Search

Background. The current development of the postgraduate pedagogical education system is determined by increasing requirements for the professional competence of teachers and by new semantic guidelines relating to their activity-role and personal characteristics. The personalized model of advanced training for teachers, implemented in institutions of the additional professional education in St. Petersburg, totally takes into account these trends and is developing in accordance with them. Besides, this model is based on the mainstream methodology that is the deficit approach which involves carrying out monitoring procedures to identify teachers’ methodological and subject deficiencies and designing the professional development taking into account the results of these procedures. One of the promising options for organizing classes with teachers is the use of scenario-type educational technologies. Purpose. The purpose of the research is thesubstantiation of effective using the scenario-type educational technologies in the implementation of an advanced training personalized model for teachers of foreign languages. Materials and methods. The main theoretical method of the research is modeling; the leading empirical method is the collection and accumulation of data (observation, questioning, analyses of teachers’ activity output and their work experience). The article is based on a set of sources presented by statutory regulations, scientific articles and methodological materials on the topic of research.Разработанная модель была внедрена в практику работы, а также проверена на основе методики BARS, аспект Assessment. The designed model was implemented into practice as well as tested on the basis of the BARS methodology, Assessment aspect. Results. The system of postgraduate teacher education is currently developing in accordance with the latest requirements and challenges of our time to the level of teachers’ professional competencies. One of the priority demands is to personalize the process of professional development for teachers and focus on compensating for their professional deficiencies. Therefore, it seems methodologically justified to build the educational process on the basis of the deficit approach taking into account the methodological and subject deficiencies for the teachers of foreign languages.The requirement to personalize the process of advanced training can be effectively fulfilled on the basis of the implication of the deficit approach with a scenario approach, which involves using scenarios that is real situations of professional interaction. The methodological and technological solution in this case is the applying of scenario-type educational technologies in the implementation of the advanced training personalized model for teachers of foreign languages.These educational technologies can be classified into two large groups: technologies of group dynamics and technologies of dialogue-professional scenarios. The testing of educational technologies “vortex groups”, “gamification”, “conversation in four corners” and “cubing” in the implementation of additional professional education programs included in the order register of the Education Committee of St. Petersburg for 2024 has proven their effectiveness on the BARS scale, Assessment aspect.

Machine learning with considering data privacy-preservation and personalized models has received attentions, especially in the manufacturing field. The data often exist in the form of isolated islands and cannot be shared because of data privacy in real industrial scenarios. It is difficult to gather the data to train a personalized model without compromising data privacy. To address this issue, we proposed a Federated Transfer Learning framework based on Auxiliary Classifier Generative Adversarial Networks named ACGAN-FTL. In the framework, Federated Learning (FL) trains a global model on decentralized datasets of the clients with data privacy-preservation and Transfer Learning (TL) transfers the knowledge from the global model to a personalized model with a relatively small data volume. ACGAN acts as a data bridge to connect FL and TL by generating similar probability distribution data of clients since the client datasets in FL cannot be directly used in TL for data privacy-preservation. A real industrial scenario of pre-baked carbon anode quality prediction is applied to verify the performance of the proposed framework. The results show that ACGAN-FTL can not only obtain acceptable performance on 0.81 accuracy, 0.86 precision, 0.74 recall, and 0.79 F1 but also ensure data privacy-preservation in the whole learning process. Compared to the baseline method without FL and TL, the former metrics have increased by 13%, 11%, 16%, and 15% respectively. The experiments verify that the performance of the proposed ACGAN-FTL framework fulfills the requirements of industrial scenarios. [ABSTRACT FROM AUTHOR]

This paper presents the experience of using the V Flow high-frame-rate ultrasound vector imaging method to study the pulsatile velocity fields in the area of the proximal anastomosis for femoral popliteal bypass surgery in vitro and in vivo. A representative (average) anastomosis model and the experimental setup designed for in vitro studies covering forward and reverse flow phases throughout the cycle are described. The results of the measurements are presented for areas with a relatively uniform velocity distribution and for areas with pronounced spatial inhomogeneities due to the jet or recirculating nature of the flow. The results of ultrasonic studies of the velocity field of the three-dimensional pulsatile flow in vitro and in vivo are compared with the data of numerical simulations carried out for the average and personalized models based on the Navier–Stokes equations. Acceptable consistency between the results of experimental and numerical studies is demonstrated. [ABSTRACT FROM AUTHOR]

