An optimized experimental protocol based on neuro-evolutionary algorithms application to the classification of dyspeptic patients and to the prediction of the effectiveness of their treatment

Bone Strain Index predicts fragility fracture in osteoporotic women: an artificial intelligence-based study

BACKGROUND

We applied an artificial intelligence-based model to predict fragility fractures in postmenopausal women, using different dual-energy x-ray absorptiometry (DXA) parameters.

METHODS

One hundred seventy-four postmenopausal women without vertebral fractures (VFs) at baseline (mean age 66.3 ± 9.8) were retrospectively evaluated. Data has been collected from September 2010 to August 2018. All subjects performed a spine x-ray to assess VFs, together with lumbar and femoral DXA for bone mineral density (BMD) and the bone strain index (BSI) evaluation. Follow-up exams were performed after 3.34 ± 1.91 years. Considering the occurrence of new VFs at follow-up, two groups were created: fractured versus not-fractured. We applied an artificial neural network (ANN) analysis with a predictive tool (TWIST system) to select relevant input data from a list of 13 variables including BMD and BSI. A semantic connectivity map was built to analyse the connections among variables within the groups. For group comparisons, an independent-samples t-test was used; variables were expressed as mean ± standard deviation.

RESULTS

For each patient, we evaluated a total of n = 6 exams. At follow-up, n = 69 (39.6%) women developed a VF. ANNs reached a predictive accuracy of 79.56% within the training testing procedure, with a sensitivity of 80.93% and a specificity of 78.18%. The semantic connectivity map showed that a low BSI at the total femur is connected to the absence of VFs.

CONCLUSION

We found a high performance of ANN analysis in predicting the occurrence of VFs. Femoral BSI appears as a useful DXA index to identify patients at lower risk for lumbar VFs.

Bone strain index as a predictor of further vertebral fracture in osteoporotic women: An artificial intelligence-based analysis

BACKGROUND

Osteoporosis is an asymptomatic disease of high prevalence and incidence, leading to bone fractures burdened by high mortality and disability, mainly when several subsequent fractures occur. A fragility fracture predictive model, Artificial Intelligence-based, to identify dual X-ray absorptiometry (DXA) variables able to characterise those patients who are prone to further fractures called Bone Strain Index, was evaluated in this study.

METHODS

In a prospective, longitudinal, multicentric study 172 female outpatients with at least one vertebral fracture at the first observation were enrolled. They performed a spine X-ray to calculate spine deformity index (SDI) and a lumbar and femoral DXA scan to assess bone mineral density (BMD) and bone strain index (BSI) at baseline and after a follow-up period of 3 years in average. At the end of the follow-up, 93 women developed a further vertebral fracture. The further vertebral fracture was considered as one unit increase of SDI. We assessed the predictive capacity of supervised Artificial Neural Networks (ANNs) to distinguish women who developed a further fracture from those without it, and to detect those variables providing the maximal amount of relevant information to discriminate the two groups. ANNs choose appropriate input data automatically (TWIST-system, Training With Input Selection and Testing). Moreover, we built a semantic connectivity map usingthe Auto Contractive Map to provide further insights about the convoluted connections between the osteoporotic variables under consideration and the two scenarios (further fracture vs no further fracture).

RESULTS

TWIST system selected 5 out of 13 available variables: age, menopause age, BMI, FTot BMC, FTot BSI. With training testing procedure, ANNs reached predictive accuracy of 79.36%, with a sensitivity of 75% and a specificity of 83.72%. The semantic connectivity map highlighted the role of BSI in predicting the risk of a further fracture.

CONCLUSIONS

Artificial Intelligence is a useful method to analyse a complex system like that regarding osteoporosis, able to identify patients prone to a further fragility fracture. BSI appears to be a useful DXA index in identifying those patients who are at risk of further vertebral fractures.

Development of machine learning models to predict RT-PCR results for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in patients with influenza-like symptoms using only basic clinical data

BACKGROUND

Reverse Transcription-Polymerase Chain Reaction (RT-PCR) for Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-COV-2) diagnosis currently requires quite a long time span. A quicker and more efficient diagnostic tool in emergency departments could improve management during this global crisis. Our main goal was assessing the accuracy of artificial intelligence in predicting the results of RT-PCR for SARS-COV-2, using basic information at hand in all emergency departments.

METHODS

This is a retrospective study carried out between February 22, 2020 and March 16, 2020 in one of the main hospitals in Milan, Italy. We screened for eligibility all patients admitted with influenza-like symptoms tested for SARS-COV-2. Patients under 12 years old and patients in whom the leukocyte formula was not performed in the ED were excluded. Input data through artificial intelligence were made up of a combination of clinical, radiological and routine laboratory data upon hospital admission. Different Machine Learning algorithms available on WEKA data mining software and on Semeion Research Centre depository were trained using both the Training and Testing and the K-fold cross-validation protocol.

RESULTS

Among 199 patients subject to study (median [interquartile range] age 65 [46-78] years; 127 [63.8%] men), 124 [62.3%] resulted positive to SARS-COV-2. The best Machine Learning System reached an accuracy of 91.4% with 94.1% sensitivity and 88.7% specificity.

