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.

Human Activity Recognition (HAR) has emerged as a vital measure of quality of life, holding significant implications for human health. The need for effective mobility monitoring in diverse settings, both indoors and outdoors, necessitates the development of scientific and technological tools. To broaden accessibility, wearable devices like smartphones and smartwatches are commonly employed, leveraging sensor data analysis for valuable insights into user activities. HAR, treated as a classification task, involves training a classifier using sensor data and corresponding activity labels. The classifier aims to automatically recognize and classify various activities in future instances. However, the performance of a generic HAR model trained on a diverse population diminishes significantly when applied to a specific user due to inter-subject variability, encompassing variations in activity patterns, behavioral status, gait, and posture. To address this challenge, we propose an innovative auto-supervised domain adaptation approach based on particle filter theory, aiming to automatically construct a personalized HAR classifier. Our approach integrates multiple steps inspired by the particle filter formalism, enabling iterative approximation of the target distribution through temporal samples to personalize the HAR model for the specific user. In extensive experiments on public HAR datasets, we emphasize the critical role of personalization when deploying an HAR classifier for a new user. The results demonstrate that our framework significantly enhances the accuracy of HAR for new users compared to a non-personalized model, achieving an average improvement of 50% across most datasets. Furthermore, we implement our personalized HAR model on an embedded wearable device, enhancing its accessibility for real-world applications. [ABSTRACT FROM AUTHOR]

This paper proposed an innovative customized service framework for the federated learning model of the Internet of Vehicles (IoV), which addressed the difficulty of meeting the needs of users to train personalized models in IoV federated learning services. This framework adopted a federated learning aggregation algorithm that integrated device contribution and dataset similarity, achieving personalized federated learning. This algorithm used different weight allocation methods and similarity calculations to enable different users to choose appropriate model training schemes based on their own needs and data characteristics. The framework also proposed a dual sampling validation method to address model performance and credibility issues, and utilized smart contracts to support data collaboration, ensuring data security. The experimental results show that the proposed algorithm exhibits high accuracy in most experimental scenarios, and this framework can significantly improve the personalized level of IoV services while ensuring the accuracy and reliability of the model. [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. [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]

Background: The application of cone‐beam computed tomography (CBCT) in image‐guided radiotherapy and adaptive radiotherapy remains limited due to its poor image quality. Purpose: In this study, we aim to develop a deep learning framework to generate high‐quality CBCT images for therapeutic applications. Methods: The synthetic CT (sCT) generation from the CBCT was proposed using a transformer‐based network with a hybrid loss function. The network was trained and validated using the data from 176 patients to produce a general model that can be extensively applied to enhance CBCT images. After the first therapy, each patient can receive paired CBCT/planning CT (pCT) scans, and the obtained data were used to fine‐tune the general model for further improvement. For subsequent treatment, a patient‐specific, personalized model was made available. In total, 34 patients were examined for general model testing, and another six patients who underwent rescanned pCT scan were used for personalized model training and testing. Results: The general model decreased the mean absolute error (MAE) from 135 HU to 59 HU as compared to the CBCT. The hybrid loss function demonstrated superior performance in CT number correction and noise/artifacts reduction. The proposed transformer‐based network also showed superior power in CT number correction compared to the classical convolutional neural network. The personalized model showed improvement based on the general model in some details, and the MAE was reduced from 59 HU (for the general model) to 57 HU (p < 0.05 Wilcoxon signed‐rank test). Conclusion: We established a deep learning framework based on transformer for clinical needs. The deep learning model demonstrated potential for continuous improvement with the help of a suggested personalized training strategy compatible with the clinical workflow. [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.