Complex systems are generating more and more high-dimensional data for which tensor analysis showed promising results by capturing complex correlation structures of data. Such data is often distributed among various sites creating challenges for developing data-driven models. Specifically, data privacy and security concerns have been exacerbated in recent years and drove the demand to store and analyze data at the edge of networks rather than sharing it with a centralized server. Federated learning frameworks have been introduced as a solution to these concerns. These frameworks allow local clients to learn local models and collaborate with others to develop a more generalizable aggregated model while handling data privacy issues. In this article, we propose a federated generalized scalar-on-tensor regression framework where multiple local tensor models are learned at the edge, and their parameters are shared with and updated by an aggregator. Experiments on synthetic data sets and two real-world data sets from agriculture and manufacturing domains show the superiority of our approach over several benchmarks. [ABSTRACT FROM AUTHOR]

This paper presents the experience of using the V Flow high-frame-rate ultrasound vector imaging method to study the pulsatile velocity fields in the area of the proximal anastomosis for femoral popliteal bypass surgery in vitro and in vivo. A representative (average) anastomosis model and the experimental setup designed for in vitro studies covering forward and reverse flow phases throughout the cycle are described. The results of the measurements are presented for areas with a relatively uniform velocity distribution and for areas with pronounced spatial inhomogeneities due to the jet or recirculating nature of the flow. The results of ultrasonic studies of the velocity field of the three-dimensional pulsatile flow in vitro and in vivo are compared with the data of numerical simulations carried out for the average and personalized models based on the Navier–Stokes equations. Acceptable consistency between the results of experimental and numerical studies is demonstrated.

With the flourishing development of the Internet of Things (IoT), federated learning has garnered significant attention as a distributed learning method aimed at preserving the privacy of participant data. However, certain IoT devices, such as sensors, face challenges in effectively employing conventional federated learning approaches due to limited computational and storage resources, which hinder their ability to train complex local models. Additionally, in IoT environments, devices often face problems of data heterogeneity and uneven benefit distribution between them. To address these challenges, a personalized and fair split learning framework is proposed for resource-constrained clients. This framework first adopts a U-shaped structure, dividing the model to enable resource-constrained clients to offload subsets of the foundational model to a central server while retaining personalized model subsets locally to meet the specific personalized requirements of different clients. Furthermore, to ensure fair benefit distribution, a model-aggregation method with optimized aggregation weights is used. This method reasonably allocates model-aggregation weights based on the contributions of clients, thereby achieving collaborative fairness. Experimental results demonstrate that, in three distinct data heterogeneity scenarios, employing personalized training through this framework exhibits higher accuracy compared to existing baseline methods. Simultaneously, the framework ensures collaborative fairness, fostering a more balanced and sustainable cooperation among IoT devices.

PurposeHigh-grade serous ovarian cancer (HGSOC) is aggressive and has a high mortality rate. A Vit-based deep learning model was developed to predicting overall survival in HGSOC patients based on preoperative CT images.Methods734 patients with HGSOC were retrospectively studied at Qilu Hospital of Shandong University with preoperative CT images and clinical information. The whole dataset was randomly split into training cohort (n = 550) and validation cohort (n = 184). A Vit-based deep learning model was built to output an independent prognostic risk score, afterward, a nomogram was then established for predicting overall survival.ResultsOur Vit-based deep learning model showed promising results in predicting survival in the training cohort (AUC = 0.822) and the validation cohort (AUC = 0.823). The multivariate Cox regression analysis indicated that the image score was an independent prognostic factor in the training (HR = 9.03, 95% CI: 4.38, 18.65) and validation cohorts (HR = 9.59, 95% CI: 4.20, 21.92). Kaplan-Meier survival analysis indicates that the image score obtained from model yields promising prognostic significance to refine the risk stratification of patients with HGSOC, and the integrative nomogram achieved a C-index of 0.74 in the training cohort and 0.72 in the validation cohort.ConclusionsOur model provides a non-invasive, simple, and feasible method to predicting overall survival in patients with HGSOC based on preoperative CT images, which could help predicting the survival prognostication and may facilitate clinical decision making in the era of individualized and precision medicine.

