A pilot study on AI-driven approaches for classification of mental health disorders.

The increasing prevalence of mental disorders among youth worldwide is one of society's most pressing issues. The proposed methodology introduces an artificial intelligence-based approach for comprehending and analyzing the prevalence of neurological disorders. This work draws upon the analysis of the Cities Health Initiative dataset. It employs advanced machine learning and deep learning techniques, integrated with data science, statistics, optimization, and mathematical modeling, to correlate various lifestyle and environmental factors with the incidence of these mental disorders. In this work, a variety of machine learning and deep learning models with hyper-parameter tuning are utilized to forecast trends in the occurrence of mental disorders about lifestyle choices such as smoking and alcohol consumption, as well as environmental factors like air and noise pollution. Among these models, the convolutional neural network (CNN) architecture, termed as DNN1 in this paper, accurately predicts mental health occurrences relative to the population mean with a maximum accuracy of 99.79%. Among the machine learning models, the XGBoost technique yields an accuracy of 95.30%, with an area under the ROC curve of 0.9985, indicating robust training. The research also involves extracting feature importance scores for the XGBoost classifier, with Stroop test performance results attaining the highest importance score of 0.135. Attributes related to addiction, namely smoking and alcohol consumption, hold importance scores of 0.0273 and 0.0212, respectively. Statistical tests on the training models reveal that XGBoost performs best on the mean squared error and R-squared tests, achieving scores of 0.013356 and 0.946481, respectively. These statistical evaluations bolster the models' credibility and affirm the best-fit models' accuracy. The proposed research in the domains of mental health, addiction, and pollution stands to aid healthcare professionals in diagnosing and treating neurological disorders in both youth and adults promptly through the use of predictive models. Furthermore, it aims to provide valuable insights for policymakers in formulating new regulations on pollution and addiction.

AuCFSR: Authentication and Color Face Self-Recovery Using Novel 2D Hyperchaotic System and Deep Learning Models.

Color face images are often transmitted over public channels, where they are vulnerable to tampering attacks. To address this problem, the present paper introduces a novel scheme called Authentication and Color Face Self-Recovery (AuCFSR) for ensuring the authenticity of color face images and recovering the tampered areas in these images. AuCFSR uses a new two-dimensional hyperchaotic system called two-dimensional modular sine-cosine map (2D MSCM) to embed authentication and recovery data into the least significant bits of color image pixels. This produces high-quality output images with high security level. When tampered color face image is detected, AuCFSR executes two deep learning models: the CodeFormer model to enhance the visual quality of the recovered color face image and the DeOldify model to improve the colorization of this image. Experimental results demonstrate that AuCFSR outperforms recent similar schemes in tamper detection accuracy, security level, and visual quality of the recovered images.

SCovNet: A skip connection-based feature union deep learning technique with statistical approach analysis for the detection of COVID-19.

Background and Objective: The global population has been heavily impacted by the COVID-19 pandemic of coronavirus. Infections are spreading quickly around the world, and new spikes (Delta, Delta Plus, and Omicron) are still being made. The real-time reverse transcription-polymerase chain reaction (RT-PCR) is the method most often used to find viral RNA in a nasopharyngeal swab. However, these diagnostic approaches require human involvement and consume more time per prediction. Moreover, the existing conventional test mainly suffers from false negatives, so there is a chance for the virus to spread quickly. Therefore, a rapid and early diagnosis of COVID-19 patients is needed to overcome these problems. Methods: Existing approaches based on deep learning for COVID detection are suffering from unbalanced datasets, poor performance, and gradient vanishing problems. A customized skip connection-based network with a feature union approach has been developed in this work to overcome some of the issues mentioned above. Gradient information from chest X-ray (CXR) images to subsequent layers is bypassed through skip connections. In the script's title, "SCovNet" refers to a skip-connection-based feature union network for detecting COVID-19 in a short notation. The performance of the proposed model was tested with two publicly available CXR image databases, including balanced and unbalanced datasets. Results: A modified skip connection-based CNN model was suggested for a small unbalanced dataset (Kaggle) and achieved remarkable performance. In addition, the proposed model was also tested with a large GitHub database of CXR images and obtained an overall best accuracy of 98.67% with an impressive low false-negative rate of 0.0074. Conclusions: The results of the experiments show that the proposed method works better than current methods at finding early signs of COVID-19. As an additional point of interest, we must mention the innovative hierarchical classification strategy provided for this work, which considered both balanced and unbalanced datasets to get the best COVID-19 identification rate.

