Variations In Features Research Articles

A translationally informed approach to vital signs for psychiatry: a preliminary proof of concept

The nature of data obtainable from the commercial smartphone – bolstered by a translational model emphasizing the impact of social and physical zeitgebers on circadian rhythms and mood – offers the possibility of scalable and objective vital signs for major depression. Our objective was to explore associations between passively sensed behavioral smartphone data and repeatedly measured depressive symptoms to suggest which features could eventually lead towards vital signs for depression. We collected continuous behavioral data and bi-weekly depressive symptoms (PHQ-8) from 131 psychiatric outpatients with a lifetime DSM-5 diagnosis of depression and/or anxiety over a 16-week period. Using linear mixed-effects models, we related depressive symptoms to concurrent passively sensed behavioral summary features (mean and variability of sleep, activity, and social engagement metrics), considering both between- and within-person associations. Individuals with more variable wake-up times across the study reported higher depressive symptoms relative to individuals with less variable wake-up times (B [95% CI] = 1.53 [0.13, 2.93]). On a given week, having a lower step count (−0.16 [−0.32, −0.01]), slower walking rate (−1.46 [−2.60, −0.32]), lower normalized location entropy (−3.01 [−5.51, −0.52]), more time at home (0.05 [0.00, 0.10]), and lower distances traveled (−0.97 [−1.72, −0.22]), relative to one’s own typical levels, were each associated with higher depressive symptoms. With replication in larger samples and a clear understanding of how these components are best combined, a behavioral composite measure of depression could potentially offer the kinds of vital signs for psychiatric medicine that have proven invaluable to assessment and decision-making in physical medicine. Clinical Trials Registration: The data that form the basis of this report were collected as part of clinical trial number NCT03152864.

Machine learning predict the degradation efficiency of aqueous refractory organic pollutants by ultrasound-based advanced oxidation processes

Ultrasound based advanced oxidation processes (AOPs) are effective for removing refractory organic pollutants by generating reactive species. Machine learning (ML) can systematically provide an excellent opportunity to determine the relationship between feature variables and output variables through large amounts of data, thereby reducing the need for experimental measurements. In this study, a ML approach was applied for the first time to predict the degradation efficiency of aqueous refractory organic pollutants by ultrasound-based AOPs. The obtained dataset was normalized and introduced into four ML models. Among them, Random Forest Regressor and XGB Regressor models had the best fit with 0.9546 and 0.9749 for training set R2 and 0.9548 and 0.9740 for test set R2, respectively. Relative importance analysis and SHAP analysis showed that catalyst type, oxidant concentration, reaction time, and catalyst dosage are the important variables affecting the degradation efficiency of refractory organic pollutants. In addition, excessive ultrasonic intensity, ultrasonic frequency, oxidant concentration, and catalyst dosage had little effect on the increase of degradation efficiency. Based on the descriptive data analysis, high degradation efficiency was obtained with ultrasonic intensity of 2 W·mL−1, ultrasonic frequency of 20–40 kHz, reaction time of 60–90 min, pH 3–7, oxidant concentration of ∼7.5 mmol·L−1, and catalyst dosage of >0.4 g·L−1. This work demonstrated a new predictive method to understand the comprehensive influence of operative factors on degradation efficiency in US-based AOPs, guiding parameter and process optimization.

NIR and MIR spectral feature information fusion strategy for multivariate quantitative analysis of tobacco components