After hearing loss retrograde degeneration of spiral ganglion neurons (SGNs) has been described. Studies modeling the effects of degeneration mostly omitted peripheral processes (dendrites). Recent experimental observations indicated that degenerating SGNsmanifested also a reduced diameter of their dendrites.We simulated populations of 400 SGNs inside a high resolution cochlear model with a cochlear implant, based on μCT scans of a human temporal bone. Cochlear implant stimuli were delivered as biphasic pulses in a monopolar configuration. Three SGN situations were simulated, based on our previous measurements of human SGN dendrites: (A) SGNs with intact dendrites (before degeneration), (B) degenerating SGNs, dendrites with a smaller diameter but original length, (C) degenerating SGNs, dendrites omitted. SGN fibers were mapped to characteristic frequency, and place pitch was estimated from excitation profiles. Results from degenerating SGNs (B, C) were similar. Most action potentials were initiated in the somatic area for all cases (A, B, C), except for areas near stimulating electrodes in the apex with intact SGNs (A), where action potentials were initiated in the distal dendrite. In most cases, degenerating SGNs had lower thresholds than intact SGNs (A) (down to –2 dB). Excitation profiles showed increased ectopic activation, i.e., activation of unintended neuronal regions, as well as similar neuronal regions excited by different apical electrodes, for degenerating SGNs (B, C). The estimated pitch showed cases of pitch reversals in apical electrodes for intact SGNs (A), as well as mostly identical pitches evoked by the four most apical electrodes for degenerating SGNs (B, C). In conclusion, neuronal excitation profiles to electrical stimulation exhibited similar traits in both ways of modeling SGN degeneration. Models showed degeneration of dendrites caused increased ectopic activation, aswell as similar excitation profiles and pitch evoked by different apical electrodes. Therefore, insertion of electrodes beyond approximately 450◦ may not provide any benefit if SGN dendrites are degenerated. [ABSTRACT FROM AUTHOR]

Objective: Pain assessment is of great importance in both clinical research and patient care. Facial expression analysis is becoming a key part of pain detection because it is convenient, automatic, and real-time. The aim of this study is to present a cold pain intensity estimation experiment, investigate the importance of the spatial-temporal information on facial expression based cold pain, and study the performance of the personalized model as well as the generalized model. Methods: A cold pain experiment was carried out and facial expressions from 29 subjects were extracted. Three different architectures (Inception V3, VGG-LSTM, and Convolutional LSTM) were used to estimate three intensities of cold pain: No pain, Moderate pain, and Severe Pain. Architectures with Sequential information were compared with single-frame architecture, showing the importance of spatial-temporal information on pain estimation. The performances of the personalized model and the generalized model were also compared. Results: A mean F1 score of 79.48% was achieved using Convolutional LSTM based on the personalized model. Conclusion: This study demonstrates the potential for the estimation of cold pain intensity from facial expression analysis and shows that the personalized spatial-temporal framework has better performance in cold pain intensity estimation. Significance: This cold pain intensity estimator could allow convenient, automatic, and real-time use to provide continuous objective pain intensity estimations of subjects and patients.

The Internet of Things (IoT) is on the verge of a major paradigm shift. In the IoT system of the future, IoFT, the “cloud” will be substituted by the “crowd” where model training is brought to the edge, allowing IoT devices to collaboratively extract knowledge and build smart analytics/models while keeping their personal data stored locally. This paradigm shift was set into motion by the tremendous increase in computational power on IoT devices and the recent advances in decentralized and privacy-preserving model training, coined as federated learning (FL). This article provides a vision for IoFT and a systematic overview of current efforts towards realizing this vision. Specifically, we first introduce the defining characteristics of IoFT and discuss FL data-driven approaches, opportunities, and challenges that allow decentralized inference within three dimensions: (i) a global model that maximizes utility across all IoT devices, (ii) a personalized model that borrows strengths across all devices yet retains its own model, (iii) a meta-learning model that quickly adapts to new devices or learning tasks. We end by describing the vision and challenges of IoFT in reshaping different industries through the lens of domain experts. Those industries include manufacturing, transportation, energy, healthcare, quality & reliability, business, and computing.

