Quality Prediction Research Articles

Prediction of Pellet Durability Index (PDI) in a Commercial Feed Mill Using Multiple Linear Regression with Variable Selection and Dimensionality Reduction.

Pellet quality, measured as Pellet Durability Index (PDI), is an important key performance indicator (KPI) for commercial feed manufacturing, as it can impact both mill efficiency and downstream performance of animals fed the manufactured diets. However, it is an ongoing challenge for the feed industry to control pellet quality, due to the complexity of feed manufacturing and the large number of variables influencing the process. Previous studies have explored prediction of pellet quality using either simple empirical models with a few variables or machine learning models with many variables. The objective of the current study was to develop statistical regression models to predict PDI, and to describe the relationship between pellet quality and 55 available variables based on a dataset with 2691 observations collected from a commercial feed mill. In the current study, the response variable (PDI) was transformed using the Box-Cox approach into the transformed response variable (tPDI), that was more normally distributed. Three multiple regression models were developed based on subsets of variables processed by variable selection and dimensionality reduction methods: Forward Selection, Principal Component Analysis, and Partial Least Squares. The results indicated that Model 1 (Forward Selection with manual removal of sparse variables), built on 9 variables, performed better than the other two models. It exhibited consistent model prediction performance on the training data and testing data, in terms of MAE (1.93 ± 0.063 versus 1.96), RMSPE (2.45 ± 0.079 versus 2.45), and CCC (0.549 ± 0.0273 versus 0.550), with a better prediction precision based on the fit plot. Expanding Temperature (℃), Fat Content (%), and ADF Content (%), and Indoor Humidity (Pelletizer) (%) were identified as more influential than other variables on the transformed response variable (tPDI) in Model 1, based on a behavior analysis. The models developed in the current study can be helpful to feed mills for predicting and comprehending the effect of a number of commonly measured variables on pellet quality in the commercial setting.

Synthetic Data Generation for AI-Informed End-of-Line Testing for Lithium-Ion Battery Production

Lithium-ion batteries are a key technology in supply chains for modern electric vehicles. Their production is complex and can be prone to defects. As such, the detection of defective batteries is critical to ensure performance and consumer safety. Existing end-of-line testing relies heavily on electrical measurements for identifying defective cells. However, it is possible that not all pertinent information is encoded within the electrical measurements alone. Reversible expansion in lithium-ion cells is an indicator of lithiation within the cell, while irreversible expansion is a consequence of the ageing process; unexpected expansion may indicate the presence of undesirable defects. By measuring expansion in addition to electrical measurements, we aim to make better and faster quality predictions during end-of-line testing, thereby facilitating the early detection of potential defects. To make these predictions, we implement artificial intelligence algorithms to extract information from the measurements. Training these networks requires large training datasets, which are expensive to produce. In this paper, we demonstrate a first-order physical modelling approach for generating synthetic data to pre-train artificial intelligence algorithms that perform anomaly detection on lithium-ion battery cells at the end-of-line. The equivalent circuit model used to generate voltage curves could be fit to real data with a mean absolute error of less than 1%, and the expansion model could be fit to a mean absolute error of less than 2% of the measured values. By pretraining the artificial intelligence network using synthetic data, we can leverage existing physical models to reduce the amount of training data required.

Data-Driven Soft Sensor Model Based on Multi-Timescale Feature Fusion for Crystal Quality Prediction in Czochralski Process

The accurate real-time prediction of the crystal quality index v/G is an important reference for the real-time monitoring of the growth quality status and the process optimization adjustment of semiconductor silicon single crystals. This paper proposes a data-driven crystal quality indicator v/G soft sensor prediction model based on multi-timescale feature fusion to achieve the effective prediction of the crystal quality indicator v/G. Firstly, the characteristics of the crystal quality index v/G in the growth process of Czochralski silicon single crystal are analyzed. Secondly, the crystal quality index v/G is broken down into several natural components using something called complete ensemble empirical mode decomposition with adaptive noise (CEEMDAD), which provides more stable data. On this basis, each intrinsic mode component is reconstructed according to the sample entropy. Then, the maximum mutual information coefficient (MIC) method is applied to identify the characteristic variables most closely associated with each reconstructed component of the crystal quality index v/G from the process-influencing factors. Then, a long short-term memory network with a self-attention mechanism is used to establish a prediction model of the reconstructed components to extract the multi-timescale feature information from the different components of the crystal quality index v/G. Finally, the prediction results of the crystal quality index v/G are obtained by fusing each subsequent prediction model. According to the actual field data, the comprehensive experimental results validate the efficacy of the proposed soft sensor modeling method for the crystal quality index v/G. Compared with the single model, the proposed prediction model has a smaller MAE, RMSE, and prediction performance index and a higher HR prediction hit rate.

