Detecting Parkinson's disease from shoe-mounted accelerometer sensors using convolutional neural networks optimized with modified metaheuristics.

Neurodegenerative conditions significantly impact patient quality of life. Many conditions do not have a cure, but with appropriate and timely treatment the advance of the disease could be diminished. However, many patients only seek a diagnosis once the condition progresses to a point at which the quality of life is significantly impacted. Effective non-invasive and readily accessible methods for early diagnosis can considerably enhance the quality of life of patients affected by neurodegenerative conditions. This work explores the potential of convolutional neural networks (CNNs) for patient gain freezing associated with Parkinson's disease. Sensor data collected from wearable gyroscopes located at the sole of the patient's shoe record walking patterns. These patterns are further analyzed using convolutional networks to accurately detect abnormal walking patterns. The suggested method is assessed on a public real-world dataset collected from parents affected by Parkinson's as well as individuals from a control group. To improve the accuracy of the classification, an altered variant of the recent crayfish optimization algorithm is introduced and compared to contemporary optimization metaheuristics. Our findings reveal that the modified algorithm (MSCHO) significantly outperforms other methods in accuracy, demonstrated by low error rates and high Cohen's Kappa, precision, sensitivity, and F1-measures across three datasets. These results suggest the potential of CNNs, combined with advanced optimization techniques, for early, non-invasive diagnosis of neurodegenerative conditions, offering a path to improve patient quality of life.

Tuning attention based long-short term memory neural networks for Parkinson's disease detection using modified metaheuristics.

Parkinson's disease (PD) is a progressively debilitating neurodegenerative disorder that primarily affects the dopaminergic system in the basal ganglia, impacting millions of individuals globally. The clinical manifestations of the disease include resting tremors, muscle rigidity, bradykinesia, and postural instability. Diagnosis relies mainly on clinical evaluation, lacking reliable diagnostic tests and being inherently imprecise and subjective. Early detection of PD is crucial for initiating treatments that, while unable to cure the chronic condition, can enhance the life quality of patients and alleviate symptoms. This study explores the potential of utilizing long-short term memory neural networks (LSTM) with attention mechanisms to detect Parkinson's disease based on dual-task walking test data. Given that the performance of networks is significantly inductance by architecture and training parameter choices, a modified version of the recently introduced crayfish optimization algorithm (COA) is proposed, specifically tailored to the requirements of this investigation. The proposed optimizer is assessed on a publicly accessible real-world clinical gait in Parkinson's disease dataset, and the results demonstrate its promise, achieving an accuracy of 87.4187 % for the best-constructed models.

Decomposition aided attention-based recurrent neural networks for multistep ahead time-series forecasting of renewable power generation.

Renewable energy plays an increasingly important role in our future. As fossil fuels become more difficult to extract and effectively process, renewables offer a solution to the ever-increasing energy demands of the world. However, the shift toward renewable energy is not without challenges. While fossil fuels offer a more reliable means of energy storage that can be converted into usable energy, renewables are more dependent on external factors used for generation. Efficient storage of renewables is more difficult often relying on batteries that have a limited number of charge cycles. A robust and efficient system for forecasting power generation from renewable sources can help alleviate some of the difficulties associated with the transition toward renewable energy. Therefore, this study proposes an attention-based recurrent neural network approach for forecasting power generated from renewable sources. To help networks make more accurate forecasts, decomposition techniques utilized applied the time series, and a modified metaheuristic is introduced to optimized hyperparameter values of the utilized networks. This approach has been tested on two real-world renewable energy datasets covering both solar and wind farms. The models generated by the introduced metaheuristics were compared with those produced by other state-of-the-art optimizers in terms of standard regression metrics and statistical analysis. Finally, the best-performing model was interpreted using SHapley Additive exPlanations.

Applying Recurrent Neural Networks for Anomaly Detection in Electrocardiogram Sensor Data.

Monitoring heart electrical activity is an effective way of detecting existing and developing conditions. This is usually performed as a non-invasive test using a network of up to 12 sensors (electrodes) on the chest and limbs to create an electrocardiogram (ECG). By visually observing these readings, experienced professionals can make accurate diagnoses and, if needed, request further testing. However, the training and experience needed to make accurate diagnoses are significant. This work explores the potential of recurrent neural networks for anomaly detection in ECG readings. Furthermore, to attain the best possible performance for these networks, training parameters, and network architectures are optimized using a modified version of the well-established particle swarm optimization algorithm. The performance of the optimized models is compared to models created by other contemporary optimizers, and the results show significant potential for real-world applications. Further analyses are carried out on the best-performing models to determine feature importance.