• Sleep staging models were constructed based on core body temperature rhythm. • Accuracy for overall males' and females' models stood at 0.70 and 0.77 respectively. • Personalized sleep staging models had promising prospects with peak accuracy reaching at 95%. Core body temperature (CBT) rhythm and sleep process are closely intertwined, yet there is a lack of quantitative methods to link them. To accomplish the research link of "sleeping thermal environment - body temperature - sleep quality" and also to establish a new sleep staging methodology, sleep staging models based on random forest algorithm were constructed. These models fulfilled the requirement of distinguishing four sleep stages (Waking stage, REM stage, Light sleep and Deep sleep) based on the CBT rhythm. The database utilized for model construction emanated from a winter sleep experiment, wherein seven male and seven female participants underwent a consecutive three-night testing protocol within laboratory settings. Data analysis results confirmed that (1) For the separately constructed sleep staging models for males and females, their accuracies stood at 0.70 and 0.77, respectively. The overall performance of the female model marginally surpassed that of the male model; (2) The predominant error in existing models arisen from misjudgments of adjacent sleep stages, such as misclassifying REM stage and deep sleep as light sleep, and this phenomenon may arise from the individual difference in CBT distribution; (3) Personalized sleep staging models have promising prospects, with their peak accuracy reaching an impressive 95% - nearly approaching the performance of gold standard polysomnography (PSG). In summary, this study elucidates the inherent quantitative connection between CBT rhythm and sleep stages, and the proposed sleep staging model has practical value. [ABSTRACT FROM AUTHOR]

The greatest challenge in engineering of human mandible implants lies in its customization for each patient individually, by adapting them to the patient’s anatomical, morphological and physiological characteristics. This customization maximizes the efficiency of the patient‘s health recovery process. The application of anatomically shaped and personalized bone endoprosthesis, fixation plate and scaffold models bring great improvement to the clinical practice in maxillofacial surgery. It ensures that implant meets the biomechanical and dentofacial aesthetic requirements and, ultimately, reduces complications during recovery. In order to create such implants, novel procedure based on personalized models of mandible and its parts, and also plates and scaffold implants is presented in this paper. Design procedures for the creation of the personalized models are based on the application of Method of Anatomical Features, which has been already applied for the creation of geometrical models of human bones. This procedure improves pre-surgical planning, enables better execution of surgical intervention, and as a consequence improves patient recovery processes.

BackgroundThe aim was to develop a personalized survival prediction deep learning model for adenosarcoma patients using the surveillance, epidemiology and end results (SEER) database.MethodsA total of 797 uterine adenosarcoma patients were enrolled in this study. Duplicated and useless variables were excluded, and 15 variables were selected for further analyses, including age, grade, positive lymph nodes or not, marital status, race, tumor extension, stage, and surgery or not. We created our deep survival learning (DSL) model to manipulate the data, which was randomly split into a training set (n = 519, 65%), validation set (n = 143, 18%) and testing set (n = 143, 18%). The Cox proportional hazard (CPH) model was also included comparatively. Finally, personalized survival curves were plotted for randomly selected patients.ResultsThe c-index for the CPH model was 0.726, and the Brier score was 0.17. For our deep survival learning model, we achieved a c-index of 0.774 and a Brier score of 0.14 in the external testing set. In addition, the limitations of the traditional staging system were revealed, and a personalized survival prediction system based on our risk scoring grouping was developed.ConclusionsOur study developed a deep neural network model for adenosarcoma. The performance of this model was superior to that of the traditional Cox proportional hazard model. In addition, a personalized survival prediction system was developed based on our deep survival learning model, which provided more accurate prognostic information for adenosarcoma patients.