Outcome prediction, quality of life, and life expectancy in metastatic spine tumors: WFNS spine committee recommendation.

This review aimed to formulate the most current, evidence-based recommendations for the prediction of outcome, life expectancy, and quality of life in patients with metastatic vertebral tumors. A systematic literature search on PubMed and Google Scholar from 2012-2022 was done, using the keywords "metastatic vertebral tumors + outcome prediction + prognoses," "quality of life + spine metastases," and "spine metastases + life expectancy." Our PubMed search yielded 402 articles for outcome prediction, whereas 40 articles were identified for life expectancy in spine metastases. These were carefully screened by the co-authors, resulting in 61 and 11 final articles analyzed for this study. Our PubMed search for quality of life yielded 137 articles, of which 63 were carefully analyzed for this study. This up-to-date information was reviewed at two separate Spine Committee meetings of the World Federation of Neurosurgical Societies (WFNS). Two rounds of the Delphi method were used to vote and arrive at a positive or negative consensus. The WFNS Spine Committee finalized seven recommendation guidelines on the prediction of outcome, life expectancy, and quality of life in metastatic vertebral tumors. Irrespective of the primary tumor, surgical decompression in appropriately selected patients potentially improves the quality of life. Pre-operative ambulatory status, overall performance, and age are independent predictors of outcome and overall survival. Prognostic scoring systems have evolved to principle-based algorithms, amongst which NOMS is the most widely used.The best tools to measure the quality of life are EUQOL5-D and SOSGOQ in patients with metastatic spine disease.

3D Nonequilibrium Ionization and Spectroscopic Modeling of Coronal Mass Ejections. I. Comparison with Hinode/EIS Observations

Abstract High-resolution EUV spectroscopy provides an informative diagnostic tool for the early evolution of coronal mass ejections (CMEs) since it can directly measure many physical properties of CME plasma close to the Sun that cannot be determined from white-light coronagraphs or narrowband imagers. The Hinode/EUV Imaging Spectrometer (EIS) captured a series of high-resolution EUV spectra of the 2008 April 9 event, also known as the “Cartwheel CME,” during its initial acceleration period. CMEs are inherently dynamic events, driving the plasma from ionization equilibrium, which makes the nonequilibrium ionization (NEI) effects important when analyzing spectral observations. In this work, we performed simulations of the Cartwheel CME using the Alfvén Wave Solar atmosphere Model (AWSoM) to provide insight into the plasma structure and dynamics during the early evolution of this CME. For the first time, we combined self-consistent nonequilibrium charge state calculations in the EUV spectral line synthesis for a CME simulation to account for the plasma departures from ionization equilibrium everywhere in the CME. In this first paper of the series, we compare the simulation results to the EIS observations of this event to assess the quality of the AWSoM predictions. We find that the CME is out of equilibrium, indicating that NEI effects should be considered in the EUV spectral synthesis. Overall, the model reproduces the thermodynamic evolution of the CME in the low solar corona well, as compared to the EIS observations.

Data-driven water quality prediction using hybrid machine learning approaches for sustainable development goal 6

Data-driven water quality prediction using hybrid machine learning approaches for sustainable development goal 6

Enhancing urban air quality prediction using time-based-spatial forecasting framework.

Air quality forecasting plays a pivotal role in environmental management, public health and urban planning. This research presents a comprehensive approach for forecasting the Air Quality Index (AQI). The proposed Time-Based-Spatial (TBS) forecasting framework is integrated with spatial and temporal information using machine learning techniques on data collected from a wide range of cities. The TBS employs Convolutional Neural Networks (CNNs) to capture spatial dependencies based on normalized latitude and longitude coordinates of the cities. Simultaneously, time series model, specifically the ARIMA (AutoRegressive Integrated Moving Average)was employed to capture temporal dependencies using pollutant concentration readings over time. The dataset included information such as date, time, pollutant concentrations and AQI was further preprocessed and divided into training and testing sets. The CNN was configured to utilize the normalized latitude and longitude grid, while the ARIMA model concurrently processed the pollutant concentrations. The model was trained on the training dataset, and a 6hour forecast is generated for each test instance. The outcomes demonstrate the TBS model's ability to accurately predict AQI values. The integration of CNNs and time series model allowed for an clearer and deeper understanding of geographical and pollutant concentration factors that contribute to air quality variations.