CONCLUSION

Our study suggests that properly trained artificial intelligence algorithms may be able to predict correct results in RT-PCR for SARS-COV-2, using basic clinical data. If confirmed, on a larger-scale study, this approach could have important clinical and organizational implications.

A New Approach towards Minimizing the Risk of Misdosing Warfarin Initiation Doses

It is a challenge to be able to prescribe the optimal initial dose of warfarin. There have been many studies focused on an efficient strategy to determine the optimal initial dose. Numerous clinical, genetic, and environmental factors affect the warfarin dose response. In practice, it is common that the initial warfarin dose is substantially different from the stable maintenance dose, which may increase the risk of bleeding or thrombosis prior to achieving the stable maintenance dose. In order to minimize the risk of misdosing, despite popular warfarin dose prediction models in the literature which create dose predictions solely based on patients' attributes, we have taken physicians' opinions towards the initial dose into consideration. The initial doses selected by clinicians, along with other standard clinical factors, are used to determine an estimate of the difference between the initial dose and estimated maintenance dose using shrinkage methods. The selected shrinkage method was LASSO (Least Absolute Shrinkage and Selection Operator). The estimated maintenance dose was more accurate than the original initial dose, the dose predicted by a linear model without involving the clinicians initial dose, and the values predicted by the most commonly used model in the literature, the Gage clinical model.

Networks in Coronary Heart Disease Genetics As a Step towards Systems Epidemiology

We present the use of innovative machine learning techniques in the understanding of Coronary Heart Disease (CHD) through intermediate traits, as an example of the use of this class of methods as a first step towards a systems epidemiology approach of complex diseases genetics. Using a sample of 252 middle-aged men, of which 102 had a CHD event in 10 years follow-up, we applied machine learning algorithms for the selection of CHD intermediate phenotypes, established markers, risk factors, and their previously associated genetic polymorphisms, and constructed a map of relationships between the selected variables. Of the 52 variables considered, 42 were retained after selection of the most informative variables for CHD. The constructed map suggests that most selected variables were related to CHD in a context dependent manner while only a small number of variables were related to a specific outcome. We also observed that loss of complexity in the network was linked to a future CHD event. We propose that novel, non-linear, and integrative epidemiological approaches are required to combine all available information, in order to truly translate the new advances in medical sciences to gains in preventive measures and patients care.

An improved I-FAST system for the diagnosis of Alzheimer's disease from unprocessed electroencephalograms by using robust invariant features

OBJECTIVE

This paper proposes a new, complex algorithm for the blind classification of the original electroencephalogram (EEG) tracing of each subject, without any preliminary pre-processing. The medical need in this field is to reach an early differential diagnosis between subjects affected by mild cognitive impairment (MCI), early Alzheimer's disease (AD) and the healthy elderly (CTR) using only the recording and the analysis of few minutes of their EEG.

METHODS AND MATERIAL

This study analyzed the EEGs of 272 subjects, recorded at Rome's Neurology Unit of the Policlinico Campus Bio-Medico. The EEG recordings were performed using 19 electrodes, in a 0.3-70Hz bandpass, positioned according to the International 10-20 System. Many powerful learning machines and algorithms have been proposed during the last 20 years to effectively resolve this complex problem, resulting in different and interesting outcomes. Among these algorithms, a new artificial adaptive system, named implicit function as squashing time (I-FAST), is able to diagnose, with high accuracy, a few minutes of the subject's EEG track; whether it manifests an AD, MCI or CTR condition. An updating of this system, carried out by adding a new algorithm, named multi scale ranked organizing maps (MS-ROM), to the I-FAST system, is presented, in order to classify with greater accuracy the unprocessed EEG's of AD, MCI and control subjects.

RESULTS

The proposed system has been measured on three independent pattern recognition tasks from unprocessed EEG tracks of a sample of AD subjects, MCI subjects and CTR: (a) AD compared with CTR; (b) AD compared with MCI; (c) CTR compared with MCI. While the values of accuracy of the previous system in distinguishing between AD and MCI were around 92%, the new proposed system reaches values between 94% and 98%. Similarly, the overall accuracy with best artificial neural networks (ANNs) is 98.25% for the distinguishing between CTR and MCI.

CONCLUSIONS

This new version of I-FAST makes different steps forward: (a) avoidance of pre-processing phase and filtering procedure of EEG data, being the algorithm able to directly process an unprocessed EEG; (b) noise elimination, through the use of a training variant with input selection and testing system, based on naïve Bayes classifier; (c) a more robust classification phase, showing the stability of results on nine well known learning machine algorithms; (d) extraction of spatial invariants of an EEG signal using, in addition to the unsupervised ANN, the principal component analysis and the multi scale entropy, together with the MS-ROM; a more accurate performance in this specific task.