The determination of total nicotine, total sugar, reducing sugar and total nitrogen contents in tobacco is of great significance to tobacco quality evaluation and formulation design. To quickly detect the content of 4 components of tobacco, using near-infrared (NIR) and mid-infrared (MIR) spectral data from 129 solid samples of tobacco powder provided by Shanghai Tobacco Group Co., Ltd., Two NIR-MIR spectral fusion techniques are studied, that is, fusion technology 1 is to establish a model by fusing feature variables after variable selection of each spectrum. The fusion technology 2 is to first fuse the NIR-MIR spectral data and then select the variables to establish the model. Both fusion technologies use successive projections algorithm (SPA), competitive adaptive reweighted sampling (CARS), backward interval PLS (biPLS), forward interval PLS (fiPLS), synergy interval PLS (siPLS), and interval interaction moving window partial least squares (iMWPLS) algorithms to filter wavelength variables. The results showed that for total nicotine and total sugar, the PLSR model established by fusion technology method 2 combined with iMWPLS algorithm is the best, and its RMSEP decreases from 0.2314 to 1.3225 to 0.0821 and 0.8079 respectively compared with the full spectrum fusion method, which is superior to the single NIR and MIR models and NIR-MIR fusion technology 1. For reducing sugars, the simple full-spectrum fusion model has the best analytical ability and the lowest RMSEP, which is superior to the single NIR-MIR models and all models established by two spectral fusion techniques combined with six wavelength selection algorithms. For total nitrogen, the prediction effect of fusion technology 1 combined with iMWPLS algorithm model was significantly improved compared with single NIR and MIR models and NIR-MIR fusion technology 2, and its RMSEP was 0.0634. The results showed that the two NIR-MIR spectral fusion techniques made full use of the complementary information provided by NIR and MIR spectroscopy, and successfully applied them to the rapid detection of total nicotine, total sugar, reducing sugar and total nitrogen content in tobacco, which provided a new method and idea for the rapid detection of tobacco components.

An interpretable machine learning method for risk stratification of patients with acute coronary syndrome

Risk stratification for major adverse cardiovascular events (MACE) within one year in patients with acute coronary syndrome (ACS) undergoing percutaneous coronary intervention (PCI) remains a challenge. Although several predictive models based on machine learning have emerged, they are difficult to understand. This study aimed to develop a machine learning prediction model that is easy to understand and trustworthy by lay people to assess the risk of MACE in ACS patients undergoing PCI within one year of the procedure. This retrospective cohort study used medical data from 1105 patients to construct a machine-learning model. To ensure thoroughness and multidimensionality of model parsing, Shapley Additive explanations (SHAP) analysis and Local interpretable model-agnostic explanations (LIME) interpretation techniques were used to systematically and deeply interpret the constructed models from a global to a detailed level. The study assessed 12 machine learning methods' prediction models and found that the Random Forest model was the most effective in predicting the risk of MACE in ACS patients after undergoing PCI. The model achieved an AUC value of 0.807 in the validation set, with an accuracy of 0.82, and a stable F1 score of 0.51. SHAP analysis ranked eight key feature variables, such as LVEF, in global importance. The weights of each feature range in the prediction model were revealed using LIME analysis. The machine learning prediction model we developed is capable of accurately predicting the likelihood of patients with ACS experiencing a MACE within one year of surgery.

Homeostatic Regulation of Spike Rate within Bursts in Two Distinct Preparations.

Homeostatic plasticity represents a set of mechanisms thought to stabilize some function of neural activity. Here, we identified the specific features of cellular or network activity that were maintained after the perturbation of GABAergic blockade in two different systems: mouse cortical neuronal cultures where GABA is inhibitory and motoneurons in the isolated embryonic chick spinal cord where GABA is excitatory (males and females combined in both systems). We conducted a comprehensive analysis of various spiking activity characteristics following GABAergic blockade. We observed significant variability in many features after blocking GABAA receptors (e.g., burst frequency, burst duration, overall spike frequency in culture). These results are consistent with the idea that neuronal networks achieve activity goals using different strategies (degeneracy). On the other hand, some features were consistently altered after receptor blockade in the spinal cord preparation (e.g., overall spike frequency). Regardless, these features did not express strong homeostatic recoveries when tracking individual preparations over time. One feature showed a consistent change and homeostatic recovery following GABAA receptor block. We found that spike rate within a burst (SRWB) increased after receptor block in both the spinal cord preparation and cortical cultures and then returned to baseline within hours. These changes in SRWB occurred at both single cell and population levels. Our findings indicate that the network prioritizes the burst spike rate, which appears to be a variable under tight homeostatic regulation. The result is consistent with the idea that networks can maintain an appropriate behavioral response in the face of challenges.