The terminal settling velocity and drag coefficient are crucial for hydrochory in plant seeds, but predictive models for seeds with highly irregular shapes are rarely studied. The settling velocities and shape characteristics of seven plant seed types are investigated using particle Reynolds number, shape factor, and relative density difference. Irregularly shaped seeds exhibit profile symmetry and sink with their maximum projection area perpendicular to the fall direction, regardless of settling velocity. Irregular-shaped plant seeds are classified into three categories based on elongation-flatness and dimensionless settling velocity, with shape factors proposed for each category. The relationship between drag coefficient and particle Reynolds number of irregular-shaped plant seeds are analyzed in comparison to regular-shaped particles. A global optimization method based on generalized genetic algorithm are used for calibrating the calculation formulas. Compared with other general models, the newly presented formulae predict more accurately the settling velocity of irregular plant seeds. [Display omitted] • New formulas for drag coefficient of irregular plant seeds for various shapes and Reynolds numbers. • Settling characteristics are classified into three types with appropriate shape factors. • One profile of irregular seeds is symmetric, sinking with maximum area perpendicular to fall direction. [ABSTRACT FROM AUTHOR]

For individuals diagnosed with diabetes mellitus, it is crucial to keep a record of the carbohydrates consumed during meals, as this should be done at least three times daily, amounting to an average of six meals. Unfortunately, many individuals tend to overlook this essential task. For those who use an artificial pancreas, carbohydrate intake proves to be a critical factor, as it can activate the insulin pump in the artificial pancreas to deliver insulin to the body. To address this need, we have developed personalized deep learning model that can accurately detect carbohydrate intake with a high degree of accuracy. Our study employed a publicly available dataset gathered by an Inertial Measurement Unit (IMU), which included accelerometer and gyroscope data. The data was sampled at a rate of 15 Hz, necessitating preprocessing. For our tailored to the patient model, we utilized a recurrent network comprising Long short-term memory (LSTM) layers. Our findings revealed a median F1 score of 0.99, indicating a high level of accuracy. Additionally, the confusion matrix displayed a difference of only 6 s, further validating the model's accuracy. Therefore, we can confidently assert that our model architecture exhibits a high degree of accuracy. Our model performed well above 90% on the dataset, with most results between 98%-99%. The recurrent networks improved the problem-solving capabilities significantly, though some outliers remained. The model's average prediction latency was 5.5 s, suggesting that later meal predictions result in extended meal progress predictions. The dataset's limitation of mostly single-day data points raises questions about multi-day performance, which could be explored by collecting multi-day data, including night periods. Future enhancements might involve transformer networks and shorter time windows to improve model responsiveness and accuracy. Therefore, we can confidently assert that our model exhibits a high degree of accuracy., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 The Author(s). Published by Elsevier Ltd.. All rights reserved.)

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]

• Constructed an index to rank the gaps in risk perception between FCW and drivers. • Proposed a framework for aligning the gap in risk perception between FCW and drivers. • Constructed personalized models for 74 drivers, the average precision increased from 53.5% to 78.2%. • Reduce the average gaps between warning sending and braking from 2.4 s to 1.6 s. The Forward Collision Warning (FCW) system has been widely equipped on vehicles to reduce rear-end crashes, which are considered the most common type of crash. However, existing FCW systems have the problem of low response rates, which restrict their safety improvement effects. This study aims to address this issue by building personalized FCW models based on human risk preferences. First, a warning feedback index ranks the gaps between drivers' risk perceptions and FCW models. Then, reward models are developed to characterize the risk perception preferences of each individual driver. After that, the reward models serve as guidelines to fine-tune the benchmark FCW model using the Proximal Policy Optimization (PPO) algorithm. In the empirical analyses, a total of 95,814 warning fragments collected from 74 drivers are used, and the proposed method generates pseudo warning results. By comparing the pseudo and historical warnings, it shows that the precision of pseudo warning results increases from 53.5% to 78.2%. Furthermore, the average differences between the moment of warning and the moment of braking behavior decrease from 2.4 s to 1.6 s. This demonstrates a higher synchronization level in the timing of risk perception between the personalized FCW models and individual drivers, which enhances the driver's trust in the warning system. [ABSTRACT FROM AUTHOR]