ECG-COVID: An end-to-end deep model based on electrocardiogram for COVID-19 detection.

The early and accurate detection of COVID-19 is vital nowadays to avoid the vast and rapid spread of this virus and ease lockdown restrictions. As a result, researchers developed methods to diagnose COVID-19. However, these methods have several limitations. Therefore, presenting new methods is essential to improve the diagnosis of COVID-19. Recently, investigation of the electrocardiogram (ECG) signals becoming an easy way to detect COVID-19 since the ECG process is non-invasive and easy to use. Therefore, we proposed in this paper a novel end-to-end deep learning model (ECG-COVID) based on ECG for COVID-19 detection. We employed several deep Convolutional Neural Networks (CNNs) on a dataset of 1109 ECG images, which is built for screening the perception of COVID-19 and cardiac patients. After that, we selected the most efficient model as our model for evaluation. The proposed model is end-to-end where the input ECG images are fed directly to the model for the final decision without using any additional stages. The proposed method achieved an average accuracy of 98.81%, Precision of 98.8%, Sensitivity of 98.8% and, F1-score of 98.81% for COVID-19 detection. As cases of corona continue to rise and hospitalizations continue again, hospitals may find our study helpful when dealing with these patients who did not get significantly worse.

Real Time Multipurpose Smart Waste Classification Model for Efficient Recycling in Smart Cities Using Multilayer Convolutional Neural Network and Perceptron.

Urbanization is a big concern for both developed and developing countries in recent years. People shift themselves and their families to urban areas for the sake of better education and a modern lifestyle. Due to rapid urbanization, cities are facing huge challenges, one of which is waste management, as the volume of waste is directly proportional to the people living in the city. The municipalities and the city administrations use the traditional wastage classification techniques which are manual, very slow, inefficient and costly. Therefore, automatic waste classification and management is essential for the cities that are being urbanized for the better recycling of waste. Better recycling of waste gives the opportunity to reduce the amount of waste sent to landfills by reducing the need to collect new raw material. In this paper, the idea of a real-time smart waste classification model is presented that uses a hybrid approach to classify waste into various classes. Two machine learning models, a multilayer perceptron and multilayer convolutional neural network (ML-CNN), are implemented. The multilayer perceptron is used to provide binary classification, i.e., metal or non-metal waste, and the CNN identifies the class of non-metal waste. A camera is placed in front of the waste conveyor belt, which takes a picture of the waste and classifies it. Upon successful classification, an automatic hand hammer is used to push the waste into the assigned labeled bucket. Experiments were carried out in a real-time environment with image segmentation. The training, testing, and validation accuracy of the purposed model was 0.99% under different training batches with different input features.

Multistage Segmentation of Prostate Cancer Tissues Using Sample Entropy Texture Analysis.

In this study, a multistage segmentation technique is proposed that identifies cancerous cells in prostate tissue samples. The benign areas of the tissue are distinguished from the cancerous regions using the texture of glands. The texture is modeled based on wavelet packet features along with sample entropy values. In a multistage segmentation process, the mean-shift algorithm is applied on the pre-processed images to perform a coarse segmentation of the tissue. Wavelet packets are employed in the second stage to obtain fine details of the structured shape of glands. Finally, the texture of the gland is modeled by the sample entropy values, which identifies epithelial regions from stroma patches. Although there are three stages of the proposed algorithm, the computation is fast as wavelet packet features and sample entropy values perform robust modeling for the required regions of interest. A comparative analysis with other state-of-the-art texture segmentation techniques is presented and dice ratios are computed for the comparison. It has been observed that our algorithm not only outperforms other techniques, but, by introducing sample entropy features, identification of cancerous regions of tissues is achieved with 90% classification accuracy, which shows the robustness of the proposed algorithm.

Pre-treatment growth and IGF-I deficiency as main predictors of response to growth hormone therapy in neural models.

Mathematical models have been applied in prediction of growth hormone treatment effectiveness in children since the end of 1990s. Usually they were multiple linear regression models; however, there are also examples derived by empirical non-linear methods. Proposed solution consists in application of machine learning technique - artificial neural networks - to analyse this problem. This new methodology, contrary to previous ones, allows detection of both linear and non-linear dependencies without assuming their character a priori The aims of this work included: development of models predicting separately growth during 1st year of treatment and final height as well as identification of important predictors and in-depth analysis of their influence on treatment's effectiveness. The models were derived on the basis of clinical data of 272 patients treated for at least 1 year, 133 of whom have already attained final height. Starting from models containing 17 and 20 potential predictors, respectively for 1st year and final height model, we were able to reduce their number to 9 and 10. Basing on the final models, IGF-I concentration and earlier growth were indicated as belonging to most important predictors of response to GH therapy, while results of GH secretion tests were automatically excluded as insignificant. Moreover, majority of the dependencies were observed to be non-linear, thus using neural networks seems to be reasonable approach despite it being more complex than previously applied methods.