New insights into the relationship between optical response and physicochemical properties in apple flesh: Hyperspectral microscope imaging technology

Hyperspectral microscope imaging (HMI) technique was employed to assess the changes in physicochemical parameters and microstructure of ‘Golden Delicious’ apples flesh during storage. Four regions of interest (ROIs), including whole-cell ROI, intercellular space ROI, cytoplasm ROI, and cell wall ROI were investigated to assess their relationships with physicochemical parameters. Different ROIs presented similar vibrational profiles, but with slight differences in spectral intensity, especially in the range of 800–1000 nm. Spectral angle mapper (SAM) was applied to the HMI of apple tissues at different storage stages to clearly show the structural changes of parenchyma cells, while principal component analysis (PCA) could highlight the distribution of sugars, water and pigments in apple flesh at the cellular scale. Simultaneously with the degradation of acid-soluble pectin (ASP), middle lamella dissolution and increased intercellular space were observed using SEM and TEM. Single feature variables were used to construct linear models based on pearson correlation analysis, with R2 of 0.96 for moisture at 982 nm, 0.85 for water-soluble pectin (WSP) at 420 nm, 0.82 for L* at 946 nm, 0.77 for soluble solids content (SSC) at 484 nm, and 0.66 for firmness at 490 nm. This work demonstrated the great potential of HMI technology as a fast, accurate and efficient solution for assessing the quality of ‘Golden Delicious’ apples.

Hierarchical multi-level dynamic hyperparameter deformable image registration with convolutional neural network

Objective. To enable the registration network to be trained only once, achieving fast regularization hyperparameter selection during the inference phase, and to improve registration accuracy and deformation field regularity. Approach. Hyperparameter tuning is an essential process for deep learning deformable image registration (DLDIR). Most DLDIR methods usually perform a large number of independent experiments to select the appropriate regularization hyperparameters, which are time-consuming and resource-consuming. To address this issue, we propose a novel dynamic hyperparameter block, which comprises a distributed mapping network, dynamic convolution, attention feature extraction layer, and instance normalization layer. The dynamic hyperparameter block encodes the input feature vectors and regularization hyperparameters into learnable feature variables and dynamic convolution parameters which changes the feature statistics of the high-dimensional features layer feature variables, respectively. In addition, the proposed method replaced the single-level structure residual blocks in LapIRN with a hierarchical multi-level architecture for the dynamic hyperparameter block in order to improve registration performance. Main results. On the OASIS dataset, the proposed method reduced the percentage of |Jϕ|⩽0 by 28.01 % , 9.78 % and improved Dice similarity coefficient by 1.17 % , 1.17 % , compared with LapIRN and CIR, respectively. On the DIR-Lab dataset, the proposed method reduced the percentage of |Jϕ|⩽0 by 10.00 % , 5.70 % and reduced target registration error by 10.84 % , 10.05 % , compared with LapIRN and CIR, respectively. Significance. The proposed method can fast achieve the corresponding registration deformation field for arbitrary hyperparameter value during the inference phase. Extensive experiments demonstrate that the proposed method reduces training time compared to DLDIR with fixed regularization hyperparameters while outperforming the state-of-the-art registration methods concerning registration accuracy and deformation smoothness on brain dataset OASIS and lung dataset DIR-Lab.

Photometric stereo multi-information fusion unsupervised anomaly detection algorithm

Due to different materials, product surfaces are susceptible to light, shadow, reflection, and other factors. Coupled with the appearance of defects of various shapes and types, as well as dust, impurities, and other interfering influences, normal and abnormal samples are difficult to distinguish and a common problem in the field of defect detection. Given this, this paper proposes an end-to-end photometric stereo multi-information fusion unsupervised anomaly detection model. First, the photometric stereo feature generator is used to obtain normal, reflectance, depth, and other information to reconstruct the 3D topographic details of the object’s surface. Second, a multi-scale channel attention mechanism is constructed to fully use the feature associations of different layers of the backbone network, and the limited feature information is used to enhance the defect characterization ability. Finally, the original image is fused with normal and depth features to find the feature variability between defects and defects, as well as between defects and background. The feature differences between the source and clone networks are utilized to achieve multi-scale detection and improve detection accuracy. In this paper, the model performance is verified on the PSAD dataset. The experimental results show that the algorithm in this paper has higher detection accuracy compared with other algorithms. Among them, the multi-scale attention mechanism and multi-information fusion input improve the detection accuracy by 2.56% and 1.57%, respectively. In addition, the ablation experiments further validate the effectiveness of the detection algorithm in this paper.

