Self-organizing Map Research Articles

Determining the onset of summer rainfall over Vietnam using self-organizing maps

Determining the onset of summer rainfall over Vietnam using self-organizing maps

Revealing the distribution characteristics and key driving factors of dissolved organic matter in Baiyangdian Lake inflow rivers from different seasons and sources

The river course is a transitional area connecting the source and receiving water bodies. The dissolved organic matter (DOM) in the river course is an important factor affecting the aquatic environment and ecological health. However, there are shortcomings in studying the differences and quantitative contributions of river DOM in different seasons and sources. In this study, ultraviolet-visible (UV–vis) and three-dimensional fluorescence spectra were used to characterize the optical properties, analyze the spatiotemporal changes, and establish the quantitative relationship between environmental factors and DOM in the inflow rivers of Baiyangdian Lake. The results showed that the relative DOM concentrations in summer and autumn were significantly higher than those in the other seasons (P < 0.001) and that the DOM source (SR < 1) was mainly exogenous. The fluorescence abundance of protein-like substances (C1 + C2 + C3) was the highest in spring, whereas that of humus C4 was the highest in autumn. Moreover, the inflow rivers exhibited strong autogenetic characteristics (BIX > 1) throughout the year. Self-organizing maps (SOM) indicated that the main driving factors of water quality were NO3−-N in spring, autumn, and winter and DO, pH, and chemical oxygen demand (COD) in summer. Random forest analysis showed that the fluorescent components (C1–C4) were closely related to the migration and transformation of nitrogen, and pH and nitrogen were the main predictors of each component. The Mantel test and structural equation model (SEM) showed that temperature and NO3−-N significantly influenced the DOM concentration, components, and molecular properties in different seasons. Moreover, the river source also affected the distribution mechanism of DOM in the water body. Our study comprehensively analyzed the response of DOM in inflow rivers in different seasons and water sources, providing a basis for further understanding the driving mechanisms of water quality.

Spatial-temporal pattern of ecosystem services and sustainable development in representative mountainous cities: A case study of Chengdu-Chongqing Urban Agglomeration

The Sustainable Development Goals (SDGs) are essential measure for preserving the balance between human well-being and natural ecosystems. The benefit of preserving ecosystems health play a crucial role in promoting the SDGs by providing stable ecosystem services (ESs). However, the ecological health of mountainous cities is vulnerable, with relative low ecological resilience. To investigate the conflict between ecosystems and sustainable development, this study takes the Chengdu-Chongqing Urban Agglomeration as the study area. The major tasks and results in this study include: (1) using the entropy weighting method and the InVEST model, we combined remote sensing, geographic, and statistical data to quantify three types of SDGs (economic, social, environmental) and four ESs (water yield, soil conservation, habitat quality, carbon storage), and establish a localized sustainable development assessment framework that is applicable to the Chengdu-Chongqing Urban Agglomeration. The results show that from 2014 to 2020, the three types of SDGs exhibited an overall upward trend, with the lowest values occurring in 2016. The gap between different counties has narrowed, but significant regional differences still remain, indicating an unbalanced development status quo. Among the 142 counties, water yield and soil conservation values show a consistent downward trend but occupies significant interannual variations, while habitat quality and carbon storage values increases consistently each year. (2) using Spearman's nonparametric correlation analysis and multiscale geographically weighted regression model to explore the temporal variation and spatial heterogeneity of correlations between county ESs and SDGs. The results showed significant heterogeneity in the spatial trade-offs and synergies between ESs and SDGs, with two pairs of synergies weakening, seven pairs of trade-offs increasing, and the strongest negative correlation between Economic Sustainable Development Goals and habitat quality. (3) we applied the self-organizing mapping neural networks to analyze the spatial clustering characteristics of ESs-SDGs. Based on the spatial clustering effects, we divides the Chengdu-Chongqing Urban Agglomeration into four zones, and different zones have different levels of ESs and SDGs. The targeted strategies should be adopted according to local conditions. This work is of great practical importance in maintaining the stability and sustainable development of the Chengdu-Chongqing Urban Agglomeration ecosystem and provides a scientific reference for the optimal regulation of mountainous cities.