Pitfalls in anorexia nervosa research: The risk of artifacts linked to denial of illness and methods of preventing them.

One of the most serious problems faced by researchers studying eating disorders is denial of illness in individuals with anorexia nervosa. Importantly, the term "denial" not only has different meanings, but in the case of anorexia nervosa its very nature still remains obscure. It is not even known whether it is deliberate or unintentional. Denial of illness in anorexic patients has serious consequences for evaluation of the reliability of information obtained from those individuals. Indeed, the most frequently used screening questionnaires, such as the Eating Attitudes Test (EAT) (Garner & Garfinkel 1979) and the Eating Disorder Inventory (EDI) (Garner et al. 1983), may not reflect the psychological state of the subjects due to distorted responses. The objective of this review article is to elucidate, at least in part, the nature of denial of illness in anorexic individuals and, importantly, to present methods for direct or indirect measurement of this variable. The authors emphasize the detrimental effect of denial of illness on the quality of information obtained from the patients and the notorious unreliability of self-report data. The final part of the paper contains suggestions as to methods of bypassing the pitfalls associated with the influence of denial of illness on the results of studies involving anorexic individuals; for instance, it is recommended that one should build an honest and trustful relationship with the patient. Last but not least, the focus is placed on the potential of experimental psychology, which offers tools producing robust data, resistant to deliberate distortion by patients.

Colour-Based Binary Discrimination of Scarified Quercus robur Acorns under Varying Illumination.

Efforts to predict the germination ability of acorns using their shape, length, diameter and density are reported in the literature. These methods, however, are not efficient enough. As such, a visual assessment of the viability of seeds based on the appearance of cross-sections of seeds following their scarification is used. This procedure is more robust but demands significant effort from experienced employees over a short period of time. In this article an automated method of acorn scarification and assessment has been announced. This type of automation requires the specific setup of a machine vision system and application of image processing algorithms for evaluation of sections of seeds in order to predict their viability. In the stage of the analysis of pathological changes, it is important to point out image features that enable efficient classification of seeds in respect of viability. The article shows the results of the binary separation of seeds into two fractions (healthy or spoiled) using average components of regular red-green-blue and perception-based hue-saturation-value colour space. Analysis of accuracy of discrimination was performed on sections of 400 scarified acorns acquired using two various setups: machine vision camera under uncontrolled varying illumination and commodity high-resolution camera under controlled illumination. The accuracy of automatic classification has been compared with predictions completed by experienced professionals. It has been shown that both automatic and manual methods reach an accuracy level of 84%, assuming that the images of the sections are properly normalised. The achieved recognition ratio was higher when referenced to predictions provided by professionals. Results of discrimination by means of Bayes classifier have been also presented as a reference.

Telerehabilitation approach for patients with hand impairment.

PURPOSE: Telerehabilitation is one of the newest branches of telemedicine which has been developed because patients need regular trainings outside the medical institution but still under specialist supervision. It helps maintain regularity of exercises and reduces costs. The professional and advanced systems for telerehabilitation are presented in papers, however, there is still lack of development of minor systems which provide therapeutic values and are more accessible to people. Therefore we focus on a solution for hand telerehabilitation of poststroke patients, based solely on a personal computer and camera. METHODS: We focused on the manipulative hand (fingers, metacarpus, wrist) movements trainings for patients with cerebral palsy. The contact between patient and physiotherapist is provided by using web cameras and web service. Additionally, the camera can be used to monitor the effectiveness of performed exercises. Computer vision system keeps track of the patient's hand movement. The digital image processing is used to detect if the patient performs exercises correctly. RESULTS: We created web service and software application TeleReh that provides therapeutic values for the hand impaired people. The system created was evaluated by three physiotherapists, one doctor and a cerebral palsy patient. CONCLUSIONS: Our solution applies to all patients who have undergone basic rehabilitation in hospital and need to continue hand rehabilitation at home. The main advantages are: easy adaptation to the individual needs and abilities, monitoring the progress by using automatically generated reports after each training session. It is worth noticing that discussion between IT specialists, rehabilitants and patients was necessary to achieve good results.

