Application of fuzzy neural network model and current-voltage analysis of biologically active points for prediction post-surgery risks

Using Fuzzy Mathematical Model in the Differential Diagnosis of Pancreatic Lesions Using Ultrasonography and Echographic Texture Analysis

Malignant tumors of the pancreas are the fourth leading cause of cancer-related deaths. This is mainly because they are often diagnosed at a late stage. One of the challenges in diagnosing focal lesions in the pancreas is the difficulty in distinguishing them from other conditions due to the unique location and anatomy of the organ, as well as the similarity in their ultrasound characteristics. One of the most sensitive imaging modalities of the pancreas is endoscopic ultrasonography. However, clinicians recognize that EUS is a difficult and highly operator-dependent method, while its results are highly dependent on the experience of the investigator. Hybrid technologies based on artificial intelligence methods can improve the accuracy and objectify the results of endosonographic diagnostics. Endoscopic ultrasonography was performed on 272 patients with focal lesions of the pancreatobiliary zone, who had been treated in the surgical section of the Kursk Regional Clinical Hospital in 2014-2023. The study utilized an Olympus EVIS EXERA II video information endoscopic system, along with an EU-ME1 ultrasound unit equipped with GF UM160 and GF UC140P-AL5 echo endoscopes. Out of the focal formations in the pancreatobiliary zone, pancreatic cancer was detected in 109 patients, accounting for 40.1% of the cases. Additionally, 40 patients (14.7%) were diagnosed with local forms of chronic pancreatitis. The reference sonograms displayed distinguishable focal pancreatic pathologies, leading to the development of hybrid fuzzy mathematical decision-making rules at the South-West State University in Kursk, Russian Federation. This research resulted in the creation of a fuzzy hybrid model for the differential diagnosis of chronic focal pancreatitis and pancreatic cancer. Endoscopic ultrasonography, combined with hybrid fuzzy logic methodology, has made it possible to create a model for differentiating between chronic focal pancreatitis and pancreatic ductal adenocarcinoma. Statistical testing on control samples has shown that the diagnostic model, based on reference endosonograms of the echographic texture of pancreatic focal pathology, has a confidence level of 0.6 for the desired diagnosis. By incorporating additional information about the contours of focal formations obtained through endosonography, the reliability of the diagnosis can be increased to 0.9. This level of reliability is considered acceptable in clinical practice and allows for the use of the developed model, even with data that is not well-structured.

Classifier for the functional state of the respiratory system via descriptors determined by using multimodal technology

Currently, intelligent systems built on a multimodal basis are used to study the functional state of living objects. Its essence lies in the fact that a decision is made through several independent information channels with the subsequent aggregation of these decisions. The method of forming descriptors for classifiers of the functional state of the respiratory system includes the study of the spectral range of the respiratory rhythm and the construction of the wavelet plane of the monitoring electrocardiosignal overlapping this range. Then, variations in the breathing rhythm are determined along the corresponding lines of the wavelet plane. Its analysis makes it possible to select slow waves corresponding to the breathing rhythm and systemic waves of the second order. Analysis of the spectral characteristics of these waves makes it possible to form a space of informative features for classifiers of the functional state of the respiratory system. To construct classifiers of the functional state of the respiratory system, hierarchical classifiers were used. As an example, we took a group of patients with pneumonia with a well-defined diagnosis (radiography, X-ray tomography, laboratory data) and a group of volunteers without pulmonary pathology. The diagnostic sensitivity of the obtained classifier was 76% specificity with a diagnostic specificity of 82%, which is comparable to the results of X-ray studies. It is shown that the corresponding lines of the wavelet planes are correlated with the respiratory system and, using their Fourier analysis, descriptors can be obtained for training neural network classifiers of the functional state of the respiratory system.

Development of a Fuzzy Diagnostic Model of Ischemic Disease of the Lower Limbs for Different Stages of Patient Management

Ischemic disease has severe impact on patients which makes accurate diagnosis vital for health protection. Improving the quality of prediction of patients with ischemic extremity disease by using hybrid fuzzy model allows for early and accurate prognosis of the development of the disease at various stages. The prediction of critical ischemia of lower extremity (CLI) at various disease stages is complex problem due to inter-related factors. We developed hybrid fuzzy decision rules to classify ischemic severity using clinical thinking (natural intelligence) with artificial intelligence, which allows achieving a new quality in solving complex systemic problems and is innovative. In this study mathematical model was developed to classify the risk level of CLI into: subcritical ischemia, favorable outcome, questionable outcome, and unfavorable outcome. The prognosis is made using such complex indicators as confidence that the patient will develop gangrene of the lower extremity (unfavorable outcome), complex coefficient of variability, and reversibility of the ischemic process. Model accuracy was calculated using representative control samples that showed high diagnostic accuracy and specificity characterizing the quality of prediction are 0.9 and higher, which makes it possible to recommend their use in medical practice.