Performance evaluation of metaheuristics-tuned recurrent neural networks for electroencephalography anomaly detection.

Electroencephalography (EEG) serves as a diagnostic technique for measuring brain waves and brain activity. Despite its precision in capturing brain electrical activity, certain factors like environmental influences during the test can affect the objectivity and accuracy of EEG interpretations. Challenges associated with interpretation, even with advanced techniques to minimize artifact influences, can significantly impact the accurate interpretation of EEG findings. To address this issue, artificial intelligence (AI) has been utilized in this study to analyze anomalies in EEG signals for epilepsy detection. Recurrent neural networks (RNNs) are AI techniques specifically designed to handle sequential data, making them well-suited for precise time-series tasks. While AI methods, including RNNs and artificial neural networks (ANNs), hold great promise, their effectiveness heavily relies on the initial values assigned to hyperparameters, which are crucial for their performance for concrete assignment. To tune RNN performance, the selection of hyperparameters is approached as a typical optimization problem, and metaheuristic algorithms are employed to further enhance the process. The modified hybrid sine cosine algorithm has been developed and used to further improve hyperparameter optimization. To facilitate testing, publicly available real-world EEG data is utilized. A dataset is constructed using captured data from healthy and archived data from patients confirmed to be affected by epilepsy, as well as data captured during an active seizure. Two experiments have been conducted using generated dataset. In the first experiment, models were tasked with the detection of anomalous EEG activity. The second experiment required models to segment normal, anomalous activity as well as detect occurrences of seizures from EEG data. Considering the modest sample size (one second of data, 158 data points) used for classification models demonstrated decent outcomes. Obtained outcomes are compared with those generated by other cutting-edge metaheuristics and rigid statistical validation, as well as results' interpretation is performed.

Human spinal cord tissue is an underutilised resource in degenerative cervical myelopathy: findings from a systematic review of human autopsies.

STUDY DESIGN: Systematic review. BACKGROUND: Although degenerative cervical myelopathy (DCM) is the most prevalent spinal cord condition worldwide, the pathophysiology remains poorly understood. Our objective was to evaluate existing histological findings of DCM on cadaveric human spinal cord tissue and explore their consistency with animal models. METHODS: MEDLINE and Embase were systematically searched (CRD42021281462) for primary research reporting on histological findings of DCM in human cadaveric spinal cord tissue. Data was extracted using a piloted proforma. Risk of bias was assessed using Joanna Briggs Institute critical appraisal tools. Findings were compared to a systematic review of animal models (Ahkter et al. 2020 Front Neurosci 14). RESULTS: The search yielded 4127 unique records. After abstract and full-text screening, 19 were included in the final analysis, reporting on 150 autopsies (71% male) with an average age at death of 67.3 years. All findings were based on haematoxylin and eosin (H&E) staining. The most commonly reported grey matter findings included neuronal loss and cavity formation. The most commonly reported white matter finding was demyelination. Axon loss, gliosis, necrosis and Schwann cell proliferation were also reported. Findings were consistent amongst cervical spondylotic myelopathy and ossification of the posterior longitudinal ligament. Cavitation was notably more prevalent in human autopsies compared to animal models. CONCLUSION: Few human spinal cord tissue studies have been performed. Neuronal loss, demyelination and cavitation were common findings. Investigating the biological basis of DCM is a critical research priority. Human spinal cord specimen may be an underutilised but complimentary approach.

HIV-1 subtype B spread through cross-border clusters in the Balkans: a molecular analysis in view of incidence trends.