Outcome predictors in autism spectrum disorders preschoolers undergoing treatment as usual: insights from an observational study using artificial neural networks

BACKGROUND

Treatment as usual (TAU) for autism spectrum disorders (ASDs) includes eclectic treatments usually available in the community and school inclusion with an individual support teacher. Artificial neural networks (ANNs) have never been used to study the effects of treatment in ASDs. The Auto Contractive Map (Auto-CM) is a kind of ANN able to discover trends and associations among variables creating a semantic connectivity map. The matrix of connections, visualized through a minimum spanning tree filter, takes into account nonlinear associations among variables and captures connection schemes among clusters. Our aim is to use Auto-CM to recognize variables to discriminate between responders versus no responders at TAU.

METHODS

A total of 56 preschoolers with ASDs were recruited at different sites in Italy. They were evaluated at T0 and after 6 months of treatment (T1). The children were referred to community providers for usual treatments.

RESULTS

At T1, the severity of autism measured through the Autism Diagnostic Observation Schedule decreased in 62% of involved children (Response), whereas it was the same or worse in 37% of the children (No Response). The application of the Semeion ANNs overcomes the 85% of global accuracy (Sine Net almost reaching 90%). Consequently, some of the tested algorithms were able to find a good correlation between some variables and TAU outcome. The semantic connectivity map obtained with the application of the Auto-CM system showed results that clearly indicated that "Response" cases can be visually separated from the "No Response" cases. It was possible to visualize a response area characterized by "Parents Involvement high". The resultant No Response area strongly connected with "Parents Involvement low".

CONCLUSION

The ANN model used in this study seems to be a promising tool for the identification of the variables involved in the positive response to TAU in autism.

Prediction of optimal warfarin maintenance dose using advanced artificial neural networks

BACKGROUND

In recent years, pharmacogenetic algorithms were developed for estimating the appropriate dose of vitamin K antagonists.

AIM

To evaluate the performance of new generation artificial neural networks (ANNs) to predict the warfarin maintenance dose.

METHODS

Demographic, clinical and genetic data (CYP2C9 and VKORC1 polymorphisms) from 377 patients treated with warfarin were used. The final prediction model was based on 23 variables selected by TWIST® system within a bipartite division of the data set (training and testing) protocol.

RESULTS

The ANN algorithm reached high accuracy, with an average absolute error of 5.7 mg of the warfarin maintenance dose. In the subset of patients requiring ≤21 mg and 21-49 mg (45 and 51% of the cohort, respectively) the absolute error was 3.86 mg and 5.45 with a high percentage of subjects being correctly identified (71 and 73%, respectively).

CONCLUSION

ANN appears to be a promising tool for vitamin K antagonist maintenance dose prediction.

Artificial Neural Networks, and Evolutionary Algorithms as a systems biology approach to a data-base on fetal growth restriction

One of the specific aims of systems biology is to model and discover properties of cells, tissues and organisms functioning. A systems biology approach was undertaken to investigate possibly the entire system of intra-uterine growth we had available, to assess the variables of interest, discriminate those which were effectively related with appropriate or restricted intrauterine growth, and achieve an understanding of the systems in these two conditions. The Artificial Adaptive Systems, which include Artificial Neural Networks and Evolutionary Algorithms lead us to the first analyses. These analyses identified the importance of the biochemical variables IL-6, IGF-II and IGFBP-2 protein concentrations in placental lysates, and offered a new insight into placental markers of fetal growth within the IGF and cytokine systems, confirmed they had relationships and offered a critical assessment of studies previously performed.

Application of artificial neural networks to investigate one-carbon metabolism in Alzheimer's disease and healthy matched individuals

Folate metabolism, also known as one-carbon metabolism, is required for several cellular processes including DNA synthesis, repair and methylation. Impairments of this pathway have been often linked to Alzheimer's disease (AD). In addition, increasing evidence from large scale case-control studies, genome-wide association studies, and meta-analyses of the literature suggest that polymorphisms of genes involved in one-carbon metabolism influence the levels of folate, homocysteine and vitamin B12, and might be among AD risk factors. We analyzed a dataset of 30 genetic and biochemical variables (folate, homocysteine, vitamin B12, and 27 genotypes generated by nine common biallelic polymorphisms of genes involved in folate metabolism) obtained from 40 late-onset AD patients and 40 matched controls to assess the predictive capacity of Artificial Neural Networks (ANNs) in distinguish consistently these two different conditions and to identify the variables expressing the maximal amount of relevant information to the condition of being affected by dementia of Alzheimer's type. Moreover, we constructed a semantic connectivity map to offer some insight regarding the complex biological connections among the studied variables and the two conditions (being AD or control). TWIST system, an evolutionary algorithm able to remove redundant and noisy information from complex data sets, selected 16 variables that allowed specialized ANNs to discriminate between AD and control subjects with over 90% accuracy. The semantic connectivity map provided important information on the complex biological connections among one-carbon metabolic variables highlighting those most closely linked to the AD condition.