Precise management and control around the landfill integrating artificial intelligence and groundwater pollution risks

The Landfill plays an important role in urban development and waste disposal. However, landfill leachate may also bring more serious pollution and health risks to the surrounding groundwater environment. Compared with other areas, the area around the landfill needs more precise management. To solve this problem, based on the “pressure-state-response” framework, a method for the identification and evaluation of groundwater pollution around the landfill was constructed. The LPI method was used to assess the contamination potential of the leachate. The comprehensive quality of groundwater was evaluated by the entropy-AHP water quality assessment method, sodium adsorption ratio and sodium percentage. The probabilistic health risks of groundwater were assessed based on a Monte Carlo algorithm. The sources of pollutants were identified by comprehensively using the PCA-APCS-MLR model and the PMF model. Finally, the self-organizing map algorithm and the Kmeans algorithm were integrated to enhance the precision of groundwater management and control measures. The results showed that the leachate of the landfill was in the mature stage, and the concentration of inorganic substances was relatively high. Leachate had the potential to contaminate surrounding groundwater. The groundwater quality of 68.14% of the study area was in the poor or lower level. The groundwater near the landfill was unsuitable not only for drinking but also for irrigation purposes. Cl− was the main non-carcinogenic risk factor. Reducing pollutant concentration and controlling exposure time are effective strategies for mitigating health risks caused by high-concentration pollutants (Cl−, NO3−) and low-concentration pollutants (F−), respectively. The groundwater around the landfill was jointly affected by six pollution sources. The PMF model has better analytical ability in mixed pollution areas. The groundwater in the study area was divided into five clusters, of which cluster Ⅰ was significantly affected by leachate, and cluster Ⅴ had the lowest pollution and health risk.

Research on low-power driving fatigue monitoring method based on spiking neural network.

Fatigue driving is one of the leading causes of traffic accidents, and the rapid and accurate detection of driver fatigue is of paramount importance for enhancing road safety. However, the application of deep learning models in fatigue driving detection has long been constrained by high computational costs and power consumption. To address this issue, this study proposes an approach that combines Self-Organizing Map (SOM) and Spiking Neural Networks (SNN) to develop a low-power model capable of accurately recognizing the driver's mental state. Initially, spatial features are extracted from electroencephalogram (EEG) signals using the SOM network. Subsequently, the extracted weight vectors are encoded and fed into the SNN for fatigue driving classification. The research results demonstrate that the proposed method effectively considers the spatiotemporal characteristics of EEG signals, achieving efficient fatigue detection. Simultaneously, this approach successfully reduces the model's power consumption. When compared to traditional artificial neural networks, our method reduces energy consumption by approximately 12.21-42.59%.

Stratification of Heart Sounds Morphology Through Unsupervised Learning.

The use of heart sounds for the assessment of the hemodynamic condition of the heart in telemonitoring applications is object of wide research at date. Many different approaches have been tried out for the analysis of the first (S1) and second (S2) heart sounds, but their morphological interpretation is still to be explored: in fact, the sound morphology is not unique and this impact the separability of the heart sounds components with methods based on envelopes or model optimization. In this study, we propose a method to stratify S1 and S2 according to their morphology to explore their diversity and increase their morphological interpretability. The method we propose is based on unsupervised learning, which we obtain using the cascade of four Self-Organizing Maps (SOMs) of decreasing dimensions. When tested on a publicly available heart sounds dataset, the proposed clustering approach proved to be robust and consistent, with over 80% of the heartbeats of the same patient being clustered together. The identified heart sounds templates highlight differences in the time and energy domains which may open to new directions of analysis in the future.

Self-organizing surrogate-assisted non-dominated sorting differential evolution

Multi-objective optimization problems (MOPs) involve optimizing multiple conflicting objectives simultaneously, resulting in a set of Pareto optimal solutions. Due to the high computational or financial cost associated with evaluating fitness, expensive multi-objective optimization problems (EMOPs) further complicate the optimization process. Surrogate-assisted evolutionary algorithms (SAEAs) have emerged as a promising approach to address EMOPs by substituting costly evaluations with computationally efficient surrogate models. This paper introduces the self-organizing surrogate-assisted non-dominated sorting differential evolution (SSDE), which uses surrogate model based on a self-organizing map (SOM) to approximate the fitness function. SSDE offers advantages such as reduced computational cost, improved accuracy, and the speed of enhanced convergence. The SOM-based surrogate models effectively capture the underlying structure of the Pareto optimal set and Pareto optimal front, leading to superior approximations of the fitness function. Experimental results on benchmark functions and real-world problems, including Model-Free Adaptive Control (MFAC) and the Yagi-Uda Antenna design, demonstrate the competitiveness and efficiency of SSDE compared to other algorithms.