Photovoltaic power one-day and multistep-hourly AI predictions using node-by-node evolved binomial tree structures to form L-transformed PDE modules

Abstract Photovoltaic (PV) power is generated by two common types of solar components that are primarily affected by fluctuations and development in cloud structures as a result of uncertain and chaotic processes. Local PV forecasting is unavoidable in supply and load planning necessary in integration of smart systems into electrical grids. Intra- or day-ahead modelling of weather patterns based on Artificial Intelligence (AI) allows one to refine available 24 h. cloudiness forecast or predict PV production at a particular plant location during the day. AI usually gets an adequate prediction quality in shorter-level horizons, using the historical meteo- and PV record series as compared to Numerical Weather Prediction (NWP) systems. NWP models are produced every 6 h to simulate grid motion of local cloudiness, which is additionally delayed and usually scaled in a rough less operational applicability. Differential Neural Network (DNN) is based on a newly developed neurocomputing strategy that allows the representation of complex weather patterns analogous to NWP. DNN parses the n-variable linear Partial Differential Equation (PDE), which describes the ground-level patterns, into sub-PDE modules of a determined order at each node. Their derivatives are substituted by the Laplace transforms and solved using adapted inverse operations of Operation Calculus (OC). DNN fuses OC mathematics with neural computing in evolution 2-input node structures to form sum modules of selected PDEs added step-by-step to the expanded composite model. The AI multi- 1…9-h and one-stage 24-h models were evolved using spatio-temporal data in the preidentified daily learning sequences according to the applied input–output data delay to predict the Clear Sky Index (CSI). The prediction results of both statistical schemes were evaluated to assess the performance of the AI models. Intraday models obtain slightly better prediction accuracy in average errors compared to those applied in the second-day-ahead evening approach. However, the first-morning strategy application includes a much more complicated and time-consuming procedure, so in conclusion, the second one-mode series processing could be recommended. Model statistics are generally more successful than NWP data processing due to limitations, delays, and unavailability of free radiation or cloudiness forecasts in the systems. An all-day model enables the prediction of complete PVP periods in sequenced computing with acceptable operational quality in the late afternoon, which is inevitable in day-ahead management and load scheduling in smart grids based on PV energy.

Optimized Random Forest Models for Rock Mass Classification in Tunnel Construction

The accurate prediction of rock mass quality ahead of the tunnel face is crucial for optimizing tunnel construction strategies, enhancing safety, and reducing geological risks. This study developed three hybrid models using random forest (RF) optimized by moth-flame optimization (MFO), gray wolf optimizer (GWO), and Bayesian optimization (BO) algorithms to classify the surrounding rock in real time during tunnel boring machine (TBM) operations. A dataset with 544 TBM tunneling samples included key parameters such as thrust force per cutter (TFC), revolutions per minute (RPM), penetration rate (PR), advance rate (AR), penetration per revolution (PRev), and field penetration index (FPI), with rock classification based on the Rock Mass Rating (RMR) method. To address the class imbalance, the Borderline Synthetic Minority Over-Sampling Technique was applied. Performance assessments revealed the MFO-RF model’s superior performance, with training and testing accuracies of 0.992 and 0.927, respectively, and key predictors identified as PR, AR, and RPM. Additional validation using 91 data sets confirmed the reliability of the MFO-RF model on unseen data, achieving an accuracy of 0.879. A graphical user interface was also developed, enabling field engineers and technicians to make instant and reliable rock classification predictions, greatly supporting safe tunnel construction and operational efficiency. These models contribute valuable tools for real-time, data-driven decision-making in tunneling projects.

Wastewater quality prediction based on channel attention and TCN-BiGRU model.

