characteristic curve) at the time it reduces the class biasing.

Aid decision algorithms to estimate the risk in congenital heart surgery

BACKGROUND AND OBJECTIVE

In this paper, we have tested the suitability of using different artificial intelligence-based algorithms for decision support when classifying the risk of congenital heart surgery. In this sense, classification of those surgical risks provides enormous benefits as the a priori estimation of surgical outcomes depending on either the type of disease or the type of repair, and other elements that influence the final result. This preventive estimation may help to avoid future complications, or even death.

METHODS

We have evaluated four machine learning algorithms to achieve our objective: multilayer perceptron, self-organizing map, radial basis function networks and decision trees. The architectures implemented have the aim of classifying among three types of surgical risk: low complexity, medium complexity and high complexity.

RESULTS

Accuracy outcomes achieved range between 80% and 99%, being the multilayer perceptron method the one that offered a higher hit ratio.

CONCLUSIONS

According to the results, it is feasible to develop a clinical decision support system using the evaluated algorithms. Such system would help cardiology specialists, paediatricians and surgeons to forecast the level of risk related to a congenital heart disease surgery.

Multivariate Calibration Approach for Quantitative Determination of Cell-Line Cross Contamination by Intact Cell Mass Spectrometry and Artificial Neural Networks

Cross-contamination of eukaryotic cell lines used in biomedical research represents a highly relevant problem. Analysis of repetitive DNA sequences, such as Short Tandem Repeats (STR), or Simple Sequence Repeats (SSR), is a widely accepted, simple, and commercially available technique to authenticate cell lines. However, it provides only qualitative information that depends on the extent of reference databases for interpretation. In this work, we developed and validated a rapid and routinely applicable method for evaluation of cell culture cross-contamination levels based on mass spectrometric fingerprints of intact mammalian cells coupled with artificial neural networks (ANNs). We used human embryonic stem cells (hESCs) contaminated by either mouse embryonic stem cells (mESCs) or mouse embryonic fibroblasts (MEFs) as a model. We determined the contamination level using a mass spectra database of known calibration mixtures that served as training input for an ANN. The ANN was then capable of correct quantification of the level of contamination of hESCs by mESCs or MEFs. We demonstrate that MS analysis, when linked to proper mathematical instruments, is a tangible tool for unraveling and quantifying heterogeneity in cell cultures. The analysis is applicable in routine scenarios for cell authentication and/or cell phenotyping in general.

Swarm intelligence applied to the risk evaluation for congenital heart surgery

Particle Swarm Optimization is an optimization technique based on the positions of several particles created to find the best solution to a problem. In this work we analyze the accuracy of a modification of this algorithm to classify the levels of risk for a surgery, used as a treatment to correct children malformations that imply congenital heart diseases.

Neural Network Prediction of ICU Length of Stay Following Cardiac Surgery Based on Pre-Incision Variables

Background

Advanced predictive analytical techniques are being increasingly applied to clinical risk assessment. This study compared a neural network model to several other models in predicting the length of stay (LOS) in the cardiac surgical intensive care unit (ICU) based on pre-incision patient characteristics.

Methods

Thirty six variables collected from 185 cardiac surgical patients were analyzed for contribution to ICU LOS. The Automatic Linear Modeling (ALM) module of IBM-SPSS software identified 8 factors with statistically significant associations with ICU LOS; these factors were also analyzed with the Artificial Neural Network (ANN) module of the same software. The weighted contributions of each factor (“trained” data) were then applied to data for a “new” patient to predict ICU LOS for that individual.

Results

Factors identified in the ALM model were: use of an intra-aortic balloon pump; O₂ delivery index; age; use of positive cardiac inotropic agents; hematocrit; serum creatinine ≥ 1.3 mg/deciliter; gender; arterial pCO₂. The r² value for ALM prediction of ICU LOS in the initial (training) model was 0.356, p <0.0001. Cross validation in prediction of a “new” patient yielded r² = 0.200, p <0.0001. The same 8 factors analyzed with ANN yielded a training prediction r² of 0.535 (p <0.0001) and a cross validation prediction r² of 0.410, p <0.0001. Two additional predictive algorithms were studied, but they had lower prediction accuracies. Our validated neural network model identified the upper quartile of ICU LOS with an odds ratio of 9.8(p <0.0001).

Conclusions

ANN demonstrated a 2-fold greater accuracy than ALM in prediction of observed ICU LOS. This greater accuracy would be presumed to result from the capacity of ANN to capture nonlinear effects and higher order interactions. Predictive modeling may be of value in early anticipation of risks of post-operative morbidity and utilization of ICU facilities.