HybGBS: A hybrid neural network and grey wolf optimizer for intrusion detection in a cloud computing environment

SummaryThe cloud computing environment is subject to unprecedented cyber‐attacks as its infrastructure and protocols may contain vulnerabilities and bugs. Among these, Distributed Denial of Service (DDoS) is chosen by most cyber extortionists, creating unusual traffic that drains cloud resources, making them inaccessible to customers and end users. Hence, security solutions to combat this attack are in high demand. The existing DDoS detection techniques in literature have many drawbacks, such as overfitting, delay in detection, low detection accuracy for attacks that target multiple victims, and high False Positive Rate (FPR). In this proposed study, an Artificial Neural Network (ANN) based hybrid GBS (Grey Wolf Optimizer (GWO) + Back Propagation Network (BPN) + Self Organizing Map (SOM)) Intrusion Detection System (IDS) is proposed for intrusion detection in the cloud computing environment. The base classifier, BPN, was chosen for our research after evaluating the performance of a comprehensive set of neural network algorithms on the standard benchmark UNSW‐NS 15 dataset. BPN intrusion detection performance is further enhanced by combining it with SOM and GWO. Hybrid Feature Selection (FS) is made using a correlation‐based approach and Stratified 10‐fold cross‐validation (STCV) ranking based on Weight matrix value (W). These selected features are further fine‐tuned using metaheuristic GWO hyperparameter tuning based on a fitness function. The proposed IDS technique is validated using the standard benchmark UNSW‐NS 15 dataset, which consists of 1,75,341 and 82,332 attack cases in the training and testing datasets. This study's findings demonstrate that the proposed ANN‐based hybrid GBS IDS model outperforms other existing IDS models with a higher intrusion detection accuracy of 99.40%, fewer false alarms (0.00389), less error rate (0.001), and faster prediction time (0.29 ns).

Segmentation of beach plastic fragments’ contours based on self-organizing map and multi-shape descriptors: A rapid indication of fragmentation and wearing types

Environmental plastic fragments have been verified as byproducts of large plastic and its secondary pollutants including micro and nanoplastics. There are few quantitative studies available, but their contours have values for the weathering mechanisms. We used geometric descriptors, fractal dimensions, and Fourier descriptors to characterize field and artificial polyethylene and polypropylene samples as a means of investigating the contour characteristics. It provides a methodological framework for contour classification. Unsupervised classification was performed using self-organizing neural networks with size-invariance parameters. We revealed the isometric phenomenon of plastic fragments during fragmentation, i.e., that the degree of contour rounding and complexity increase and decrease, respectively, with decreasing fragment size. With an average error rate of 8.9 %, we can distinguish artificial samples from field samples. It was also validated by the difference in Carbonyl Index between groups. We propose a two-stage process for plastic fragmentation and give three types of contour features which were key in the description of fragmented contours, i.e., size, complexity, and rounding. Our work will improve the accuracy of characterizations regarding the weathering and fragmentation processes of certain kinds of plastic fragments. The contour parameters also have the potential to be applied in more realistic scenarios and varied polymers.

A novel coaxial air-R134a spray cooling for heat transfer enhancement of laser dermatology

Efficient cooling is essential in laser dermatology to prevent thermal damage, but current methods are often hindered by thermal barrier film deposition. This study presents coaxial air-R134a spray cooling for heat transfer enhancement and deposited-film mitigation. Surface temperature was measured using a TFTC thermocouple, while heat flux was estimated through the improved Duhamel theorem. High-speed camera visualization with the Mie-scattering technique captured the spray and film behavior. The results showed uniform and effective cooling via transient surface heat transfer and film behaviors. The boiling curve identified three distinct cooling phases: transition boiling, nucleate boiling, and single-phase film deposition. Airflow adjustments not only enhance the heat transfer coefficient but also control spray dynamics, droplet rebounding, and film spreading. Self-organizing maps revealed that medium-to-high nozzle spacing (y/D) reduces surface temperature and increases the heat transfer coefficient. Film resistance time is minimized by either high airflow and low y/D or low airflow and high y/D. For critical heat flux, low y/D and airflow are optimal. A y/D = 32 and 60 L/min airflow provided appropriate operating conditions. However, airflow increased frost deposition during off-duty, suggesting moisture-free air in the future.