Background: The aim was to develop a personalized survival prediction deep learning model for adenosarcoma patients using the surveillance, epidemiology and end results (SEER) database. Methods: A total of 797 uterine adenosarcoma patients were enrolled in this study. Duplicated and useless variables were excluded, and 15 variables were selected for further analyses, including age, grade, positive lymph nodes or not, marital status, race, tumor extension, stage, and surgery or not. We created our deep survival learning (DSL) model to manipulate the data, which was randomly split into a training set (n = 519, 65%), validation set (n = 143, 18%) and testing set (n = 143, 18%). The Cox proportional hazard (CPH) model was also included comparatively. Finally, personalized survival curves were plotted for randomly selected patients. Results: The c-index for the CPH model was 0.726, and the Brier score was 0.17. For our deep survival learning model, we achieved a c-index of 0.774 and a Brier score of 0.14 in the external testing set. In addition, the limitations of the traditional staging system were revealed, and a personalized survival prediction system based on our risk scoring grouping was developed. Conclusions: Our study developed a deep neural network model for adenosarcoma. The performance of this model was superior to that of the traditional Cox proportional hazard model. In addition, a personalized survival prediction system was developed based on our deep survival learning model, which provided more accurate prognostic information for adenosarcoma patients. [ABSTRACT FROM AUTHOR]

• Dynamic modeling of energy expenditure can capture transient thermoregulation and food intake effect. • LSTM networks promise high accuracy of time-series data prediction and capturing patterns. • Adequate accuracy of dynamic EE prediction should be within 5–10 % of MAPE. • The ensemble of CNN-LSTM and LGBM models can lead to the high accuracy of personalized EE prediction. • Personalized modeling architecture and features selection can lead to capturing transient EE effects in individuals. Dynamic human energy expenditure (EE) consists of energy costs for resting metabolism, food digestion, physical activity, and thermoregulation. Currently, multiple models predict EE mainly concerning physical activity, thus, discarding other factors contributing to the dynamic variation of EE. This paper aimed to demonstrate that (i) a dynamic human body EE prediction requires the time series approach, (ii) personalization of input features and models can outperform the generalized approach. To achieve these objectives, data were collected from 3 sets of experiments with 6 test subjects wearing multiple sensors. The analysis of features' importance showed that the selection of features varies for activity-dominated and non-activity-dominated cases and also varies between individuals. Long-Short Term Memory (LSTM) networks were used to develop personalized models such as a simple LSTM, a convolutional LSTM (CNN-LSTM), and also an ensemble model combining CNN-LSTM with a Gradient Boosting algorithm (LSTM-LGBM). A personalized autoregressive linear model and a generalized approach of the LSTM-LGBM method were also developed to have a base of comparison. The results show that the personalized models provide good prediction accuracy, with the mean absolute percentage error (MAPE) mostly lying in the range of 5–15 %. The CNN-LSTM outperforms a simple LSTM model by 3–5 % in MAPE values, and the ensemble model outperforms the by 5–8 % the simple LSTM. The personalized modeling approach with LSTM has shown the potential to improve the prediction accuracy of dynamic EE and capture the non-activity-related effects such as thermoregulation and postprandial thermogenesis. [ABSTRACT FROM AUTHOR]

Atypical atrial flutter (AFlut) is a reentrant arrhythmia which patients frequently develop after ablation for atrial fibrillation (AF). Indeed, substrate modifications during AF ablation can increase the likelihood to develop AFlut and it is clinically not feasible to reliably and sensitively test if a patient is vulnerable to AFlut. Here, we present a novel method based on personalized computational models to identify pathways along which AFlut can be sustained in an individual patient. We build a personalized model of atrial excitation propagation considering the anatomy as well as the spatial distribution of anisotropic conduction velocity and repolarization characteristics based on a combination of a priori knowledge on the population level and information derived from measurements performed in the individual patient. The fast marching scheme is employed to compute activation times for stimuli from all parts of the atria. Potential flutter pathways are then identified by tracing loops from wave front collision sites and constricting them using a geometric snake approach under consideration of the heterogeneous wavelength condition. In this way, all pathways along which AFlut can be sustained are identified. Flutter pathways can be instantiated by using an eikonal-diffusion phase extrapolation approach and a dynamic multifront fast marching simulation. In these dynamic simulations, the initial pattern eventually turns into the one driven by the dominant pathway, which is the only pathway that can be observed clinically. We assessed the sensitivity of the flutter pathway maps with respect to conduction velocity and its anisotropy. Moreover, we demonstrate the application of tailored models considering disease-specific repolarization properties (healthy, AF-remodeled, potassium channel mutations) as well as applicabiltiy on a clinical dataset. Finally, we tested how AFlut vulnerability of these substrates is modulated by exemplary antiarrhythmic drugs (amiodarone, dronedarone). Our novel method allows to assess the vulnerability of an individual patient to develop AFlut based on the personal anatomical, electrophysiological, and pharmacological characteristics. In contrast to clinical electrophysiological studies, our computational approach provides the means to identify all possible AFlut pathways and not just the currently dominant one. This allows to consider all relevant AFlut pathways when tailoring clinical ablation therapy in order to reduce the development and recurrence of AFlut.