Multi-energy load forecasting for integrated energy system based on sequence decomposition fusion and factors correlation analysis

Considering the seasonal and cyclical fluctuation of loads and the complexity of multi-energy coupling, this paper proposes a novel load forecasting model based on sequence decomposition fusion and factors correlation analysis. Firstly, the variational mode decomposition (VMD) is used to decompose the highly complex load sequences and the novel influencing factors correlation analysis (ICA) is proposed to select strong factors and remove weak feature variables to construct the input and output set. Secondly, this paper proposes the combined forecasting framework MTL-CNN-BiGRU-Attention to simultaneously forecast the cooling, heat, and electricity loads, along with BiGRU used as the hard sharing layer to deeply explore the coupling information between different types of loads. Meanwhile, the gray wolf algorithm (GWO) is improved to accurately and quickly search for the optimal hyperparameters of the model. Finally, the dataset of a comprehensive energy station in Shanghai is used to test our model, and the results show that the MAPE of the cooling and electricity loads forecasting achieve 5.501% and 5.821% in summer and 5.921%, 7.899% and 7.541% for the cooling, heat and electricity loads in transition season and winter, which confirms the effectiveness and superiority of our model.

Prediction and Classification of Phenol Contents in Cnidium officinale Makino Using a Stacking Ensemble Model in Climate Change Scenarios

Recent studies have focused on using big-data-based machine learning to address the effects of climate change scenarios on the production and quality of medicinal plants. Challenges relating to data collection can hinder the analysis of key feature variables that affect the quality of medicinal plants. In the study presented herein, we analyzed feature variables that affect the phenolic content of Korean Cnidium officinale Makino (C. officinale Makino) under different climate change scenarios. We applied different climate change scenarios based on environmental information obtained from Yeongju city, Gyeongsangbuk-do, Republic of Korea, and cultivated C. officinale Makino to collect data. The collected data included 3237, 75, and 45 records, and data augmentation was performed to address this data imbalance. We designed a function based on the DPPH value to set the phenolic content grade in C. officinale Makino and proposed a stacking ensemble model for predicting the total phenol contents and classifying the phenolic content grades. The regression model in the performance evaluation presented an improvement of 6.23–7.72% in terms of the MAPE; in comparison, the classification model demonstrated a 2.48–3.34% better performance in terms of accuracy. The classification accuracy was >0.825 when classifying phenol content grades using the predicted total phenol content values from the regression model, and the area under the curve values of the model indicated high model fitness (0.987–0.981). We plan to identify the key feature variables for the optimal cultivation of C. officinale Makino and explore the relationships among these feature variables.

Textural heterogeneity of liver lesions in CT imaging - comparison of colorectal and pancreatic metastases

PurposeTumoral heterogeneity poses a challenge for personalized cancer treatments. Especially in metastasized cancer, it remains a major limitation for successful targeted therapy, often leading to drug resistance due to tumoral escape mechanisms. This work explores a non-invasive radiomics-based approach to capture textural heterogeneity in liver lesions and compare it between colorectal cancer (CRC) and pancreatic cancer (PDAC).Materials and methodsIn this retrospective single-center study 73 subjects (42 CRC, 31 PDAC) with 1291 liver metastases (430 CRC, 861 PDAC) were segmented fully automated on contrast-enhanced CT images by a UNet for medical images. Radiomics features were extracted using the Python package Pyradiomics. The mean coefficient of variation (CV) was calculated patient-wise for each feature to quantify the heterogeneity. An unpaired t-test identified features with significant differences in feature variability between CRC and PDAC metastases.ResultsIn both colorectal and pancreatic liver metastases, interlesional heterogeneity in imaging can be observed using quantitative imaging features. 75 second-order features were extracted to compare the varying textural characteristics. In total, 18 radiomics features showed a significant difference (p < 0.05) in their expression between the two malignancies. Out of these, 16 features showed higher levels of variability within the cohort of pancreatic metastases, which, as illustrated in a radar plot, suggests greater textural heterogeneity for this entity.ConclusionsRadiomics has the potential to identify the interlesional heterogeneity of CT texture among individual liver metastases. In this proof-of-concept study for the quantification and comparison of imaging-related heterogeneity in liver metastases a variation in the extent of heterogeneity levels in CRC and PDAC liver metastases was shown.