Increase in porosity and permeability resolution for thin-bedded Miocene formation in Carpathian Foredeep using different clustering methods

AbstractThe accurate interpretation of well-logging data is a crucial stage in the exploration of gas- and oil-bearing reservoirs. Geological formations, such as the Miocene deposits, present many challenges related to thin layers, whose thickness is often less than the measurement resolution. This research emphasizes the potential of utilizing electrofacies in such challenging environments. The application of electrofacies not only allows for the grouping of intervals with similar physical characteristics but can also be useful for estimating porosity and permeability parameters. For this purpose, various clustering methods were tested, including the 2D indexed and probabilized self-organizing map (IPSOM) method with and without supervision. Subsequently, the usefulness of the obtained results to improve the estimation of porosity and permeability parameters with the help of artificial neural networks was verified. As a result of the conducted analyses, significantly better results were obtained compared to classical petrophysical interpretation. The calculated porosity and permeability parameters were characterized by much greater variability and alignment with laboratory measurements on porosity and permeability. The best results were obtained for the IPSOM method, but the other methods did not differ significantly. In conclusion, the studies have shown a positive result of applying clustering methods, including the IPSOM method, to improve the estimation of permeability and porosity parameters in complicated, thinly-layered formations.

Ultra-short-term photovoltaic power prediction based on similar day clustering and temporal convolutional network with bidirectional long short-term memory model: A case study using DKASC data

Due to its strong dependence on weather conditions, photovoltaic (PV) power is highly intermittent in nature. In light of this, this paper introduces a PV power prediction model based on similar-day clustering and temporal convolutional network (TCN) with bidirectional long short-term memory (BiLSTM) model. Firstly, in the data preprocessing stage, improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) method is used to decompose and denoise the original PV power data to obtain subsequences of different frequencies. Subsequently, the sample entropy (SE) method is incorporated to reconstruct the subsequences and generate low-frequency and high-frequency subsequences with more pronounced temporal features. In addition, a self-organizing map (SOM) with the K-means algorithm is used to categorize historical days into four weather types: sunny, cloudy, rainy, and intermittent. Through grey relational analysis (GRA), meteorological factors significantly affecting PV power prediction are identified to construct a multidimensional feature set. During the model prediction phase, historical PV power data and related meteorological impact factors are input into the TCN-BiLSTM model to perform multi-data-driven PV power prediction. Finally, the predicted results of each subsequence are linearly combined to obtain the ultimate PV power prediction. The proposed model is compared with convolutional neural network (CNN)-LSTM, LSTM, and TCN models, achieving a reduction in the root mean square error (RMSE) of 0.129 kW, 0.238 kW, and 0.257 kW, respectively. This demonstrates that the proposed model exhibits superior overall performance in time series modeling and information capture, enhancing the understanding of seasonal, periodic, and irregular patterns in PV power generation data.

Modeling carbon dioxide emissions reduction

The paper proposes a new scientific approach to modeling carbon dioxide (CO2) emissions in individual countries, based on the tools of Kohonen self-organizing maps and the provisions of economic theory. This work is based on conducted a comprehensive analysis of wide set of indicators influencing carbon dioxide emissions and decarbonization processes in 40 selected countries for 10 years time span from 2013 till 2022. During clustering these nations using self-organizing maps, 14 key indicators were chosen that cover economic and demographic growth, energy consumption, CO2 emission data, and includes trade of energy recourses for wider analysis and identification of countries with similar decarbonization potential, economic development and trade energy recourses possibilities. This approach exposed distinct clusters of nations with varying decarbonization capabilities and tracked the progress of analyzed countries in terms of CO2 emission changes in dynamic.The analysis of clustering was focused on identification leaders and followers in decarbonization activities. The leaders cluster includes Sweden, Norway, New Zealand and other countries exhibited leadership in developing a climate-resilient economy among 40 countries. Hence, all countries positioned closer to the leader cluster on the map demonstrated higher efficiency in their decarbonization pathways. Consequently, these properties of Kohonen maps provide a foundation for formulating general recommendations to achieve an efficient and effective low-carbon economy for followers including Ukraine, Kazakhstan and other energy intensive countries.Distinct countries decarbonization profiles resulting from clustering were used to develop practical recommendations towards estimating and establishing targeted CO2 emission levels for analyzed countries, conducting continued promoting renewable energy utilization, enhancing cross-border energy trade, diversifying energy portfolios, etc. Also, the proposed methodology can be utilized to project future emissions trends for each country based on cluster-specific models and facilitate policy decisions making process related to mitigating carbon emissions.