Artificial Neural Networks and risk stratification models in Emergency Departments: The policy maker's perspective

The primary goal of Emergency Department (ED) physicians is to discriminate between individuals at low risk, who can be safely discharged, and patients at high risk, who require prompt hospitalization. The problem of correctly classifying patients is an issue involving not only clinical but also managerial aspects, since reducing the rate of admission of patients to EDs could dramatically cut costs. Nevertheless, a trade-off might arise due to the need to find a balance between economic interests and the health conditions of patients. This work considers patients in EDs after a syncope event and presents a comparative analysis between two models: a multivariate logistic regression model, as proposed by the scientific community to stratify the expected risk of severe outcomes in the short and long run, and Artificial Neural Networks (ANNs), an innovative model. The analysis highlights differences in correct classification of severe outcomes at 10 days (98.30% vs. 94.07%) and 1 year (97.67% vs. 96.40%), pointing to the superiority of Neural Networks. According to the results, there is also a significant superiority of ANNs in terms of false negatives both at 10 days (3.70% vs. 5.93%) and at 1 year (2.33% vs. 10.07%). However, considering the false positives, the adoption of ANNs would cause an increase in hospital costs, highlighting the potential trade-off which policy makers might face.

An Automated System for Skeletal Maturity Assessment by Extreme Learning Machines

Assessing skeletal age is a subjective and tedious examination process. Hence, automated assessment methods have been developed to replace manual evaluation in medical applications. In this study, a new fully automated method based on content-based image retrieval and using extreme learning machines (ELM) is designed and adapted to assess skeletal maturity. The main novelty of this approach is it overcomes the segmentation problem as suffered by existing systems. The estimation results of ELM models are compared with those of genetic programming (GP) and artificial neural networks (ANNs) models. The experimental results signify improvement in assessment accuracy over GP and ANN, while generalization capability is possible with the ELM approach. Moreover, the results are indicated that the ELM model developed can be used confidently in further work on formulating novel models of skeletal age assessment strategies. According to the experimental results, the new presented method has the capacity to learn many hundreds of times faster than traditional learning methods and it has sufficient overall performance in many aspects. It has conclusively been found that applying ELM is particularly promising as an alternative method for evaluating skeletal age.

Finding Risk Groups by Optimizing Artificial Neural Networks on the Area under the Survival Curve Using Genetic Algorithms

We investigate a new method to place patients into risk groups in censored survival data. Properties such as median survival time, and end survival rate, are implicitly improved by optimizing the area under the survival curve. Artificial neural networks (ANN) are trained to either maximize or minimize this area using a genetic algorithm, and combined into an ensemble to predict one of low, intermediate, or high risk groups. Estimated patient risk can influence treatment choices, and is important for study stratification. A common approach is to sort the patients according to a prognostic index and then group them along the quartile limits. The Cox proportional hazards model (Cox) is one example of this approach. Another method of doing risk grouping is recursive partitioning (Rpart), which constructs a decision tree where each branch point maximizes the statistical separation between the groups. ANN, Cox, and Rpart are compared on five publicly available data sets with varying properties. Cross-validation, as well as separate test sets, are used to validate the models. Results on the test sets show comparable performance, except for the smallest data set where Rpart’s predicted risk groups turn out to be inverted, an example of crossing survival curves. Cross-validation shows that all three models exhibit crossing of some survival curves on this small data set but that the ANN model manages the best separation of groups in terms of median survival time before such crossings. The conclusion is that optimizing the area under the survival curve is a viable approach to identify risk groups. Training ANNs to optimize this area combines two key strengths from both prognostic indices and Rpart. First, a desired minimum group size can be specified, as for a prognostic index. Second, the ability to utilize non-linear effects among the covariates, which Rpart is also able to do.