Identification of genetic subtypes in follicular lymphoma

Follicular lymphoma (FL) exhibits considerable variability in biological features and clinical trajectories across patients. To dissect the diversity of FL, we utilized a Bernoulli mixture model to identify genetic subtypes in 713 pre-treatment tumor tissue samples. Our analysis revealed the existence of five subtypes with unique genetic profiles that correlated with clinicopathological characteristics. The clusters were enriched in specific mutations as follows: CS (CREBBP and STAT6), TT (TNFAIP3 and TP53), GM (GNA13 and MEF2B), Q (quiescent, for low mutation burden), and AR (mutations of mTOR pathway-related genes). The subtype Q was enriched for patients with stage I disease and associated with a lower proliferative history than the other subtypes. The AR subtype was unique in its enrichment for IgM-expressing FL cases and was associated with advanced-stage and more than 4 nodal sites. The existence of subtypes was validated in an independent cohort of 418 samples from the GALLIUM trial. Notably, patients assigned to the TT subtype consistently experienced inferior progression-free survival when treated with immunochemotherapy. Our findings offer insight into core pathways distinctly linked with each FL cluster and are expected to be informative in the era of targeted therapies.

A Longitudinal Tree-Based Framework for Lapse Management in Life Insurance

Developing an informed lapse management strategy (LMS) is critical for life insurers to improve profitability and gain insight into the risk of their global portfolio. Prior research in actuarial science has shown that targeting policyholders by maximising their individual customer lifetime value is more advantageous than targeting all those likely to lapse. However, most existing lapse analyses do not leverage the variability of features and targets over time. We propose a longitudinal LMS framework, utilising tree-based models for longitudinal data, such as left-truncated and right-censored (LTRC) trees and forests, as well as mixed-effect tree-based models. Our methodology provides time-informed insights, leading to increased precision in targeting. Our findings indicate that the use of longitudinally structured data significantly enhances the precision of models in predicting lapse behaviour, estimating customer lifetime value, and evaluating individual retention gains. The implementation of mixed-effect random forests enables the production of time-varying predictions that are highly relevant for decision-making. This paper contributes to the field of lapse analysis for life insurers by demonstrating the importance of exploiting the complete past trajectory of policyholders, which is often available in insurers’ information systems but has yet to be fully utilised.

Applying the LightGBM model and SHAP interpretation to understand the urban land price index and its influencing factors in China

ABSTRACT Investigating the key factors affecting land prices is conducive to establishing targeted market-oriented reforms of the land market and promoting its healthy development. Traditional linear analysis cannot comprehensively compare the importance of feature variables and is unable to analyse the nonlinear influence process of feature variables. Therefore, based on the land transaction dataset from 2011 to 2020 in China, we used machine learning and the Shapley Additive exPlanations method to analyse the importance ranking of factors affecting land prices and analyse the nonlinear influence processes. The government fiscal, industrial development, and financial system indices are determined as the most important factors affecting residential, commercial, and industrial land price indices, respectively, and each exhibits different characteristics. The financial system index exerts the strongest impact on land price indices in large cities, whereas transportation and communication indices are the most dominant in small cities. Moreover, industrial development index is the most important influencing factor in industrial cities, whereas regional economic index exerts the greatest impact in small cities. By employing machine learning methods, we determined nonlinear effects that traditional linear models fail to obtain. Based on the results, we propose policy recommendations for market-oriented reforms on the land market, real-estate policies, and a stable monetary system.