Edge human activity recognition using federated learning on constrained devices

Human Activity Recognition (HAR) using wearable Internet of Things (IoT) devices represents a well investigated researched field encompassing various application domains. Many current approaches rely on cloud-based methodologies for gathering data from diverse users, resulting in the creation of extensive training datasets. Although this strategy facilitates the application of powerful Machine Learning (ML) techniques, it raises significant privacy concerns, which can become particularly severe given the sensitivity of HAR data. Moreover, the labeling process can be extremely time-consuming and even more challenging for IoT wearable devices due to the absence of efficient input systems. In this paper, we address both aforementioned challenges by designing, implementing, and validating edge-based Human Activity Recognition (HAR) systems that operate on resource-constrained IoT devices, which relies on the utilization of Self-Organizing Maps (SOM) for activity detection. We incorporate a feature selection process before training to reduce data dimensionality and, consequently, the SOM size, aligning with the resource limitations of wearable IoT devices. Additionally, we explore the application of Federated Learning (FL) techniques for HAR tasks, enabling new users to leverage SOM models trained by others on their respective datasets. Our federated Extreme Edge (EE)-aware HAR system is implemented on a wearable IoT device and rigorously tested against state-of-the-art and experimental datasets. The results demonstrate that our C++-based SOM implementation achieves a consistent reduction in model size compared to state-of-the-art approaches. Furthermore, our findings highlight the effectiveness of the FL-based approach in overcoming personalized training challenges, particularly in onboarding scenarios.

Analysis on Spatial Variability of Hydrochemical Characteristics and Control Factors of Jinan Baotu Spring Area

To understand the karst groundwater hydrochemical characteristics and ion sources of the Jinan Baotu Spring area, this study focused on the three functional zones as an indirect recharge area, direct recharge area, and discharge area. Through water sample collection and testing, the spatial variability of groundwater chemical characteristics in different functional zones and the formation mechanism were analyzed using hydrochemistry parameter statistics, multivariate statistics, self-organizing map, hydrochemistry graphical analysis, ion ratios, and other methods, guided by the theory of groundwater flow system and combined with regional physical geography and hydrogeological conditions. The results showed that： the groundwater of each functional zone was alkaline as is typical in the dissolution of carbonate minerals. Owing to the different groundwater runoff pathways, the variability of water chemistry parameters in different functional areas was obvious. The groundwater of the discharge area was recharged by both the direct recharge area and the indirect recharge area. The hydrochemistry type changed from HCO3-Ca type to HCO3·SO4-Ca type through the indirect recharge area to the discharge area. The presence of a small amount of gypsum dissolution within the aquifer generated Ca2+ and SO42-. The ions in groundwater mainly came from the dissolution and filtration of aquifer rock minerals, and at the same time, they were affected by cation exchange, mineral dissolution equilibrium, and the combined effects of human activities. The groundwater in the Baotu Spring area was greatly influenced by human activities, which to some extent affected the evolution of groundwater hydrochemistry in the spring area.

Retraction Note: Simple hemogram to support the decision-making of COVID-19 diagnosis using clusters analysis with self-organizing maps neural network

Retraction Note: Simple hemogram to support the decision-making of COVID-19 diagnosis using clusters analysis with self-organizing maps neural network

Optimized Radio Frequency Footprint Identification Based on UAV Telemetry Radios.

With the widespread use of unmanned aerial vehicles (UAVs), the detection and identification of UAVs is a vital security issue for the safety of airspace and ground facilities in the no-fly zone. Telemetry radios are important wireless communication devices for UAVs, especially in UAVs beyond the visual line of sight (BVLOS) operating mode. This work focuses on the UAV identification approach using transient signals from UAV telemetry radios instead of the signals from UAV controllers that the former research work depended on. In our novel UAV Radio Frequency (RF) identification system framework based on telemetry radio signals, the EC-α algorithm is optimized to detect the starting point of the UAV transient signal and the detection accuracy at different signal-to-noise ratios (SNR) is evaluated. In the training stage, the Convolutional Neural Network (CNN) model is trained to extract features from raw I/Q data of the transient signals with different waveforms. Its architecture and hyperparameters are analyzed and optimized. In the identification stage, the extracted transient signals are clustered through the Self-Organizing Map (SOM) algorithm and the Clustering Signals Joint Identification (CSJI) algorithm is proposed to improve the accuracy of RF fingerprint identification. To evaluate the performance of our proposed approach, we design a testbed, including two UAVs as the flight platform, a Universal Software Radio Peripheral (USRP) as the receiver, and 20 telemetry radios with the same model as targets for identification. Indoor test results show that the optimized identification approach achieves an average accuracy of 92.3% at 30 dB. In comparison, the identification accuracy of SVM and KNN is 69.7% and 74.5%, respectively, at the same SNR condition. Extensive experiments are conducted outdoors to demonstrate the feasibility of this approach.