Artificial intelligence-based Medical Image Analysis (AI-MIA) has achieved significant advantages in accuracy and efficiency in various biomedical applications. However, most existing deep learning (DL) models have a fixed network structure and aim at single MIA tasks, making it difficult to meet the diverse and complex MIA requirements. To improve the universality of the AI-MIA DL models, we propose a configurable DL framework, which consists of a MIA task element library and a DL model component library. We establish a MIA task element library to accurately define various potential MIA tasks for numerous diseases. In addition, we collect DL computation modules that can be used in different steps of MIA and built a DL model component library. Given a specific MIA demand of different diseases, the framework can build a personalized DL model by defining MIA tasks and configuring DL model components. Moreover, by comparing with the existing machine learning (ML)/DL models, the structure of the DL model for each configuration can be further adjusted to improve its performance. To demonstrate the effectiveness of the proposed framework, two personalized DL models are designed for upper abdominal organ detection and colon cancer cell classification, achieving high accuracy and high performance. Experimental results on actual medical image datasets show that the configurable DL framework can define specific MIA requirements for different diseases and flexibly configure DL model components to build suitable personalized DL models. Compared with the state-of-the-art methods, the significant advantage of our framework is that it can generate personalized DL models with a flexible structure and can continuously fine-tune the model structure to improve model performance. [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.

Purpose: Applications of deep learning (DL) are essential to realizing an effective adaptive radiotherapy (ART) workflow. Despite the promise demonstrated by DL approaches in several critical ART tasks, there remain unsolved challenges to achieve satisfactory generalizability of a trained model in a clinical setting. Foremost among these is the difficulty of collecting a task‐specific training dataset with high‐quality, consistent annotations for supervised learning applications. In this study, we propose a tailored DL framework for patient‐specific performance that leverages the behavior of a model intentionally overfitted to a patient‐specific training dataset augmented from the prior information available in an ART workflow—an approach we term Intentional Deep Overfit Learning (IDOL). Methods: Implementing the IDOL framework in any task in radiotherapy consists of two training stages: (1) training a generalized model with a diverse training dataset of N patients, just as in the conventional DL approach, and (2) intentionally overfitting this general model to a small training dataset‐specific the patient of interest (N+1) generated through perturbations and augmentations of the available task‐ and patient‐specific prior information to establish a personalized IDOL model. The IDOL framework itself is task‐agnostic and is, thus, widely applicable to many components of the ART workflow, three of which we use as a proof of concept here: the autocontouring task on replanning CTs for traditional ART, the MRI super‐resolution (SR) task for MRI‐guided ART, and the synthetic CT (sCT) reconstruction task for MRI‐only ART. Results: In the replanning CT autocontouring task, the accuracy measured by the Dice similarity coefficient improves from 0.847 with the general model to 0.935 by adopting the IDOL model. In the case of MRI SR, the mean absolute error (MAE) is improved by 40% using the IDOL framework over the conventional model. Finally, in the sCT reconstruction task, the MAE is reduced from 68 to 22 HU by utilizing the IDOL framework. Conclusions: In this study, we propose a novel IDOL framework for ART and demonstrate its feasibility using three ART tasks. We expect the IDOL framework to be especially useful in creating personally tailored models in situations with limited availability of training data but existing prior information, which is usually true in the medical setting in general and is especially true in ART. [ABSTRACT FROM AUTHOR]

This paper investigates whether adding a vertical distance sensor to a shank-mounted inertialmeasurement unit can improvewalking distance estimation accuracy. Some important features that can be extracted from the vertical distance sensor and a gyroscope were proposed, and the model was validated with personalized and generalized parameters. Results showthat the vertical distance sensor is unnecessary for personalized model, but can retain the estimation accuracy of generalized model at 96%. As a comparison, many existing methods yield high accuracy for personalized model, but some has the accuracy reduced by 5–10% for generalized model. [ABSTRACT FROM AUTHOR]