Breath analysis as a potential and non-invasive frontier in disease diagnosis: an overview

Currently, a small number of diseases, particularly cardiovascular (CVDs), oncologic (ODs), neurodegenerative (NDDs), chronic respiratory diseases, as well as diabetes, form a severe burden to most of the countries worldwide. Hence, there is an urgent need for development of efficient diagnostic tools, particularly those enabling reliable detection of diseases, at their early stages, preferably using non-invasive approaches. Breath analysis is a non-invasive approach relying only on the characterisation of volatile composition of the exhaled breath (EB) that in turn reflects the volatile composition of the bloodstream and airways and therefore the status and condition of the whole organism metabolism. Advanced sampling procedures (solid-phase and needle traps microextraction) coupled with modern analytical technologies (proton transfer reaction mass spectrometry, selected ion flow tube mass spectrometry, ion mobility spectrometry, e-noses, etc.) allow the characterisation of EB composition to an unprecedented level. However, a key challenge in EB analysis is the proper statistical analysis and interpretation of the large and heterogeneous datasets obtained from EB research. There is no standard statistical framework/protocol yet available in literature that can be used for EB data analysis towards discovery of biomarkers for use in a typical clinical setup. Nevertheless, EB analysis has immense potential towards development of biomarkers for the early disease diagnosis of diseases.

A Predictive E-Health Information System

Diabetes Mellitus is a chronic disease and a major cause of several severe complications and death in both developing and developed countries. The number of diabetes cases world-wide has been climbed up drastically over last decades. Hence, it was of utmost important to manage this illness and to develop tools that help clinicians do their job professionally. Artificial neural networks play a major role herein. In this research, a clinical decision support system that helps in diagnosing diabetes has been developed. The system was implemented using a multilayer perceptron artificial neural network. Due to the fact that there is no systematic way to follow in order to determine the number of hidden layers and neurons in MLP, an algorithm was proposed and followed based on the rules-of-thumb previously defined around this issue. As a result, two different topologies were trained and verified using cross validation technique. The topology that exhibited the best averaged accuracy was that of one hidden layer. The data set was obtained from King Abdullah University Hospital in Jordan.

Artificial neural networks approach to early lung cancer detection

Lung cancer is rated with the highest incidence and mortality every year compared with other forms of cancer, therefore early detection and diagnosis is essential. Artificial Neural Networks (ANNs) are “artificial intelligence” software which have been used to assess a few prognostic situations. In this study, a database containing 193 patients from Diagnostic and Monitoring of Tuberculosis and Illness of Lungs Ward in Kuyavia and Pomerania Centre of the Pulmonology (Bydgoszcz, Poland) was analysed using ANNs. Each patient was described using 48 factors (i.e. age, sex, data of patient history, results from medical examinations etc.) and, as an output value, the expected presence of lung cancer was established. All 48 features were retrospectively collected and the database was divided into a training set (n=97), testing set (n=48) and a validating set (n=48). The best prediction score of the ANN model (MLP 48-9-2) was above 0.99 of the area under a receiver operator characteristic (ROC) curve. The ANNs were able to correctly classify 47 out of 48 test cases. These data suggest that Artificial Neural Networks can be used in prognosis of lung cancer and could help the physician in diagnosis of patients with the suspicion of lung cancer.

Artificial neural network predicts the need for therapeutic ERCP in patients with suspected choledocholithiasis

BACKGROUND

Selection of patients with the highest probability for therapeutic ERCP remains an important task in a clinical workup of patients with suspected choledocholithiasis (CDL).

OBJECTIVE

To determine whether an artificial neural network (ANN) model can improve the accuracy of selecting patients with a high probability of undergoing therapeutic ERCP among those with strong clinical suspicion of CDL and to compare it with our previously reported prediction model.

DESIGN

Prospective, observational study.

SETTING

Single, tertiary-care endoscopy center.

PATIENTS

Between January 2010 and September 2012, we prospectively recruited 291 consecutive patients who underwent ERCP after being referred to our center with firm suspicion for CDL.

INTERVENTIONS

Predictive scores for CDL based on a multivariate logistic regression model and ANN model.

MAIN OUTCOME MEASUREMENTS

The presence of common bile duct stones confirmed by ERCP.

RESULTS

There were 80.4% of patients with positive findings on ERCP. The area under the receiver-operating characteristic curve for our previously established multivariate logistic regression model was 0.787 (95% CI, 0.720-0.854; P < .001), whereas area under the curve for the ANN model was 0.884 (95% CI, 0.831-0.938; P < .001). The ANN model correctly classified 92.3% of patients with positive findings on ERCP and 69.6% patients with negative findings on ERCP.

LIMITATIONS

Only those variables believed to be related to the outcome of interest were included. The majority of patients in our sample had positive findings on ERCP.

CONCLUSIONS

An ANN model has better discriminant ability and accuracy than a multivariate logistic regression model in selecting patients for therapeutic ERCP.