EFFECT OF SERVICE FEATURES AND WORD OF MOUTH STRATEGY ON CUSTOMER DECISIONS IN USING BSI MOBILE

This study aims to determine the effect of Service Features and word of mouth strategies on customer decisions in using BSI Mobile banking. Research using quantitative methods. The location of this research took place at Bank Syariah Indonesia Cirebon KCP Plered 2. The population in this study was 4005. Calculated based on the Slovin formula, a sample of 100 participants was obtained. The instrument used is a questionnaire or a questionnaire. instrument validity used product moment validity and reliability used Croncbach's alpha. Data collection techniques through the distribution of questionnaires, interviews, and observation. Data analysis used included research instrument tests, classical assumption tests, multiple linear regression tests, and hypothesis. The results showed that part there was a significant positive effect of service features on customer decisions in using Mobile Banking for BSI Cirebon customers. KCP Plered 2 with a significant value of 0.000. Meanwhile, according to word of mouth, most of the variables did not have a significant positive effect on the customer's decision to use mobile banking for BSI Cirebon KCP Plered 2 customers with a significant value of 0.696. Simultaneously the influence of service feature variables and word of mouth contributed 22.75% to customer decisions in using BSI Mobile banking.

Building Shadow Detection Based on Improved Quick Shift Algorithm in GF‐2 Images

Shadows in remote sensing images contain crucial information about various features on the ground. In this study, a method for detecting building shadows in GF‐2 images based on improved quick shift was proposed. First, six feature variables were constructed: first principal component (PC1), brightness component (I), normalized difference shadow index (NDSI), morphological shadow index (MSI), normalized difference water index (NDWI), and normalized difference vegetation index (NDVI). Then, the image was segmented to obtain homogeneous objects, which were then classified using a random forest model. Two improvements were added to the quick shift algorithm: using PC1, I, and MSI as input data instead of RGB images; and adding Canny edge constraints. Validation in six research areas yields Kappa coefficients of 0.928, 0.896, 0.89, 0.913, 0.879, and 0.909, confirming method feasibility. In addition, comparative experiments demonstrate its effectiveness and robustness across different land cover types while mitigating the segmentation scale effect.

Explainable highway performance degradation prediction model based on LSTM

With the dense and huge highway network in China, the highway maintenance management system has become the most concerned issue for Chinese highway managers in recent years. Therefore, based on the highway network in Guizhou Province of China, this paper established the semi-rigid asphalt pavement performance multi-output Long Short-Term Memory (LSTM) prediction model with regional applicability, including Pavement Surface Condition Index (PCI), Pavement Rutting Depth Index (RDI), Pavement Riding Quality Index (RQI) and Pavement Skidding Resistance Index (SRI) and the hyperparameters of the model were optimized by Bayesian optimization algorithm and the casual analysis of the model was conducted based on SHapley Additive exPlanations (SHAP), providing a reference for China's highway network-level preventive maintenance decision-making. The results prove that the established model can effectively predict the performance of the following year through the data of the previous two years. The established model has a better comprehensive prediction effect in comparison to similar multi-output models. The average coefficient of determination (R2) of the four prediction indexes can reach 0.823. Besides, the prediction effect among the indexes is stable and multiple pavement performance indexes can be effectively and reliably predicted at the same time. In addition, the contribution of input feature variables to output is quantified based on SHAP. According to quantified the contribution, the association relationship of the neural network model is set-valued mapped to the causality relationship in the post hoc. The input feature variables with the main contribution are extracted for causal analysis. The causal analysis shows that the degradation of the four pavement performances all have obvious time memory characteristics and have significant differences, but their pavement performance degradation is the result of nonlinear changes caused by the coupling effects of traffic load, road age, climate, and pavement structure.

Naïve Bayes-based Student Graduation Prediction Model: Effectiveness and Implementation to Improve Timely Graduation

Studies in an educational institution, when the lack of timely graduation of students in each batch and the number of students in each batch, causes an imbalance between incoming students and outgoing students and causes a decrease in accreditation from the campus, this should not continue to happen, the solution to dealing with this problem as an early detection of students who graduate on time is to predict the length of the student study period they have. Therefore, researchers will discuss the design of a prediction system for graduating on time using the Naïve Bayes method, to predict student graduation so that there is no imbalance of incoming and outgoing students. The construction of this system also uses the Naïve Bayes method and the CRISP-DM (Cross Industry Standard Process Data Mining) development method. In this case study, the Naïve Bayes method predicts data into 2, namely 1 (graduated on time) and 0 (did not graduate on time) by labeling the data used. In this model using 3247 data with the selection of features, namely semester achievement index 1 (ips1), ips2, ips3, ips4, ips5, semester credit units1 (credits1), credits2, credits3, credits4, credits5, semester credit units not passed 1 (skstidaklulus1), skstidaklulus2, skstidaklulus3, skstidaklulus4, skstidaklulus5 and labels. Using these feature variables results in model performance with 80% accuracy, with 80% accuracy it can be said that the model works well.