We proposed a personalized bolus advisor for patients with type 1 diabetes (T1D). A bolus advisor is a decision support system that recommends insulin doses based on an open‐loop model‐based optimization. To construct the bolus advisor, the optimal open‐loop control of blood glucose (BG) concentration in T1D patients was represented as a multi‐objective optimization problem. The insulin types, doses, and times for each injection were provided by the bolus advisor based on a personalized model and an average daily diet, which should be re‐tuned frequently in specific time intervals. The constructed personalized model for T1D patients incorporates effects of the patient's age and body weight. Two treatment schemes using three types of insulin (regular, lispro, and NPH) were investigated. The proposed bolus advisor was tested in silico on three virtual patients with different ages (from 9 to 50 years old) and body weights (from 28 to 100 kg) considering ±40% under‐ and over‐eating scenarios. The fluctuations in blood glucose and insulin levels are obviously wider in younger virtual subjects, which is showing the difficulties of the BG control problem in younger patients. [ABSTRACT FROM AUTHOR]

• 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]

Enhanced external counterpulsation (EECP) is a treatment and rehabilitation approach for ischemic diseases, including coronary artery disease. Its therapeutic benefits are primarily attributed to the improved blood circulation achieved through sequential mechanical compression of the lower extremities. However, despite the crucial role that hemodynamic effects in the lower extremity arteries play in determining the effectiveness of EECP treatment, most studies have focused on the diastole phase and ignored the systolic phase. In the present study, a novel siphon model (SM) was developed to investigate the interdependence of several hemodynamic parameters, including pulse wave velocity, femoral flow rate, the operation pressure of cuffs, and the mean blood flow changes in the femoral artery throughout EECP therapy. To verify the accuracy of the SM, we coupled the predicted afterload in the lower extremity arteries during deflation using SM with the 0D-1D patient-specific model. Finally, the simulation results were compared with clinical measurements obtained during EECP therapy to verify the applicability and accuracy of the SM, as well as the coupling method. The precision and reliability of the previously developed personalized approach were further affirmed in this study. The average waveform similarity coefficient between the simulation results and the clinical measurements during the rest state exceeded 90%. This work has the potential to enhance our understanding of the hemodynamic mechanisms involved in EECP treatment and provide valuable insights for clinical decision-making. [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.

INTRODUCTION: Machine learning algorithms and in silico models for the COVID-19 have been used to classify infectious people and predict their condition in time. OBJECTIVES: This study aims at creating a personalized model that combines machine learning and finite element simulation approach in order to predict development of COVID-19 infection in patients. METHODS: The methodology combines several aspects (1) classification of patients into several classes of clinical condition (2) segmentation of human lungs in X ray images (3) finite element simulation to investigate the spreading of SARS-COV-2 virion in the lungs. RESULTS: The findings show accuracy larger than 90% in all aspects of methodology. FE simulation has revealed that the distribution of airflow in the lung changes in time with the infection. CONCLUSION: The key benefit of our proposed method is that it combines several methods that will be further improved in order to create a truly unique combined methodology for predictive models in patients infected with COVID-19. [ABSTRACT FROM AUTHOR]

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]