Role of XPC, XPD, XRCC1, GSTP genetic polymorphisms and Barrett's esophagus in a cohort of Italian subjects. A neural network analysis

BACKGROUND

Barrett's esophagus (BE), a metaplastic premalignant disorder, represents the primary risk factor for the development of esophageal adenocarcinoma. Chronic gastroesophageal reflux disease and central obesity have been associated with BE and esophageal adenocarcinoma, but relatively little is known about the specific genes that confer susceptibility to BE carcinogenesis.

METHODS

A total of 74 patients with BE and 67 controls coming from six gastrointestinal Italian units were evaluated for six polymorphisms in four genes: XPC, XPD nucleotide excision repair (NER) genes, XRCC1 (BER gene), and glutathione S-transferase P1. Smoking status was analyzed together with the genetic data. Statistical analysis was performed through Artificial Neural Networks.

RESULTS

Distributions of sex, smoking history, and polymorphisms among BE cases and controls did not show statistically significant differences. The r-value from linear correlation allowed us to identify possible protective factors as well as possible risk factors. The application of advanced intelligent systems allowed for the selection of a subgroup of nine variables. Artificial Neural Networks applied on the final data set reached mean global accuracy of 60%, reaching as high as 65.88%.

CONCLUSION

We report here results from an exploratory study. Results from this study failed to find an association among the tested single nucleotide polymorphisms and BE phenotype through classical statistical methods. On the contrary, advanced intelligent systems are really able to handle the disease complexity, not treating the data with reductionist approaches unable to detect multiple genes of smaller effect in predisposing to the disease.

IMPACT

To detect multiple genes of smaller effects in predisposing individuals to Barrett's esophagus.

Low bone mineral density and its predictors in type 1 diabetic patients evaluated by the classic statistics and artificial neural network analysis

OBJECTIVE

To investigate factors associated with bone mineral density (BMD) in type 1 diabetes by classic statistic and artificial neural networks.

RESEARCH DESIGN AND METHODS

A total of 175 eugonadal type 1 diabetic patients (age 32.8 ± 8.4 years) and 151 age- and BMI-matched control subjects (age 32.6 ± 4.5 years) were studied. In all subjects, BMI and BMD (as Z score) at the lumbar spine (LS-BMD) and femur (F-BMD) were measured. Daily insulin dose (DID), age at diagnosis, presence of complications, creatinine clearance (ClCr), and HbA(1c) were determined.

RESULTS

LS- and F-BMD levels were lower in patients (-0.11 ± 1.2 and -0.32 ± 1.4, respectively) than in control subjects (0.59 ± 1, P < 0.0001, and 0.63 ± 1, P < 0.0001, respectively). LS-BMD was independently associated with BMI and DID, whereas F-BMD was associated with BMI and ClCr. The cutoffs for predicting low BMD were as follows: BMI <23.5 kg/m(2), DID >0.67 units/kg, and ClCr <88.8 mL/min. The presence of all of these risk factors had a positive predictive value, and their absence had a negative predictive value for low BMD of 62.9 and 84.2%, respectively. Data were also analyzed using the TWIST system in combination with supervised artificial neural networks and a semantic connectivity map. The TWIST system selected 11 and 12 variables for F-BMD and LS-BMD prediction, which discriminated between high and low BMD with 67 and 66% accuracy, respectively. The connectivity map showed that low BMD at both sites was indirectly connected with HbA(1c) through chronic diabetes complications.

CONCLUSIONS

In type 1 diabetes, low BMD is associated with low BMI and low ClCr and high DID. Chronic complications negatively influence BMD.

Recognition of morphometric vertebral fractures by artificial neural networks: analysis from GISMO Lombardia Database

BACKGROUND

It is known that bone mineral density (BMD) predicts the fracture's risk only partially and the severity and number of vertebral fractures are predictive of subsequent osteoporotic fractures (OF). Spinal deformity index (SDI) integrates the severity and number of morphometric vertebral fractures. Nowadays, there is interest in developing algorithms that use traditional statistics for predicting OF. Some studies suggest their poor sensitivity. Artificial Neural Networks (ANNs) could represent an alternative. So far, no study investigated ANNs ability in predicting OF and SDI. The aim of the present study is to compare ANNs and Logistic Regression (LR) in recognising, on the basis of osteoporotic risk-factors and other clinical information, patients with SDI≥1 and SDI≥5 from those with SDI = 0.

METHODOLOGY

We compared ANNs prognostic performance with that of LR in identifying SDI≥1/SDI≥5 in 372 women with postmenopausal-osteoporosis (SDI≥1, n = 176; SDI = 0, n = 196; SDI≥5, n = 51), using 45 variables (44 clinical parameters plus BMD). ANNs were allowed to choose relevant input data automatically (TWIST-system-Semeion). Among 45 variables, 17 and 25 were selected by TWIST-system-Semeion, in SDI≥1 vs SDI = 0 (first) and SDI≥5 vs SDI = 0 (second) analysis. In the first analysis sensitivity of LR and ANNs was 35.8% and 72.5%, specificity 76.5% and 78.5% and accuracy 56.2% and 75.5%, respectively. In the second analysis, sensitivity of LR and ANNs was 37.3% and 74.8%, specificity 90.3% and 87.8%, and accuracy 63.8% and 81.3%, respectively.