With the flourishing development of the Internet of Things (IoT), federated learning has garnered significant attention as a distributed learning method aimed at preserving the privacy of participant data. However, certain IoT devices, such as sensors, face challenges in effectively employing conventional federated learning approaches due to limited computational and storage resources, which hinder their ability to train complex local models. Additionally, in IoT environments, devices often face problems of data heterogeneity and uneven benefit distribution between them. To address these challenges, a personalized and fair split learning framework is proposed for resource-constrained clients. This framework first adopts a U-shaped structure, dividing the model to enable resource-constrained clients to offload subsets of the foundational model to a central server while retaining personalized model subsets locally to meet the specific personalized requirements of different clients. Furthermore, to ensure fair benefit distribution, a model-aggregation method with optimized aggregation weights is used. This method reasonably allocates model-aggregation weights based on the contributions of clients, thereby achieving collaborative fairness. Experimental results demonstrate that, in three distinct data heterogeneity scenarios, employing personalized training through this framework exhibits higher accuracy compared to existing baseline methods. Simultaneously, the framework ensures collaborative fairness, fostering a more balanced and sustainable cooperation among IoT devices.

Thermal comfort and sensation are important aspects of building design and indoor climate control, as modern man spends most of the day indoors. Conventional indoor climate design and control approaches are based on static thermal comfort/sensation models that view the building occupants as passive recipients of their thermal environment. To overcome the disadvantages of static models, adaptive thermal comfort models aim to provide opportunity for personalized climate control and thermal comfort enhancement. Recent advances in wearable technologies contributed to new possibilities in controlling and monitoring health conditions and human wellbeing in daily life. The generated streaming data generated from wearable sensors are providing a unique opportunity to develop a real-time monitor of an individual's thermal state. The main goal of this work is to introduce a personalized adaptive model to predict individual's thermal sensation based on non-intrusive and easily measured variables, which could be obtained from already available wearable sensors. In this paper, a personalized classification model for individual thermal sensation with a reduced-dimension input-space, including 12 features extracted from easily measured variables, which are obtained from wearable sensors, was developed using least-squares support vector machine algorithm. The developed classification model predicted the individual's thermal sensation with an overall average accuracy of 86%. Additionally, we introduced the main framework of streaming algorithm for personalized classification model to predict an individual's thermal sensation based on streaming data obtained from wearable sensors. [ABSTRACT FROM AUTHOR]

Atypical atrial flutter (AFlut) is a reentrant arrhythmia which patients frequently develop after ablation for atrial fibrillation (AF). Indeed, substrate modifications during AF ablation can increase the likelihood to develop AFlut and it is clinically not feasible to reliably and sensitively test if a patient is vulnerable to AFlut. Here, we present a novel method based on personalized computational models to identify pathways along which AFlut can be sustained in an individual patient. We build a personalized model of atrial excitation propagation considering the anatomy as well as the spatial distribution of anisotropic conduction velocity and repolarization characteristics based on a combination of a priori knowledge on the population level and information derived from measurements performed in the individual patient. The fast marching scheme is employed to compute activation times for stimuli from all parts of the atria. Potential flutter pathways are then identified by tracing loops from wave front collision sites and constricting them using a geometric snake approach under consideration of the heterogeneous wavelength condition. In this way, all pathways along which AFlut can be sustained are identified. Flutter pathways can be instantiated by using an eikonal-diffusion phase extrapolation approach and a dynamic multifront fast marching simulation. In these dynamic simulations, the initial pattern eventually turns into the one driven by the dominant pathway, which is the only pathway that can be observed clinically. We assessed the sensitivity of the flutter pathway maps with respect to conduction velocity and its anisotropy. Moreover, we demonstrate the application of tailored models considering disease-specific repolarization properties (healthy, AF-remodeled, potassium channel mutations) as well as applicabiltiy on a clinical dataset. Finally, we tested how AFlut vulnerability of these substrates is modulated by exemplary antiarrhythmic drugs (amiodarone, dronedarone). Our novel method allows to assess the vulnerability of an individual patient to develop AFlut based on the personal anatomical, electrophysiological, and pharmacological characteristics. In contrast to clinical electrophysiological studies, our computational approach provides the means to identify all possible AFlut pathways and not just the currently dominant one. This allows to consider all relevant AFlut pathways when tailoring clinical ablation therapy in order to reduce the development and recurrence of AFlut. [ABSTRACT FROM AUTHOR]