Mitigating data quality challenges in ambulatory wrist-worn wearable monitoring through analytical and practical approaches

Chronic disease management and follow-up are vital for realizing sustained patient well-being and optimal health outcomes. Recent advancements in wearable technologies, particularly wrist-worn devices, offer promising solutions for longitudinal patient monitoring, replacing subjective, intermittent self-reporting with objective, continuous monitoring. However, collecting and analyzing data from wearables presents several challenges, such as data entry errors, non-wear periods, missing data, and wearable artifacts. In this work, we explore these data analysis challenges using two real-world datasets (mBrain21 and ETRI lifelog2020). We introduce practical countermeasures, including participant compliance visualizations, interaction-triggered questionnaires to assess personal bias, and an optimized pipeline for detecting non-wear periods. Additionally, we propose a visualization-oriented approach to validate processing pipelines using scalable tools such as tsflex and Plotly-Resampler. Lastly, we present a bootstrapping methodology to evaluate the variability of wearable-derived features in the presence of partially missing data segments. Prioritizing transparency and reproducibility, we provide open access to our detailed code examples, facilitating adaptation in future wearable research. In conclusion, our contributions provide actionable approaches for improving wearable data collection and analysis.

Machine Learning Predicts Peripherally Inserted Central Catheters-Related Deep Vein Thrombosis Using Patient Features and Catheterization Technology Features.

This study aims to use patient feature and catheterization technology feature variables to train the corresponding machine learning (ML) models to predict peripherally inserted central catheters-deep vein thrombosis (PICCs-DVT) and analyze the importance of the two types of features to PICCs-DVT from the aspect of "input-output" correlation. To comprehensively and systematically summarize the variables used to describe patient features and catheterization technical features, this study combined 18 literature involving the two types of features in predicting PICCs-DVT. A total of 21 variables used to describe the two types of features were summarized, and feature values were extracted from the data of 1,065 PICCs patients from January 1, 2021 to August 31, 2022, to construct a data sample set. Then, 70% of the sample set is used for model training and hyperparameter optimization, and 30% of the sample set is used for PICCs-DVT prediction and feature importance analysis of three common ML classification models (i.e. support vector classifier [SVC], random forest [RF], and artificial neural network [ANN]). In terms of prediction performance, this study selected four metrics to evaluate the prediction performance of the model: precision (P), recall (R), accuracy (ACC), and area under the curve (AUC). In terms of feature importance analysis, this study chooses a single feature analysis method based on the "input-output" sensitivity principle-Permutation Importance. For the mean model performance, the three ML models on the test set are P = 0.92, R = 0.95, ACC = 0.88, and AUC = 0.81. Specifically, the RF model is P = 0.95, R = 0.96, ACC = 0.92, AUC = 0.86; the ANN model is P = 0.92, R = 0.95, ACC = 0.88, AUC = 0.81; the SVC model is P = 0.88, R = 0.94, ACC = 0.85, AUC = 0.77. For feature importance analysis, Catheter-to-vein rate (RF: 91.55%, ANN: 82.25%, SVC: 87.71%), Zubrod-ECOG-WHO score (RF: 66.35%, ANN: 82.25%, SVC: 44.35%), and insertion attempt (RF: 44.35%, ANN: 37.65%, SVC: 65.80%) all occupy the top three in the ML models prediction task of PICCs-DVT, showing relatively consistent ranking results. The ML models show good performance in predicting PICCs-DVT and reveal a relatively consistent ranking of feature importance from the data. The important features revealed might help clinical medical staff to better understand and analyze the formation mechanism of PICCs-DVT from a data-driven perspective.