CONCLUSIONS

ANNs showed a better performance in identifying both SDI≥1 and SDI≥5, with a higher sensitivity, suggesting its promising role in the development of algorithm for predicting OF.

Is it possible to clinically differentiate erosive from nonerosive reflux disease patients? A study using an artificial neural networks-assisted algorithm

BACKGROUND

The use of either symptom questionnaires or artificial neural networks (ANNs) has proven to improve the accuracy in diagnosing gastroesophageal reflux disease (GERD). However, the differentiation between the erosive and nonerosive reflux disease based upon symptoms at presentation still remains inconclusive.

AIM

To assess the capability of a combined approach, that is, the use of a novel GERD questionnaire - the QUestionario Italiano Diagnostico (QUID) questionnaire - and of an ANNs-assisted algorithm, to discriminate between nonerosive gastroesophageal reflux disease (NERD) and erosive esophagitis (EE) patients.

METHODS

Five hundred and fifty-seven adult outpatients with typical GERD symptoms and 94 asymptomatic adult patients, were submitted to the QUID questionnaire. GERD patients were then submitted to upper gastrointestinal endoscopy to differentiate them between EE and NERD patients.

RESULTS

The QUID score resulted significantly (P<0.001) higher in GERD patients versus controls, but it was not statistically significantly different between EE and NERD patients. ANNs assisted diagnosis had greater specificity, sensitivity and accuracy compared with the linear discriminant analysis only to differentiate GERD patients from controls. However, no single technique was able to satisfactorily discriminate between EE and NERD patients.

CONCLUSION

Our study suggests that the combination between QUID questionnaire and an ANNs-assisted algorithm is useful only to differentiate GERD patients from healthy individuals but fails to further discriminate erosive from nonerosive patients.

Polymorphisms in folate-metabolizing genes, chromosome damage, and risk of Down syndrome in Italian women: identification of key factors using artificial neural networks

BACKGROUND

Studies in mothers of Down syndrome individuals (MDS) point to a role for polymorphisms in folate metabolic genes in increasing chromosome damage and maternal risk for a Down syndrome (DS) pregnancy, suggesting complex gene-gene interactions. This study aimed to analyze a dataset of genetic and cytogenetic data in an Italian group of MDS and mothers of healthy children (control mothers) to assess the predictive capacity of artificial neural networks assembled in TWIST system in distinguish consistently these two different conditions and to identify the variables expressing the maximal amount of relevant information to the condition of being mother of a DS child.The dataset consisted of the following variables: the frequency of chromosome damage in peripheral lymphocytes (BNMN frequency) and the genotype for 7 common polymorphisms in folate metabolic genes (MTHFR 677C>T and 1298A>C, MTRR 66A>G, MTR 2756A>G, RFC1 80G>A and TYMS 28bp repeats and 1494 6bp deletion). Data were analysed using TWIST system in combination with supervised artificial neural networks, and a semantic connectivity map.

RESULTS

TWIST system selected 6 variables (BNMN frequency, MTHFR 677TT, RFC1 80AA, TYMS 1494 6bp +/+, TYMS 28bp 3R/3R and MTR 2756AA genotypes) that were subsequently used to discriminate between MDS and control mothers with 90% accuracy. The semantic connectivity map provided important information on the complex biological connections between the studied variables and the two conditions (being MDS or control mother).

CONCLUSIONS

Overall, the study suggests a link between polymorphisms in folate metabolic genes and DS risk in Italian women.

Predicting the outcome of patients with subarachnoid hemorrhage using machine learning techniques

BACKGROUND

Outcome prediction for subarachnoid hemorrhage (SAH) helps guide care and compare global management strategies. Logistic regression models for outcome prediction may be cumbersome to apply in clinical practice.

OBJECTIVE

To use machine learning techniques to build a model of outcome prediction that makes the knowledge discovered from the data explicit and communicable to domain experts.

MATERIAL AND METHODS

A derivation cohort (n = 441) of nonselected SAH cases was analyzed using different classification algorithms to generate decision trees and decision rules. Algorithms used were C4.5, fast decision tree learner, partial decision trees, repeated incremental pruning to produce error reduction, nearest neighbor with generalization, and ripple down rule learner. Outcome was dichotomized in favorable [Glasgow outcome scale (GOS) = I-II] and poor (GOS = III-V). An independent cohort (n = 193) was used for validation. An exploratory questionnaire was given to potential users (specialist doctors) to gather their opinion on the classifier and its usability in clinical routine.

RESULTS

The best classifier was obtained with the C4.5 algorithm. It uses only two attributes [World Federation of Neurological Surgeons (WFNS) and Fisher's scale] and leads to a simple decision tree. The accuracy of the classifier [area under the ROC curve (AUC) = 0.84; confidence interval (CI) = 0.80-0.88] is similar to that obtained by a logistic regression model (AUC = 0.86; CI = 0.83-0.89) derived from the same data and is considered better fit for clinical use.