Differential Diagnosis of Pancreatic Cancer and Chronic Pancreatitis According to Endoscopic Ultrasonography Based on the Analysis of the Nature of the Contours of Focal Formations Based on Fuzzy Mathematical Models

One of the key echographic signs of focal pathology of the pancreas is the presence of formation contours and their nature. Endoscopic ultrasonography has a unique ability to visualize the echographic texture of the pancreatic parenchyma, and also allows you to assess in detail the boundaries and nature of the contours of the tumor formations of the organ due to the proximity of the ultrasound sensor. However, the differential diagnosis of focal pancreatic lesions remains a difficult clinical task due to the similarity of their echosemiotics. One of the ways to objectify and improve the accuracy of ultrasound data is the use of artificial intelligence methods for interpreting images. Improving the quality of differential diagnosis of focal pathology of the pancreas according to endoscopic ultrasonography based on the analysis of the nature of the contours of focal formations using fuzzy mathematical models.

Method for Determining the Body's Level of Protection According to Oxidant Status in Assessing the Influence of Industrial Risk Factors on Health

This work aims at improving the quality of health assessments, specifically under the influence of occupational risk factors. For this purpose, additional informative indicators are utilized in prognostic and diagnostic models. The models are used to characterize the level of body protection based on oxidative status. A quantitative method is proposed to assess the body's level of protection by means of the levels of lipid peroxidation and antioxidant activity, which characterize the body's oxidative status. A mechanism is developed for integrating the proposed method into prognostic and diagnostic decision rules. The developed rules are in the form of mathematical models used to synthesize hybrid fuzzy decision rules, which are then used to quantify the level of body protection (LBP) against external risk factors, based on the use of protection level functions in terms of lipid peroxidation and antioxidant activity. A mechanism for embedding LBP into predictive and diagnostic decision rules has been proposed. The proposed method is used to predict the occurrence and development of coronary heart disease in railroad locomotive drivers. It was found that to improve the predicting and diagnosing of diseases caused by external pathogenic factors, quantitative assessments of LBP, determined by oxidative status, can be implemented. It has been established that the use of the protection level indicator in predictive decision rules makes it possible to increase the efficiency of the prediction while simultaneously increasing its accuracy.

Hybrid neural networks with virtual flows in in medical risk classifiers

Modern medical risk classification systems focus on traditional risk factors and modeling methods. The available modeling tools do not allow reliable prediction of the of disease severity. In this study we develop prediction model of recurrent myocardial infarction in the rehabilitation period using several health variables generated in virtual flows. Hybrid decision modules with health data flows were used to build prognostic model for the prediction of disease. The vector of input information features consists of two subvectors: the first reflects real flows, the second reflects virtual flows. Complex interrelations among input data are modelled using Neural Network structure. The model classification quality of the intellectual cardiovascular catastrophe prediction system was tested on a sample composed of 230 patients who had acute myocardial infarction. For prediction, three categories of risk factors were identified: traditional factors, factors associated with stressful overloads, and risk factors derived from bio-impedance studies. During the rehabilitation period, the level of molecular products of lipid peroxidation and the antioxidant potential of blood serum were also studied. Experimental studies of various modifications of the proposed classifier model were conducted, consisting of sequential disconnection from the aggregator of solutions of “weak” classifiers at various hierarchical levels. The mathematical model show predictions accuracy of correct prognosis for the risk of myocardial infarction exceeding 0.86. Prediction quality indicators are higher than the known ASCORE cardiovascular catastrophe prediction system, on average, by 14%.

Biotechnical system based on fuzzy logic prediction for surgical risk classification using analysis of current-voltage characteristics of acupuncture points

OBJECTIVE

This study aimed to develop expert fuzzy logic model to assist physicians in the prediction of postoperative complications of prostatic hyperplasia before surgery.

METHODS

A method for classification of surgical risks was developed. The effect of rotation of the current-voltage characteristics at biologically active points (acupuncture points) was used for the formation of classifier descriptors. The effect determined reversible and non-reversible changes in electrical resistance at acupuncture points with periodic exposure to a sawtooth probe current. Then, the developed method was tested on the prediction of the success of surgical treatment of benign prostatic hyperplasia.

RESULTS

Input descriptors were obtained from collected data including current-voltage characteristics of 5 acupuncture points and composed of 27 arrays feeding in the model. The maximum diagnostic sensitivity of the classifier for the success of a surgical operation in the control sample was 88% and for testing data set prediction accuracy was 97%.

CONCLUSION

The use of tuples of current-voltage characteristic descriptors of acupuncture points in the classifiers could be used to predict the success of surgical treatment with satisfactory accuracy. The model can be a valuable tool to support physicians' diagnosis.

Prediction of Occupational Diseases Due to Exposure to High Radiation Electromagnetic Environment Using a Fuzzy Logic Model

Several researchers studied the health impacts of electromagnetic fields in work environment. However, the previous research focuses on the statistical analysis of past exposure. There are no studies that addressed prediction of health symptoms. Prediction and early diagnosis of occupational diseases of electric power workers with acceptable accuracy is needed. The objective of this study is to develop a data driven mathematical model for predicting and diagnosis of occupational diseases in workers in electric power industry. The complex nature of disease occurrence due to electromagnetic radiation is appropriate for the fuzzy rules set by medical experts which are analyzed and validated to produce hybrid fuzzy decision rules. The selected group of medical experts suggested using hormonal disorders, endocrine diseases, coffee abuse, chronic diseases of the internal organs, allergic diseases, cervical osteochondrosis, severe course of infectious diseases, intoxication, injury. The developed hybrid fuzzy logic model predicts high risk of developing nervous system diseases. The prediction accuracy exceeded 0.88, which is acceptable for supporting tool.