[Vascular System Infections: Characteristics, Risk Factors, Prevention Methods and Economic Impact].

One of the main problems of modern medicine are infections. They can be divided into local and general (depending on infected tissues and/or organs) or hospital and community-acquired infections (depending on the location and source of infection). The occurrence of infection reduces the ability for quick recovery, and in case of complications the ability to continue professional activity. Bacteria can be present in the vascular system causing vein, artery, capillary infection or blood infection (bacteremia). The vascular system infection can be connected with medical procedures, type and chemical composition of used medical devices or biomaterials. The usage of central or peripheral venous catheters can increase the risk factor of vascular system infection. The main risk factors of hospital infection are: patient's condition, surgical procedure and hospital aseptic procedures. Improving the current state of knowledge of medical personnel and implementation of well-designed prevention procedures can contribute to reducing hospital infection factors. The technical quality of used medical devices (e.g. anti-bacterial coat on vascular prostheses) can also reduce the risk of infection. Raising awareness and educating the patient (e.g. with infected trophic ulcers) can be an important element of control and prevention of nosocomial and communityacquired infections. Medical literature containing procedures and descriptions of specific medical cases related to development process of infections was analysed. The literature confirms the significant magnitude of the problems associated with the vascular system infections.

Neural network models - a novel tool for predicting the efficacy of growth hormone (GH) therapy in children with short stature.

INTRODUCTION: The leading method for prediction of growth hormone (GH) therapy effectiveness are multiple linear regression (MLR) models. Best of our knowledge, we are the first to apply artificial neural networks (ANN) to solve this problem. For ANN there is no necessity to assume the functions linking independent and dependent variables. The aim of study is to compare ANN and MLR models of GH therapy effectiveness. MATERIAL AND METHODS: Analysis comprised the data of 245 GH-deficient children (170 boys) treated with GH up to final height (FH). Independent variables included: patients' height, pre-treatment height velocity, chronological age, bone age, gender, pubertal status, parental heights, GH peak in 2 stimulation tests, IGF-I concentration. The output variable was FH. RESULTS: For testing dataset, MLR model predicted FH SDS with average error (RMSE) 0.64 SD, explaining 34.3% of its variability; ANN model derived on the same pre-processed data predicted FH SDS with RMSE 0.60 SD, explaining 42.0% of its variability; ANN model derived on raw data predicted FH with RMSE 3.9 cm (0.63 SD), explaining 78.7% of its variability. CONCLUSION: ANN seem to be valuable tool in prediction of GH treatment effectiveness, especially since they can be applied to raw clinical data.

Artificial neural network modelling of the results of tympanoplasty in chronic suppurative otitis media patients.

The application of computer modelling for medical purposes, although challenging, is a promising pathway for further development in the medical sciences. We present predictive neural and k-nearest neighbour (k-NN) models for hearing improvements after middle ear surgery for chronic otitis media. The studied data set comprised 150 patients characterised by the set of input variables: age, gender, preoperative audiometric results, ear pathology and details of the surgical procedure. The predicted (output) variable was the postoperative hearing threshold. The best neural models developed in this study achieved 84% correct predictions for the test data set while the k-NN model produced only 75.8% correct predictions.

[How can an otolaryngologist benefit from artificial neural networks?]. / Co moze uzyskac otolaryngolog, stosujac sztuczne sieci neuronowe?

Artificial neural networks are informatic systems that have unique computational capabilities. The principle of their functioning is based on the rules of data processing in the brain. This article discusses the most important features of the artificial neural networks with reference to their applications in otolaryngology. The cited studies concern the fields of rhinology, audiology, phoniatrics, vestibulology, oncology, sleep apnea and salivary gland diseases. The authors also refer to their own experience with predictive neural models designed in the Department of Otolaryngology of the Jagiellonian University Medical College in Krakow. The applications of artificial neural networks in clinical diagnosis, automated signal interpretation and outcome prediction are presented. Moreover, the article explains how the artificial neural networks work and how the otolaryngologists can use them in their clinical practice and research.

Obstetrical prevention of human cancers. Review article.