Neural network analysis of dynamic sequences of EUS elastography used for the differential diagnosis of chronic pancreatitis and pancreatic cancer

BACKGROUND

EUS elastography is a newly developed imaging procedure that characterizes the differences of hardness and strain between diseased and normal tissue.

OBJECTIVE

To assess the accuracy of real-time EUS elastography in pancreatic lesions.

DESIGN

Cross-sectional feasibility study.

PATIENTS

The study group included, in total, 68 patients with normal pancreas (N = 22), chronic pancreatitis (N = 11), pancreatic adenocarcinoma (N = 32), and pancreatic neuroendocrine tumors (N = 3). A subgroup analysis of 43 cases with focal pancreatic masses was also performed.

INTERVENTIONS

A postprocessing software analysis was used to examine the EUS elastography movies by calculating hue histograms of each individual image, data that were further subjected to an extended neural network analysis to differentiate benign from malignant patterns.

MAIN OUTCOME MEASUREMENTS

To differentiate normal pancreas, chronic pancreatitis, pancreatic cancer, and neuroendocrine tumors.

RESULTS

Based on a cutoff of 175 for the mean hue histogram values recorded on the region of interest, the sensitivity, specificity, and accuracy of differentiation of benign and malignant masses were 91.4%, 87.9%, and 89.7%, respectively. The positive and negative predictive values were 88.9% and 90.6%, respectively. Multilayer perceptron neural networks with both one and two hidden layers of neurons (3-layer perceptron and 4-layer perceptron) were trained to learn how to classify cases as benign or malignant, and yielded an excellent testing performance of 95% on average, together with a high training performance that equaled 97% on average.

LIMITATION

A lack of the surgical standard in all cases.

CONCLUSIONS

EUS elastography is a promising method that allows characterization and differentiation of normal pancreas, chronic pancreatitis, and pancreatic cancer. The currently developed methodology, based on artificial neural network processing of EUS elastography digitalized movies, enabled an optimal prediction of the types of pancreatic lesions. Future multicentric, randomized studies with adequate power will have to establish the clinical impact of this procedure for the differential diagnosis of focal pancreatic masses.

The Framingham study and treatment guidelines for stroke prevention

In recent years, institutional bodies and scientific societies of principal Western countries have produced several guidelines dealing with risk assessment, primary prevention, and treatment of acute stroke. From a prospective, community-based, observational cohort of patients from the Framingham Heart Study, an absolute estimate of risk for stroke alone or stroke or death was determined based on several risk factors, including advanced age, female sex, increased systolic blood pressure, prior stroke or transient ischemic attack, and diabetes mellitus. This algorithm considers many variables and expresses their results as the percentage of risk of developing a fatal or nonfatal stroke in the following 5 years. The author has identified three major pitfalls of this algorithm, which are related to the limitation of the classic statistical approach in handling this kind of nonlinear and complex information: 1) the very large confidence interval of individual risk assessment, 2) the inability to capture the process dynamics, and 3) the inability to capture the disease complexity. The artificial intelligence armamentarium may provide an advantage in the attempt to overcome these limitations. The theoretic background and some application examples related to artificial neural networks (ANNs) and fuzzy logic are reviewed and discussed. Newer approaches linked to artificial intelligence, such as fuzzy logic and ANNs, seem better at addressing the challenge of the increasing complexity of the predisposing factors linked to cerebrovascular events and at predicting future events in an individual patient.

New application of intelligent agents in sporadic amyotrophic lateral sclerosis identifies unexpected specific genetic background

BACKGROUND

Few genetic factors predisposing to the sporadic form of amyotrophic lateral sclerosis (ALS) have been identified, but the pathology itself seems to be a true multifactorial disease in which complex interactions between environmental and genetic susceptibility factors take place. The purpose of this study was to approach genetic data with an innovative statistical method such as artificial neural networks to identify a possible genetic background predisposing to the disease. A DNA multiarray panel was applied to genotype more than 60 polymorphisms within 35 genes selected from pathways of lipid and homocysteine metabolism, regulation of blood pressure, coagulation, inflammation, cellular adhesion and matrix integrity, in 54 sporadic ALS patients and 208 controls. Advanced intelligent systems based on novel coupling of artificial neural networks and evolutionary algorithms have been applied. The results obtained have been compared with those derived from the use of standard neural networks and classical statistical analysis

RESULTS

Advanced intelligent systems based on novel coupling of artificial neural networks and evolutionary algorithms have been applied. The results obtained have been compared with those derived from the use of standard neural networks and classical statistical analysis. An unexpected discovery of a strong genetic background in sporadic ALS using a DNA multiarray panel and analytical processing of the data with advanced artificial neural networks was found. The predictive accuracy obtained with Linear Discriminant Analysis and Standard Artificial Neural Networks ranged from 70% to 79% (average 75.31%) and from 69.1 to 86.2% (average 76.6%) respectively. The corresponding value obtained with Advanced Intelligent Systems reached an average of 96.0% (range 94.4 to 97.6%). This latter approach allowed the identification of seven genetic variants essential to differentiate cases from controls: apolipoprotein E arg158cys; hepatic lipase -480 C/T; endothelial nitric oxide synthase 690 C/T and glu298asp; vitamin K-dependent coagulation factor seven arg353glu, glycoprotein Ia/IIa 873 G/A and E-selectin ser128arg.