Water quality prediction is crucial for water resource management, as accurate forecasting can help identify potential issues in advance and provide a scientific basis for sustainable management. To predict key water quality indicators, including chemical oxygen demand (COD), suspended solids (SS), total phosphorus (TP), pH, total nitrogen (TN), and ammonia nitrogen (NH₃-N), we propose a novel model, CA-TCN-BiGRU, which combines channel attention mechanisms with temporal convolutional networks (TCN) and bidirectional gated recurrent units (BiGRU). The model, which uses a multi-input multi-output (MIMO) architecture, is capable of simultaneously predicting multiple water quality indicators. It is trained and tested using data from a wastewater treatment plant in Huizhou, China. This study investigates the impact of data preprocessing and the channel attention mechanism on model performance and compares the predictive capabilities of various deep learning models. The results demonstrate that data preprocessing significantly improves prediction accuracy, while the channel attention mechanism enhances the model's focus on key features. The CA-TCN-BiGRU model outperforms others in predicting multiple water quality indicators, particularly for COD, TP, and SS, where MAE and RMSE decrease by approximately 23% and 26%, respectively, and R2 improves by 5.85%. Moreover, the model shows strong robustness and real-time performance across different wastewater treatment plants, making it suitable for short-term (1-3days) water quality prediction. The study concludes that the CA-TCN-BiGRU model not only achieves high accuracy but also offers low computational overhead and fast inference speed, making it an ideal solution for real-time water quality monitoring.

Understanding Post‐Release Dispersal and Habitat Selection Helps Refine Management of Translocated Populations

ABSTRACTTranslocation outcomes in connected habitats are often uncertain, as individuals dispersing outside managed areas are exposed to threats. Post‐release monitoring can reduce uncertainty by revealing how dispersal and habitat selection influences establishment and population growth which inform future translocations. We undertook post‐release monitoring to identify habitat selection patterns following a translocation of toutouwai (North Island robin, Petroica longipes) to a large, contiguous forest habitat. Post‐release monitoring aimed to estimate survival, dispersal, and territory establishment to inform management decisions and future release site selection. We created species distribution models using monitoring data to identify differences in habitat selection during the post‐release dispersal and territory establishment phases. Toutouwai dispersed across 1312 ha but established territories within only 113 ha and 1 km from the release location. Site fidelity was higher than predicted, and there was no difference in dispersal or habitat selection across demographic groups. Critically, high site fidelity suggested that the extent of managed habitat was sufficient to protect dispersing individuals. Habitat selection preferences were stronger during territory establishment and were associated with lower slopes, higher water deficit and proximity to water reservoirs. Species distribution modelling allowed for predictions of high‐quality core habitat where dispersal and territory establishment were more likely, resulting in targeted management to improve population growth. Our results show that initial dispersal in connected habitats may be much larger than suggested by territory data alone, and that management may need to protect larger areas to support successful establishment. We demonstrate how effective post‐release monitoring can inform predictions of habitat quality and dispersal and guide management actions to improve translocations outcomes.

Rice Quality and Yield Prediction Based on Multi-Source Indicators at Different Periods

This study aims to develop an effective and reliable method for estimating rice quality indices and yield, addressing the growing need for rapid, non-destructive, and accurate predictions in modern agriculture. Field experiments were conducted in 2018 at the Suiling Water Conservancy Comprehensive Experimental Station (47°27′ N, 127°06′ E), using Longqingdao 3 as the test variety. Measurements included the leaf area index (LAI), chlorophyll content (SPAD), leaf nitrogen content (LNC), and leaf spectral reflectance during the tillering, jointing, and maturity stages. Based on these parameters, spectral indicators were calculated, and univariate linear regression models were developed to predict key rice quality indices. The results demonstrated that the optimal R2 values for brown rice rate, moisture content, and taste value were 0.866, 0.913, and 0.651, with corresponding RMSE values of 0.122, 0.081, and 1.167. After optimizing the models, the R2 values for the brown rice rate and taste value improved significantly to 0.95 (RMSE: 0.075) and 0.992 (RMSE: 0.179), respectively. Notably, the spectral index GM2 during the jointing stage achieved the highest accuracy for yield prediction, with an R2 value of 0.822. These findings confirm that integrating multiple indicators across different growth periods enhances the accuracy of rice quality and yield predictions, offering a robust and intelligent solution for practical agricultural applications.