OBJECTIVES: To analyze phylogenetic relations and assess the role of cross-border clusters in the spread of HIV-1 subtype B across the Balkans, given the general trends of new HIV diagnoses in seven Balkan countries. DESIGN: Retrospective phylogenetic and trend analysis. METHODS: In-depth phylogenetic, phylodynamic and phylogeographic analysis performed on 2415 HIV-1 subtype B sequences from 1999 to 2019 using maximal likelihood and Bayesian methods. The joinpoint regression analysis of new HIV diagnoses by country and modes of transmission using 2004-2019 ECDC data. RESULTS: Ninety-three HIV-1 Subtype B transmission clusters (68% of studied sequences) were detected of which four cross-border clusters (11% of studied sequences). Phylodynamic analysis showed activity of cross-border clusters up until the mid-2000s, with a subsequent stationary growth phase. Phylogeography analyses revealed reciprocal spread patterns between Serbia, Slovenia and Montenegro and several introductions to Romania from these countries and Croatia. The joinpoint analysis revealed a reduction in new HIV diagnoses in Romania, Greece and Slovenia, whereas an increase in Serbia, Bulgaria, Croatia and Montenegro, predominantly among MSM. CONCLUSION: Differing trends of new HIV diagnoses in the Balkans mirror differences in preventive policies implemented in participating countries. Regional spread of HIV within the countries of former Yugoslavia has continued to play an important role even after country break-up, whereas the spread of subtype B through multiple introductions to Romania suggested the changing pattern of travel and migration linked to European integration of Balkan countries in the early 2000s.

Reconstructing the Temporal Origin and the Transmission Dynamics of the HIV Subtype B Epidemic in St. Petersburg, Russia.

The HIV/AIDS epidemic in Russia is among the fastest growing in the world. HIV epidemic burden is non-uniform in different Russian regions and diverse key populations. An explosive epidemic has been documented among people who inject drugs (PWID) starting from the mid-1990s, whereas presently, the majority of new infections are linked to sexual transmission. Nationwide, HIV sub-subtype A6 (previously called AFSU) predominates, with the increasing presence of other subtypes, namely subtype B and CRF063_02A. This study explores HIV subtype B sequences from St. Petersburg, collected from 2006 to 2020, in order to phylogenetically investigate and characterize transmission clusters, focusing on their evolutionary dynamics and potential for further growth, along with a socio-demographic analysis of the available metadata. In total, 54% (107/198) of analyzed subtype B sequences were found grouped in 17 clusters, with four transmission clusters with the number of sequences above 10. Using Bayesian MCMC inference, tMRCA of HIV-1 subtype B was estimated to be around 1986 (95% HPD 1984-1991), whereas the estimated temporal origin for the four large clusters was found to be more recent, between 2001 and 2005. The results of our study imply a complex pattern of the epidemic spread of HIV subtype B in St. Petersburg, Russia, still in the exponential growth phase, and in connection to the men who have sex with men (MSM) transmission, providing a useful insight needed for the design of public health priorities and interventions.

Exploring Evolutionary and Transmission Dynamics of HIV Epidemic in Serbia: Bridging Socio-Demographic With Phylogenetic Approach.

Previous molecular studies of Serbian HIV epidemic identified the dominance of subtype B and presence of clusters related HIV-1 transmission, in particular among men who have sex with men (MSM). In order to get a deeper understanding of the complexities of HIV sub-epidemics in Serbia, epidemic trends, temporal origin and phylodynamic characteristics in general population and subpopulations were analyzed by means of mathematical modeling, phylogenetic analysis and latent class analysis (LCA). Fitting of the logistic curve of trends for a cumulative annual number of new HIV cases in 1984-2016, in general population and MSM transmission group, was performed. Both datasets fitted the logistic growth model, showing the early exponential phase of the growth curve. According to the suggested model, in the year 2030, the number of newly diagnosed HIV cases in Serbia will continue to grow, in particular in the MSM transmission group. Further, a detailed phylogenetic analysis was performed on 385 sequences from the period 1997-2015. Identification of transmission clusters, estimation of population growth (Ne), of the effective reproductive number (Re) and time of the most recent common ancestor (tMRCA) were estimated employing Bayesian and maximum likelihood methods. A substantial proportion of 53% of subtype B sequences was found within transmission clusters/network. Phylodynamic analysis revealed Re over one during the whole period investigated, with the steepest slopes and a recent tMRCA for MSM transmission group subtype B clades, in line with a growing trend in the number of transmissions in years approaching the end of the study period. Contrary, heterosexual clades in both studied subtypes - B and C - showed modest growth and stagnation. LCA analysis identified five latent classes, with transmission clusters dominantly present in 2/5 classes, linked to MSM transmission living in the capital city and with the high prevalence of co-infection with HBV and/or other STIs.Presented findings imply that HIV epidemic in Serbia is still in the exponential growth phase, in particular, related to the MSM transmission, with estimated steep growth curve until 2030. The obtained results imply that an average new HIV patient in Serbia is a young man with concomitant sexually transmitted infection.