Thermal comfort and sensation are important aspects of building design and indoor climate control, as modern man spends most of the day indoors. Conventional indoor climate design and control approaches are based on static thermal comfort/sensation models that view the building occupants as passive recipients of their thermal environment. To overcome the disadvantages of static models, adaptive thermal comfort models aim to provide opportunity for personalized climate control and thermal comfort enhancement. Recent advances in wearable technologies contributed to new possibilities in controlling and monitoring health conditions and human wellbeing in daily life. The generated streaming data generated from wearable sensors are providing a unique opportunity to develop a real-time monitor of an individual’s thermal state. The main goal of this work is to introduce a personalized adaptive model to predict individual’s thermal sensation based on non-intrusive and easily measured variables, which could be obtained from already available wearable sensors. In this paper, a personalized classification model for individual thermal sensation with a reduced-dimension input-space, including 12 features extracted from easily measured variables, which are obtained from wearable sensors, was developed using least-squares support vector machine algorithm. The developed classification model predicted the individual’s thermal sensation with an overall average accuracy of 86%. Additionally, we introduced the main framework of streaming algorithm for personalized classification model to predict an individual’s thermal sensation based on streaming data obtained from wearable sensors.

In this research we propose a novel method of face recognition based on texture and shape information. Age invariant face recognition enables matching of an image obtained at a given point in time against an image of the same individual obtained at an earlier point in time and thus has important applications, notably in law enforcement. We investigate various types of models built on different levels of data granularity. At the global level a model is built on training data that encompasses the entire set of available individuals, whereas at the local level, data from homogeneous sub-populations is used and finally at the individual level a personalized model is built for each individual. We narrow down the search space by dividing the whole database into subspaces for improving recognition time. We use a two-phased process for age invariant face recognition. In the first phase we identify the correct subspace by using a probabilistic method, and in the second phase we find the probe image within that subspace. Finally, we use a decision tree approach to combine models built from shape and texture features. Our empirical results show that the local and personalized models perform best when rated on both Rank-1 accuracy and recognition time. [ABSTRACT FROM AUTHOR]

Machine learning with considering data privacy-preservation and personalized models has received attentions, especially in the manufacturing field. The data often exist in the form of isolated islands and cannot be shared because of data privacy in real industrial scenarios. It is difficult to gather the data to train a personalized model without compromising data privacy. To address this issue, we proposed a Federated Transfer Learning framework based on Auxiliary Classifier Generative Adversarial Networks named ACGAN-FTL. In the framework, Federated Learning (FL) trains a global model on decentralized datasets of the clients with data privacy-preservation and Transfer Learning (TL) transfers the knowledge from the global model to a personalized model with a relatively small data volume. ACGAN acts as a data bridge to connect FL and TL by generating similar probability distribution data of clients since the client datasets in FL cannot be directly used in TL for data privacy-preservation. A real industrial scenario of pre-baked carbon anode quality prediction is applied to verify the performance of the proposed framework. The results show that ACGAN-FTL can not only obtain acceptable performance on 0.81 accuracy, 0.86 precision, 0.74 recall, and 0.79 F1 but also ensure data privacy-preservation in the whole learning process. Compared to the baseline method without FL and TL, the former metrics have increased by 13%, 11%, 16%, and 15% respectively. The experiments verify that the performance of the proposed ACGAN-FTL framework fulfills the requirements of industrial scenarios., (© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2023, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.)