Ensemble of machine learning techniques to predict survival in kidney transplant recipients

Kidney transplant recipients face a high cardiovascular risk, which is a leading cause of death in this patient group. This article proposes the application of clustering techniques and feature selection to predict the survival outcomes of kidney transplant recipients based on machine learning techniques and mainstream statistical methods. First, feature selection techniques (Boruta, Random Survival Forest and Elastic Net) are used to detect the most relevant variables. Subsequently, each set of variables obtained by each feature selection technique is used as input for the clustering algorithms used (Consensus Clustering, Self-Organizing Map and Agglomerative Clustering) to determine which combination of feature selection, clustering algorithm and number of clusters maximizes intercluster variability. Next, the mechanism called False Clustering Discovery Reduction is applied to obtain the minimum number of statistically differentiable populations after applying a control metric. This metric is based on a variance test to confirm that reducing the number of clusters does not generate significant losses in the heterogeneity obtained. This approach was applied to the Organ Procurement and Transplantation Network medical dataset (n = 11,332). The combination of Random Survival Forest and consensus clustering yielded the optimal result of 4 clusters starting from 8 initial ones. Finally, for each population, Kaplan-Meier survival curves are generated to predict the survival of new patients based on the predictions of the XGBoost classifier, with an overall multi-class AUC of 98.11%.

Introducing machine learning to analyze factors aimed at successful development of the individual social qualities

The paper considers an important problem of the successful development of social qualities in an individual using machine learning methods. Social qualities play an important role in forming personal and professional lives, and their development is becoming relevant in modern society. The paper presents an overview of modern research in social psychology and machine learning; besides, it describes the data analysis method to identify factors influencing success in the development of social qualities. By analyzing large amounts of data collected from various sources, the authors of the paper use machine learning algorithms, such as Kohonen maps, decision tree and neural networks, to identify relationships between different variables, including education, environment, personal characteristics, and the development of social skills. Experiments were conducted to analyze the considered datasets, which included the introduction of methods to find dependencies between the input and output parameters. Machine learning introduction to find factors influencing the development of individual social qualities has varying dependence accuracy. The study results could be useful for both practical purposes and further scientific research in social psychology and machine learning. The paper represents an important contribution to understanding the factors that contribute to the successful development of individual social skills and could be useful in the development of programs and interventions in this area. The main objective of the research was to study the functionalities of the machine learning algorithms and various models to predict the students’s success in learning.

Relative contribution of comorbid diseases to health-related quality of life in patients with Parkinson’s disease

BackgroundMultimorbidity is common in elderly people, and one of the major consequences of multimorbidity is low health-related quality of life (HRQoL). The aim of this study was to investigate the frequency of comorbid diseases in patients with Parkinson’s disease (PD) and to analyze their relative importance in HRQoL. The aim was also to examine agreement between the generic 15D questionnaire and the PD-specific Parkinson’s Disease Questionnaire (PDQ-8) to further validate 15D in the evaluation of HRQoL in patients with PD.MethodsPatients with PD (N = 551) filled a questionnaire on comorbid diseases, and the 15D questionnaire yielding a 15-dimensional health profile and a score representing the overall HRQoL. Self-organizing map was used for an unsupervised pattern recognition of the health profiles. Relative importance analysis was used to evaluate the contribution of 16 comorbid diseases to the 15D score. The agreement between 15D and PDQ-8 questionnaires was studied in a subset of 81 patients that were examined clinically.Results533 patients (96.7%) reported comorbid diseases. The most affected dimensions in the 15D questionnaire were secretion, usual activities, discomfort and symptoms, and sexual activity. Self-organizing map identified three patterns of health profiles that included patients with high, low or transition HRQoL. The transition subgroup was similar to low HRQoL subgroup in non-motor dimensions. Sixteen comorbid diseases explained 33.7% of the variance in the 15D score. Memory deficit, depression, heart failure, and atrial fibrillation had the highest relative importance. The intraclass correlation coefficient between the generic 15D and the PD-specific PDQ-8 was 0.642 suggesting moderate reliability.ConclusionsThe most marked differences in HRQoL were in the dimensions of secretion, usual activities, and sexual activity. Pattern detection of 15D health dimensions enabled the detection of a subgroup with disproportionately poor HRQoL in non-motor dimensions. The comorbid diseases affecting most to HRQoL were memory deficit and depression. The generic 15D questionnaire can be used in the evaluation of HRQoL in PD patients.