An Advanced Approach for Geometallurgical Modeling Applied to Bauxite Mines

Geometallurgy is an approach that integrates geology, mining, processing, and environmental areas, aiming to increase knowledge of the deposit and reduce risks in mining projects and/or operations. Growing changes in economic and operational scenarios require the development of robust models for metallurgical responses. Typically, the population of geometallurgical variables is small, and there are restrictions on applying geostatistical techniques, such as Ordinary Kriging, since recoveries are considered non-additive variables. This study used multiple linear regression to develop a geometallurgical model for a bauxite mine. The developed models enabled reconciliation with real results, and this application for bauxite mining represents a novelty in the literature. The model estimated the mass recovery in the coarse fraction with an accuracy greater than 97%. Additionally, the geometallurgical model developed for Mineração Rio do Norte (MRN) allows predictability and mapping potential product quality deviations.

Machine learning and GIS based groundwater quality prediction for agricultural practices - A case study form Arjunanadi River basin of South India

Machine learning and GIS based groundwater quality prediction for agricultural practices - A case study form Arjunanadi River basin of South India

Koopman-Constrained Hierarchical Deep State Space Model for Industrial Quality Prediction via Cloud-Edge Collaborative Framework

Koopman-Constrained Hierarchical Deep State Space Model for Industrial Quality Prediction via Cloud-Edge Collaborative Framework

Efficient Weather and Air Quality Forecasts Through AI

Accurate and timely weather forecasts and air quality predictions are essential for designing effective strategies to manage weather-related events and air pollution. These forecasts also play a key role in understanding aerosol-meteorology interactions within weather systems. However, traditional numerical methods, such as chemical transport models (CTMs), are computationally intensive, and their high resource demands limit their practical use in real-time air quality management and weather forecasting. In response to these challenges, we develop a novel approach called DeepCTM4D, which leverages deep learning to replicate CTM simulations, enhancing the computational efficiency of meteorology and air quality modeling in the four-dimensional chemistry space. The DeepCTM4D model is trained to accurately predict atmospheric chemical concentrations based on inputs such as precursor emissions, meteorological factors, and initial conditions. The key advantage of DeepCTM4D lies in its ability to efficiently identify the main drivers of pollution formation and assess how changes in emissions and meteorological conditions influence air quality. The relationships between emissions, meteorology, and concentration that DeepCTM4D captures align with established atmospheric chemistry mechanisms, further supporting the model’s scientific validity. Overall, DeepCTM4D offers a promising solution for simulating complex atmospheric processes, providing policymakers with critical information needed to design effective pollution control strategies and weather-caused events. This AI-driven model can also be integrated into global weather and air quality forecasting systems, serving as a powerful tool for more efficient, real-time predictions.

Bridging the gap: Phage manufacturing processes from laboratory to agri-food industry.

Interest in bacteriophages (phages) as sustainable biocontrol agents in the agri-food industry has increased because of growing worries about food safety and antimicrobial resistance (AMR). The phage manufacturing process is examined in this review, with particular attention paid to the crucial upstream and downstream processes needed for large-scale production. Achieving large phage yields requires upstream procedures, including fermentation and phage amplification. In the meantime, downstream procedures, including purification, endotoxin removal, and formulation, is essential for guaranteeing product quality and regulatory compliance. Despite advances in upstream and downstream process optimization of phage production processes, these methods are not effectively utilized in manufacturing processes. Additionally, the commercialization of phage products is hindered by fragmented rules and inconsistent regulations. Emerging technologies such as enhanced chromatography, continuous processing, and encapsulating techniques provide prospects for increased stability, efficiency, and scalability to fill these gaps. Furthermore, by facilitating real-time process optimization, predictive quality control (QC), and unique phage product creation, the integration of artificial intelligence (AI) and machine learning has the potential to transform the phage manufacturing industry completely. In order to provide consistent standards, encourage innovation, and bridge the gap between academic research and commercial applications, this review identifies gaps and highlights the necessity of cooperation between academia, industry, and regulatory agencies. To effectively utilize phages' potential to improve food safety, fight AMR, and promote sustainable agricultural practices, the agri-food industry must advance phage manufacturing techniques and harmonize regulatory frameworks.

Effect of Botulinum toxin on non-motor symptoms in adult-onset idiopathic focal/segmental dystonia.