Cancer appears in the new form of cell life as a direct consequence of self-organising pre-cancer cells (dissipathogenic systems), whose further existence is disabled by extreme impairment of their metabolism. Life itself is the highest value, since instead of necrosis or apoptosis, the cellular system in an unfavourable environment can change its genetic identity provided that the improvement in its own metabolism leads to increased chaos by higher dissipation of matter and energy in its environment. Prolongation of human life has resulted by a longer period of old age which is favourable for self-organisation of dissipative neoplastic cells. Modern medicine has explained the relation of the cervical cancer to preterm births and to inadequate use of oral contraceptives and/or replacement, instead of supplementary, neurohormonal therapy. Therefore, as early as in the period of pregnancy reproductive cells should be protected due to their prime importance in the intergenerational passage of life. Disturbance of systemic autoregulation causes development of dissipathogenic state of cells. A single zygote, whose environment is also important for the future development of the next two generations that are initiated with its formation, defines the unique identity of each person, whose life is determined by free will and neoplasms.

[Applicability of artificial neural networks in postoperative hearing improvement prognosis in patients with chronic otitis media]. / Przydatnosc sztucznych sieci neuronowych w prognozowaniu pooperacyjnej poprawy sluchu u chorych z przewleklym zapaleniem ucha srodkowego.

The primary objective of surgical treatment of chronic otitis media is to remove the pathological changes and restore the biology of the middle ear. Subsequently, ossicular chain reconstruction is performed in order to restore hearing. The aim of the study was to evaluate the efficacy of mathematical models (artificial neural networks) in postoperative hearing improvement prognosis. The models based on preoperative anamnesis and data concerning the operative procedure. The analyzed group comprised of 135 patients operated for chronic otitis media in the Department of Otolaryngology CM UJ in Krakow. The measure of postoperative hearing improvement was cochlear reserve (air-bone gap) closure depending on the type of ossicular chain reconstruction. The artificial neural networks provided 100% correct predictions basing on preoperative clinical examination, pathological changes found in the middle ear and description of the surgical procedure. The study proves that artificial neural networks are effective in functional result prognosis in tympanoplastic surgery.

Cognitive techniques in medical information systems.

This paper presents the way of application of structural methods of Artificial Intelligence-in particular, linguistic mechanisms of semantic meaning reasoning-for development of intelligent medical information systems. They also facilitate an in-depth analysis of the meaning presented in Diagnosis Support Information Systems used in analysis of selected medical examinations. This paper will present the mechanisms of pattern meaning description on selected examples of spinal cord image analysis. Presented approach for semantic reasoning will be based on the model of cognitive resonance, which will be applied to the task of interpreting the meaning of selected diagnostic images from the central nervous system. Such algorithms are aimed to construct an intelligent analysis module in medical IS. The application presented in this paper is of a research character and it serves the preparation of efficient lesion detection methods applied to a data set originating from magnetic and resonance examinations of the spine and spinal cord structures.

Application of artificial neural networks and DFT-based parameters for prediction of reaction kinetics of ethylbenzene dehydrogenase.

Artificial neural networks (ANNs) are used for classification and prediction of enzymatic activity of ethylbenzene dehydrogenase from EbN1 Azoarcus sp. bacterium. Ethylbenzene dehydrogenase (EBDH) catalyzes stereo-specific oxidation of ethylbenzene and its derivates to alcohols, which find its application as building blocks in pharmaceutical industry. ANN systems are trained based on theoretical variables derived from Density Functional Theory (DFT) modeling, topological descriptors, and kinetic parameters measured with developed spectrophotometric assay. Obtained models exhibit high degree of accuracy (100% of correct classifications, correlation between predicted and experimental values of reaction rates on the 0.97 level). The applicability of ANNs is demonstrated as useful tool for the prediction of biochemical enzyme activity of new substrates basing only on quantum chemical calculations and simple structural characteristics. Multi Linear Regression and Molecular Field Analysis (MFA) are used in order to compare robustness of ANN and both classical and 3D-quantitative structure-activity relationship (QSAR) approaches.

Modeling of ECG interpretation methods sharing based on human experts relations.

The introduction of distributed ECG interpretation is currently limited by the computational capacity of a battery-operated wearable recorder. The relations between human experts in cardiology are generalized in this paper and considered as a pattern of co-operation between computerized interpretation centers. Studying the cardiologist's practice we found out that during the years the range of family doctors skills, their adaptations to the patient-specific needs and their communication with the regional experts were optimized. Extrapolation of this approach to the artificial networks assumes that only selected and commutable basic interpretation routines are implemented in a remote recorder limiting its costs, increasing the autonomy time and adaptability. While most of cases are resolved on local level, the complicated but not very frequent events are reported directly to the interpretive center, for automated or human-assisted interpretation. Such architecture, derived from natural task sharing, is believed to fulfil all the diagnostic requirements with minimum involvement of complicated equipment and at minimum costs of data transmission.