Personalized modeling usually trains a predictive model for a new point using only observations similar to the new point. However, existing methodologies have limitations that do not reflect the target variable in the similarity calculation nor the density of neighbors. Thus, this paper proposes a new personalized modeling method. The proposed methodology transforms the input variables into the latent variables through a supervised autoencoder and calculates the similarity measure between observations in the transformed latent space. The proposed method also considers the neighborhood density around the test point. As a result of the experiments with real datasets, it was found that the proposed method outperformed other benchmark methods and showed the interpretability of the predictive model. • Trains predictive models for new observations using the sampled neighbor observations. • Proposes a new personalized modeling method through a supervised autoencoder. • Transforms the input variables into the latent variables and calculates the similarity measures. • Proposed method outperformed other benchmark methods and showed the interpretability. [ABSTRACT FROM AUTHOR]

Purpose: High-grade serous ovarian cancer (HGSOC) is aggressive and has a high mortality rate. A Vit-based deep learning model was developed to predicting overall survival in HGSOC patients based on preoperative CT images., Methods: 734 patients with HGSOC were retrospectively studied at Qilu Hospital of Shandong University with preoperative CT images and clinical information. The whole dataset was randomly split into training cohort (n = 550) and validation cohort (n = 184). A Vit-based deep learning model was built to output an independent prognostic risk score, afterward, a nomogram was then established for predicting overall survival., Results: Our Vit-based deep learning model showed promising results in predicting survival in the training cohort (AUC = 0.822) and the validation cohort (AUC = 0.823). The multivariate Cox regression analysis indicated that the image score was an independent prognostic factor in the training (HR = 9.03, 95% CI: 4.38, 18.65) and validation cohorts (HR = 9.59, 95% CI: 4.20, 21.92). Kaplan-Meier survival analysis indicates that the image score obtained from model yields promising prognostic significance to refine the risk stratification of patients with HGSOC, and the integrative nomogram achieved a C-index of 0.74 in the training cohort and 0.72 in the validation cohort., Conclusions: Our model provides a non-invasive, simple, and feasible method to predicting overall survival in patients with HGSOC based on preoperative CT images, which could help predicting the survival prognostication and may facilitate clinical decision making in the era of individualized and precision medicine., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2022 Zheng, Wang, Zhang, Li, Yang, Yang and Dong.)

This paper aims at the development and evaluation of a personalized insulin infusion advisory system (IIAS), able to provide real-time estimations of the appropriate insulin infusion rate for type 1 diabetes mellitus (T1DM) patients using continuous glucose monitors and insulin pumps. The system is based on a nonlinear model-predictive controller (NMPC) that uses a personalized glucose–insulin metabolism model, consisting of two compartmental models and a recurrent neural network. The model takes as input patient's information regarding meal intake, glucose measurements, and insulin infusion rates, and provides glucose predictions. The predictions are fed to the NMPC, in order for the latter to estimate the optimum insulin infusion rates. An algorithm based on fuzzy logic has been developed for the on-line adaptation of the NMPC control parameters. The IIAS has been in silico evaluated using an appropriate simulation environment (UVa T1DM simulator). The IIAS was able to handle various meal profiles, fasting conditions, interpatient variability, intraday variation in physiological parameters, and errors in meal amount estimations. [ABSTRACT FROM PUBLISHER]

Objective: Pain assessment is of great importance in both clinical research and patient care. Facial expression analysis is becoming a key part of pain detection because it is convenient, automatic, and real-time. The aim of this study is to present a cold pain intensity estimation experiment, investigate the importance of the spatial-temporal information on facial expression based cold pain, and study the performance of the personalized model as well as the generalized model., Methods: A cold pain experiment was carried out and facial expressions from 29 subjects were extracted. Three different architectures (Inception V3, VGG-LSTM, and Convolutional LSTM) were used to estimate three intensities of cold pain: No pain, Moderate pain, and Severe Pain. Architectures with Sequential information were compared with single-frame architecture, showing the importance of spatial-temporal information on pain estimation. The performances of the personalized model and the generalized model were also compared., Results: A mean F1 score of 79.48% was achieved using Convolutional LSTM based on the personalized model., Conclusion: This study demonstrates the potential for the estimation of cold pain intensity from facial expression analysis and shows that the personalized spatial-temporal framework has better performance in cold pain intensity estimation., Significance: This cold pain intensity estimator could allow convenient, automatic, and real-time use to provide continuous objective pain intensity estimations of subjects and patients.