CONCLUSION

This study provides an alternative and reliable method to approach complex diseases. Indeed, the application of a novel artificial intelligence-based method offers a new insight into genetic markers of sporadic ALS pointing out the existence of a strong genetic background.

Artificial Neural Networks for Early Prediction of Mortality in Patients with Non Variceal Upper GI Bleeding (UGIB)

BACKGROUND

Mortality for non variceal upper gastrointestinal bleeding (UGIB) is clinically relevant in the first 12-24 hours of the onset of haemorrhage and therefore identification of clinical factors predictive of the risk of death before endoscopic examination may allow for early corrective therapeutic intervention.

AIM

1) Identify simple and early clinical variables predictive of the risk of death in patients with non variceal UGIB; 2) assess previsional gain of a predictive model developed with conventional statistics vs. that developed with artificial neural networks (ANNs).

METHODS AND RESULTS

Analysis was performed on 807 patients with nonvariceal UGIB (527 males, 280 females), as a part of a multicentre Italian study. The mortality was considered "bleeding-related" if occurred within 30 days from the index bleeding episode. A total of 50 independent variables were analysed, 49 of which clinico-anamnestic, all collected prior to endoscopic examination plus the haemoglobin value measured on admission in the emergency department. Death occurred in 42 (5.2%). Conventional statistical techniques (linear discriminant analysis) were compared with ANNs (Twist® system-Semeion) adopting the same result validation protocol with random allocation of the sample in training and testing subsets and subsequent cross-over. ANNs resulted to be significantly more accurate than LDA with an overall accuracy rate near to 90%.

CONCLUSION

Artificial neural networks technology is highly promising in the development of accurate diagnostic tools designed to recognize patients at high risk of death for UGIB.

Artificial neural networks can classify uninvestigated patients with dyspepsia

There is consensus on investigating older patients presenting with or without alarm symptoms and/or risk factors, and irrespective of their Helicobacter pylori status. Remaining patients with uninvestigated dyspepsia, however, represents a 'grey' population for whom no clearly defined guidelines have been delineated. Physicians often struggle with the decision of whether or not to undertake noninvasive testing, treat dyspeptic patients empirically or perform an invasive endoscopy of the upper gastrointestinal tract. We have explored the contribution of artificial neural networks (ANNs) to provide appropriate interpretation of presenting complaints and clinical characteristics for these patients. By taking into account all the 86 recorded features of 101 dyspeptic patients, the overall predictive capability of ANNs in sorting out organic from functional disease amounted to 74.2% and increased to a figure of 85.0% when only the 55 best performing input variables were analyzed. The ANNs performed much better in extracting those patients with a functional dyspepsia (90% accuracy rate), but even in patients with organic disease the 80% accuracy value was remarkable. In patients with an uninvestigated dyspepsia, ANNs found a unique combination of socioenvironmental data, past medical history, risk factors for organic disease, and presenting abdominal complaints that each patient brings to the clinical encounter. With this ability, ANNs can be used to assist in the classification and treatment of patients with uninvestigated dyspepsia, and to bring a greater level of confidence to this process.

The IFAST model, a novel parallel nonlinear EEG analysis technique, distinguishes mild cognitive impairment and Alzheimer's disease patients with high degree of accuracy

OBJECTIVE

This paper presents the results obtained with the innovative use of special types of artificial neural networks (ANNs) assembled in a novel methodology named IFAST (implicit function as squashing time) capable of compressing the temporal sequence of electroencephalographic (EEG) data into spatial invariants. The aim of this study is to assess the potential of this parallel and nonlinear EEG analysis technique in distinguishing between subjects with mild cognitive impairment (MCI) and Alzheimer's disease (AD) patients with a high degree of accuracy in comparison with standard and advanced nonlinear techniques. The principal aim of the study was testing the hypothesis that automatic classification of MCI and AD subjects can be reasonably correct when the spatial content of the EEG voltage is properly extracted by ANNs.

METHODS AND MATERIAL

Resting eyes-closed EEG data were recorded in 180 AD patients and in 115 MCI subjects. The spatial content of the EEG voltage was extracted by IFAST step-wise procedure using ANNs. The data input for the classification operated by ANNs were not the EEG data, but the connections weights of a nonlinear auto-associative ANN trained to reproduce the recorded EEG tracks. These weights represented a good model of the peculiar spatial features of the EEG patterns at scalp surface. The classification based on these parameters was binary (MCI versus AD) and was performed by a supervised ANN. Half of the EEG database was used for the ANN training and the remaining half was utilised for the automatic classification phase (testing).

RESULTS

The best results distinguishing between AD and MCI reached to 92.33%. The comparative results obtained with the best method so far described in the literature, based on blind source separation and Wavelet pre-processing, were 80.43% (p<0.001).