Background: Tracking the whole body center of mass (CoM) trajectory of balance-impaired individuals with a personalized model is useful in the development of customized fall prevention strategies. A personalized CoM estimate can be obtained using the statically equivalent serial chain (SESC) method, but the subject has to perform an identification procedure to determine the set of subject-specific SESC parameters. During this identification, the subject must hold a series of static poses, some of which are unsuitable for balanced-impaired individuals.Research Question: Can non-static poses be used to replace the static poses during SESC parameter identification?Methods: A new method that extends the range of postures used to determine SESC parameters is presented. It takes advantage of CoM dynamics and can be executed by predominantly using dynamic motions with a few static frames. Furthermore, it is implemented using a Kalman filter to allow automatic switching between the dynamic and static models. The proposed method was tested with motion data obtained from seven healthy adults using a Vicon motion capture system and an AMTI force platform.Results: We found that dynamic motions could be used to estimate the SESC parameter and even reproduce ground reaction forces; however a small number of static poses are still required to determine the subject's CoM position. The SESC-based CoM estimate obtained with this new approach was similar to that obtained using conventional full-static identification, except that the subject did not have to assume and maintain static poses.Significance: Our proposed extension of the conventional SESC method would facilitate its application in the field of neuro-rehabilitation, especially in patients who need balance training. This personalized CoM method could be applicable for patients who are not able to maintain a static posture. In addition, this method helps minimize the total identification time by increasing the number of usable recorded frames. [ABSTRACT FROM AUTHOR]

With each passing day, consumers, particularly students, expect more intelligent and personalized services. The key to providing such services is the concept of a personalized model. Applying appropriate personalized learning models to students is a process that is filled with many challenges. The main purpose of the current report is to explore the challenges that impede the design and implementation of models that are strictly customized to the interests of students. In order to achieve this goal, it is necessary to explore the contemporary factors that are essential in designing personalization models. The factors are mainly related to the requirements of the business towards professionalism among students, as well as the goals set by the state for the education improvement. In order to achieve this goal, it is also necessary to investigate the technologies by which the personalization of models is realized and by which successful implementation of the personalized models is ensured in the students' education. [ABSTRACT FROM AUTHOR]

Background: Despite their high accuracy to recognize oral potentially malignant disorders (OPMDs) with cancer risk, non-invasive oral assays are poor in discerning whether the risk is high or low. However, it is critical to identify the risk levels, since high-risk patients need active intervention, while low-risk ones simply need to be follow-up. This study aimed at developing a personalized computational model to predict cancer risk level of OPMDs and explore its potential web application in OPMDs screening.Methods: Each enrolled patient was subjected to the following procedure: personal information collection, non-invasive oral examination, oral tissue biopsy and histopathological analysis, treatment, and follow-up. Patients were randomly divided into a training set (N = 159) and a test set (N = 107). Random forest was used to establish classification models. A baseline model (model-B) and a personalized model (model-P) were created. The former used the non-invasive scores only, while the latter was incremented with appropriate personal features.Results: We compared the respective performance of cancer risk level prediction by model-B, model-P, and clinical experts. Our data suggested that all three have a similar level of specificity around 90%. In contrast, the sensitivity of model-P is beyond 80% and superior to the other two. The improvement of sensitivity by model-P reduced the misclassification of high-risk patients as low-risk ones. We deployed model-P in web.opmd-risk.com, which can be freely and conveniently accessed.Conclusion: We have proposed a novel machine-learning model for precise and cost-effective OPMDs screening, which integrates clinical examinations, machine learning, and information technology. [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]

Community Question Answering (CQA) websites have become valuable knowledge repositories. Millions of internet users resort to CQA websites to seek answers to their encountered questions. CQA websites provide information far beyond a search on a site such as Google due to (1) the plethora of high-quality answers, and (2) the capabilities to post new questions toward the communities of domain experts. While most research efforts have been made to identify experts or to preliminarily detect potential experts of CQA websites, there has been a remarkable shift toward investigating how to keep the engagement of experts. Experts are usually the major contributors of high-quality answers and questions of CQA websites. Consequently, keeping the expert communities active is vital to improving the lifespan of these websites. In this article, we present an algorithm termed PALP to predict the activity level of expert users of CQA websites. To the best of our knowledge, PALP is the first approach to address a personalized activity level prediction model for CQA websites. Furthermore, it takes into consideration user behavior change over time and focuses specifically on expert users. Extensive experiments on the Stack Overflow website demonstrate the competitiveness of PALP over existing methods. [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.