Error-correction learning for artificial neural networks using the Bayesian paradigm. Application to automated medical diagnosis

Automated medical diagnosis models are now ubiquitous, and research for developing new ones is constantly growing. They play an important role in medical decision-making, helping physicians to provide a fast and accurate diagnosis. Due to their adaptive learning and nonlinear mapping properties, the artificial neural networks are widely used to support the human decision capabilities, avoiding variability in practice and errors based on lack of experience. Among the most common learning approaches, one can mention either the classical back-propagation algorithm based on the partial derivatives of the error function with respect to the weights, or the Bayesian learning method based on posterior probability distribution of weights, given training data. This paper proposes a novel training technique gathering together the error-correction learning, the posterior probability distribution of weights given the error function, and the Goodman-Kruskal Gamma rank correlation to assembly them in a Bayesian learning strategy. This study had two main purposes; firstly, to develop anovel learning technique based on both the Bayesian paradigm and the error back-propagation, and secondly,to assess its effectiveness. The proposed model performance is compared with those obtained by traditional machine learning algorithms using real-life breast and lung cancer, diabetes, and heart attack medical databases. Overall, the statistical comparison results indicate that thenovellearning approach outperforms the conventional techniques in almost all respects.

APPLYING MACHINE LEARNING TECHNIQUES IN DETECTING BACTERIAL VAGINOSIS

There are several diseases which arise because of changes in the microbial communities in the body. Scientists continue to conduct research in a quest to find the catalysts that provoke these changes in the naturally occurring microbiota. Bacterial Vaginosis (BV) is a disease that fits the above criteria. BV afflicts approximately 29% of women in child bearing age. Unfortunately, its causes are unknown. This paper seeks to uncover the most important features for diagnosis and in turn employ classification algorithms on those features. In order to fulfill our purpose, we conducted two experiments on the data. We isolated the clinical and medical features from the full set of raw data, we compared the accuracy, precision, recall and F-measure and time elapsed for each feature selection and classification grouping. We noticed that classification results were as good or better after performing feature selection although there was a wide range in the number of features produced from the feature selection process. After comparing the experiments, the algorithms performed best on the medical dataset.

Current breathomics--a review on data pre-processing techniques and machine learning in metabolomics breath analysis

We define breathomics as the metabolomics study of exhaled air. It is a strongly emerging metabolomics research field that mainly focuses on health-related volatile organic compounds (VOCs). Since the amount of these compounds varies with health status, breathomics holds great promise to deliver non-invasive diagnostic tools. Thus, the main aim of breathomics is to find patterns of VOCs related to abnormal (for instance inflammatory) metabolic processes occurring in the human body. Recently, analytical methods for measuring VOCs in exhaled air with high resolution and high throughput have been extensively developed. Yet, the application of machine learning methods for fingerprinting VOC profiles in the breathomics is still in its infancy. Therefore, in this paper, we describe the current state of the art in data pre-processing and multivariate analysis of breathomics data. We start with the detailed pre-processing pipelines for breathomics data obtained from gas-chromatography mass spectrometry and an ion-mobility spectrometer coupled to multi-capillary columns. The outcome of data pre-processing is a matrix containing the relative abundances of a set of VOCs for a group of patients under different conditions (e.g. disease stage, treatment). Independently of the utilized analytical method, the most important question, 'which VOCs are discriminatory?', remains the same. Answers can be given by several modern machine learning techniques (multivariate statistics) and, therefore, are the focus of this paper. We demonstrate the advantages as well the drawbacks of such techniques. We aim to help the community to understand how to profit from a particular method. In parallel, we hope to make the community aware of the existing data fusion methods, as yet unresearched in breathomics.

Data mining for wearable sensors in health monitoring systems: a review of recent trends and challenges

The past few years have witnessed an increase in the development of wearable sensors for health monitoring systems. This increase has been due to several factors such as development in sensor technology as well as directed efforts on political and stakeholder levels to promote projects which address the need for providing new methods for care given increasing challenges with an aging population. An important aspect of study in such system is how the data is treated and processed. This paper provides a recent review of the latest methods and algorithms used to analyze data from wearable sensors used for physiological monitoring of vital signs in healthcare services. In particular, the paper outlines the more common data mining tasks that have been applied such as anomaly detection, prediction and decision making when considering in particular continuous time series measurements. Moreover, the paper further details the suitability of particular data mining and machine learning methods used to process the physiological data and provides an overview of the properties of the data sets used in experimental validation. Finally, based on this literature review, a number of key challenges have been outlined for data mining methods in health monitoring systems.