CONCLUSION

The results confirmed the working hypothesis that a correct automatic classification of MCI and AD subjects can be obtained extracting spatial information content of the resting EEG voltage by ANNs and represent the basis for research aimed at integrating spatial and temporal information content of the EEG.

International experience on the use of artificial neural networks in gastroenterology

In this paper, we reconsider the scientific background for the use of artificial intelligence tools in medicine. A review of some recent significant papers shows that artificial neural networks, the more advanced and effective artificial intelligence technique, can improve the classification accuracy and survival prediction of a number of gastrointestinal diseases. We discuss the 'added value' the use of artificial neural networks-based tools can bring in the field of gastroenterology, both at research and clinical application level, when compared with traditional statistical or clinical-pathological methods.

The implicit function as squashing time model: a novel parallel nonlinear EEG analysis technique distinguishing mild cognitive impairment and Alzheimer's disease subjects with high degree of accuracy

Objective. This paper presents the results obtained using a protocol based on special types of artificial neural networks (ANNs) assembled in a novel methodology able to compress the temporal sequence of electroencephalographic (EEG) data into spatial invariants for the automatic classification of mild cognitive impairment (MCI) and Alzheimer's disease (AD) subjects. With reference to the procedure reported in our previous study (2007), this protocol includes a new type of artificial organism, named TWIST. The working hypothesis was that compared to the results presented by the workgroup (2007); the new artificial organism TWIST could produce a better classification between AD and MCI. Material and methods. Resting eyes-closed EEG data were recorded in 180 AD patients and in 115 MCI subjects. The data inputs for the classification, instead of being the EEG data, were the weights of the connections within a nonlinear autoassociative ANN trained to generate the recorded data. The most relevant features were selected and coincidently the datasets were split in the two halves for the final binary classification (training and testing) performed by a supervised ANN. Results. The best results distinguishing between AD and MCI were equal to 94.10% and they are considerable better than the ones reported in our previous study ( approximately 92%) (2007). Conclusion. The results confirm the working hypothesis that a correct automatic classification of MCI and AD subjects can be obtained by extracting spatial information content of the resting EEG voltage by ANNs and represent the basis for research aimed at integrating spatial and temporal information content of the EEG.

Assessment of the role of genetic polymorphism in venous thrombosis through artificial neural networks

PURPOSE

To assess the role of genetic polymorphisms in venous thrombosis events (VTE) using Artificial Neural Networks (ANNs), a model for solving non-linear problems frequently associated with complex biological systems, due to interactions between biological, genetic and environmental factors.

METHODS

A database was generated from a case-control study of venous thrombosis, using 238 patients and 211 controls. The database of 64 variables included age, gender and a panel of 62 genetic variants. Three different ANNs were compared, with logistic regression for the accuracy of predicting cases and controls.

RESULTS

ANNs yielded a better performance than the logistic regression algorithm. Indeed, through ANNs models, the 62 variables related to genetic variants were first reduced to a set of 9, and then of 3 (MTHFR 677 C/T, FV arg506gln, ICAM1 gly214arg).

CONCLUSIONS

The findings of this study illustrate the power of ANN in evaluating multifactorial data, and show that the different sensitivities of the models of elaboration are related to the characteristics of the data. This may contribute to a better understanding of the role played by genetic polymorphisms in VTE, and help to define, if possible, a test panel of genetic variants to estimate an individual's probability of developing the disease.

Possible contribution of advanced statistical methods (artificial neural networks and linear discriminant analysis) in recognition of patients with suspected atrophic body gastritis

AIM: To investigating whether ANNs and LDA could recognize patients with ABG in a database, containing only clinical and biochemical variables, of a pool of patients with and without ABG, by selecting the most predictive variables and by reducing input data to the minimum.

METHODS: Data was collected from 350 consecutive outpatients (263 with ABG, 87 with non-atrophic gastritis and/or celiac disease [controls]). Structured questionnaires with 22 items (anagraphic, anamnestic, clinical, and biochemical data) were filled out for each patient. All patients underwent gastroscopy with biopsies. ANNs and LDA were applied to recognize patients with ABG. Experiment 1: random selection on 37 variables, experiment 2: optimization process on 30 variables, experiment 3: input data reduction on 8 variables, experiment 4: use of only clinical input data on 5 variables, and experiment 5: use of only serological variables.

RESULTS: In experiment 1, overall accuracies of ANNs and LDA were 96.6% and 94.6%, respectively, for predicting patients with ABG. In experiment 2, ANNs and LDA reached an overall accuracy of 98.8% and 96.8%, respectively. In experiment 3, overall accuracy of ANNs was 98.4%. In experiment 4, overall accuracies of ANNs and LDA were, respectively, 91.3% and 88.6%. In experiment 5, overall accuracies of ANNs and LDA were, respectively, 97.7% and 94.5%.

CONCLUSION: This preliminary study suggests that advanced statistical methods, not only ANNs, but also LDA, may contribute to better address bioptic sampling during gastroscopy in a subset of patients in whom ABG may be suspected on the basis of aspecific gastrointestinal symptoms or non-digestive disorders.