Non-motor symptoms, including depression, anxiety, sleep disturbances, pain and cognitive dysfunction, are a much more important predictor of quality of life than the severity of dystonia. To assess the effect of Botulinum toxin on non-motor symptoms and quality of life in patients with adult-onset idiopathic focal dystonia. Patients aged > 18years diagnosed with idiopathic focal dystonia were recruited in this longitudinal cohort study. The severity of dystonia, non-motor symptoms, and quality of life were evaluated using the BFMDRS, DNMSQuest, and EQ-5D at baseline and 1 and 3months after botulinum toxin. 65 patients were recruited with a median age of 59years. Blepharospasm was the most common phenomenology. 49.2% of patients had depression at baseline assessed using the Beck Depression Inventory (BDI). There was a significant negative correlation between baseline BFMDRS, DNMSQuest, BDI, and HAM-A scores and quality of life, but there was no relation with the type of focal dystonia. The mean percentage improvement in the BFMDRS-M, BFMDRS-D, DNMSQuest, BDI, HAM-A and EQ-5D was 25%, 52%, 16%,20%,23% and 23%, respectively, at one month. There was no statistically significant correlation between percentage change in motor scores compared to depression and quality of life scores at one month. Botulinum toxin improved motor and non-motor scores and quality of life at 1 and 3months after botulinum toxin therapy. The motor scores did not correlate with depression and DNMSQUEST scores but showed a weak positive correlation with anxiety scores.

Occupational risk factors for sleep quality among Serbian airline pilots

IntroductionThe ability of airline pilots to maintain a good level of sleep goes a long way in ensuring operational effectiveness with regard to safety as well as personal health. The study assesses the risk factors for sleep quality of airline pilots in Serbia with the objective of determining those factors, both occupational and lifestyle that are paramount in assisting with sleep health.MethodsA cross-sectional study was conducted on a total of 66 Serbian commercial pilots, and the Pittsburgh Sleep Quality Index (PSQI) was used to assess their sleep quality. Demographic, occupational, lifestyle, biological and psychological variables were obtained through validated questionnaires. Multivariate logistic regression was used to determine the predictors of poor sleep quality (PSQI ≥5).ResultsOverall sleep quality was poor in 65.2% of the participants. Significant factors associated with poor sleep quality included years of experience (OR = 1.17, p = 0.007) and levels of stress (OR = 2.87, p = 0.004). Particular lifestyle variables, including factors such as coffee intake, had initial relationships with sleep quality but were not significant in the multivariate models. With regard to age, a significant univariate association was also revealed but was dropped in the final model because of collinearity with years of experience.ConclusionSerbian commercial pilots have a relatively high risk of inadequate sleep associated with years of experience and levels of stress. There is a need to implement organization-wide changes such as stress management schemes in order to improve sleep quality among pilots, and foster prioritizing well-being.

The quality of life and well-being of nursing home residents and their relatives during the COVID-19 pandemic: A quantitative approach

Background During the COVID-19 pandemic, nursing homes implemented protection measures to limit contact with others. Measures implemented in Switzerland included in-room isolation, prohibitions on group eating, and limitations on visiting and group activities. Nursing home residents thus experienced long periods of social isolation, which could have impacted their well-being and that of their relatives (whether direct family or other loved ones). The present study aimed to quantitatively describe and compare the quality of life, well-being and coping strategies of nursing home residents and their relatives during the implementation of protection measures against COVID-19 in the French-speaking part of Switzerland. Method 136 residents and 47 relatives from 13 nursing homes responded to the Perceived Stress Scale (PSS), the Brief Coping Orientation to Problems Experienced (Brief-COPE) questionnaire, the Impact of Event Scale-6 (IES-6) and the Post-Trauma Growth Inventory (PTGI). The only (minimal) difference between groups was that resident’s QoL was evaluated using the WHOQOL-OLD, and relatives’ QoL was evaluated using the WHOQOL-BREF scale. Non-parametric T-test were used to compare between group when possible, and linear regression to evaluate predictor of quality of life. Results Results showed that relatives coped better with lockdown isolation than did residents, residents developed lower levels of post-traumatic stress symptoms than relatives but there was no difference between the groups regarding perceived stress. Multiple linear regression calculations on indicators of quality of life showed that overall quality of life was most impacted by perceived stress. In the Psychological Health subscale, the familial link also reached significance: relatives who were part of a resident’s direct family had a better psychological QoL than other loved ones from outside. Conclusion The protection measures against COVID-19 isolated nursing residents, which impacted their quality of life and distanced residents and their relatives. In both populations, stress symptoms were the single most important predictor of quality of life.