Regression Tree Model Research Articles

Investigating a Machine Learning Approach to Predicting White Pixel Defects in Wafers-A Case Study of Wafer Fabrication Plant F.

CMOS image sensor (CIS) semiconductor products are integral to mobile phones and photographic devices, necessitating ongoing enhancements in efficiency and quality for superior photographic outcomes. The presence of white pixels serves as a crucial metric for assessing CIS product performance, primarily arising from metal impurity contamination during the wafer production process or from defects introduced by the grinding blade process. While immediately addressing metal impurity contamination during production presents challenges, refining the handling of defects attributed to grinding blade processing can notably mitigate white pixel issues in CIS products. This study zeroes in on silicon wafer manufacturers in Taiwan, analyzing white pixel defects reported by customers and leveraging machine learning to pinpoint and predict key factors leading to white pixel defects from grinding blade operations. Such pioneering practical studies are rare. The findings reveal that the classification and regression tree (CART) and random forest (RF) models deliver the most accurate predictions (95.18%) of white pixel defects caused by grinding blade operations in a default parameter setting. The analysis further elucidates critical factors like grinding load and torque, vital for the genesis of white pixel defects. The insights garnered from this study aim to arm operators with proactive measures to diminish the potential for customer complaints.

Prediction and optimisation of gasoline quality in petroleum refining: The use of machine learning model as a surrogate in optimisation framework

AbstractHardware‐based sensing frameworks such as cooperative fuel research engines are conventionally used to monitor research octane number (RON) in the petroleum refining industry. Machine learning techniques are employed to predict the RON of integrated naphtha reforming and isomerisation processes. A dynamic Aspen HYSYS model was used to generate data by introducing artificial uncertainties in the range of ±5% in process conditions, such as temperature, flow rates, etc. The generated data was used to train support vector machines (SVM), Gaussian process regression (GPR), artificial neural networks (ANN), regression trees (RT), and ensemble trees (ET). Hyperparameter tuning was performed to enhance the prediction capabilities of GPR, ANN, SVM, ET and RT models. Performance analysis of the models indicates that GPR, ANN, and SVM with R2 values of 0.99, 0.978, and 0.979 and RMSE values of 0.108, 0.262, and 0.258, respectively performed better than the remaining models and had the prediction capability to capture the RON dependence on predictor variables. ET and RT had an R2 value of 0.94 and 0.89, respectively. The GPR model was used as a surrogate model for fitness function evaluations in two optimisation frameworks based on genetic algorithm and particle swarm method. Optimal parameter values found by the optimisation methodology increased the RON value by 3.52%. The proposed methodology of surrogate‐based optimisation will provide a platform for plant‐level implementation to realise the concept of industry 4.0 in the refinery.

Predicting the properties of perovskite materials by improved compositionally restricted attention-based networks and explainable machine learning

Understanding the unique properties of perovskite materials is crucial in advancing solar energy technologies. Factors like heat of formation and bandgap significantly influence the light absorption capability and stability of perovskite solar cells. However, it is time-consuming and labor-intensive to obtain the properties of perovskites using traditional experimental or high-throughput computational methods. As a prospective method, machine learning can find regularities in the given training data and give accurate prediction results. In this article, we use deep learning models based on attention mechanisms and elemental features to predict the heat of formation and bandgap of perovskite materials. Random Forest and Gradient Boosted Regression Tree models have also been used for interpretable predictions of properties. The compositionally restricted attention-based network was improved by introducing a densely connected network and optimizing the network structure to increase data processing capabilities. The experiment results show that the mean absolute errors of the heat of formation and bandgap on the test sets are decreased by 5.77% and 3.37% respectively. The optimized model also shows better performance when used for classification tasks. In addition, we use the gradient boosting regression tree model and the shapley additive explanations tool to conduct an interpretable analysis, explaining the impact of different features on the predictions of the properties.

Response mechanism of rodent burrow density to natural environmental factors in desert areas based on multisource data

Timely and accurate prediction and early warning of rodent infestation are highly important for the sustainable development of desert ecosystems. This study combines data from 1376 UAV images of actual rodent burrows with natural environmental factor data, aiming to reveal the mechanism through which these factors affect rodent burrow density in the desert areas. The relationships between rodent burrow density and the 9 plant communities and 34 plant species in this area were analyzed by the K-W test and Spearman rank correlation analysis, respectively. The Boruta algorithm and the boosted regression trees (BRT) model were used to determine the six main factors influencing rodent burrow density from a total of 32 factors and to determine the distribution pattern of the burrows. The results show that (1) the density of rodent burrows in the Junggar Basin desert area gradually decreases from south to north and from west to east. (2) There were significant differences in rodent burrow density across the different plant community types (P = 0.048**). A total of 50.0 % of the species significantly influenced rodent burrow density at at least one level. (3) Among the 17 factors with lag effects, nine had the most significant correlation with rodent burrow density in April. Only two natural environmental factors demonstrated lag effects of less than one month. (4) The soil clay content, elevation, Normalized Difference Salinity Index (NDSI), vegetation cover, precipitation, and soil bulk density were the most important influencing factors on rodent burrow density. Considering the pairwise interactions among factors, the collective impact of these factors on rodent burrow density is at least sevenfold greater than the independent effect of a single factor. This study provides a theoretical basis for obtaining a timely and accurate understanding of the spatial distribution of rodent burrows, mitigating the degradation trend of desert ecosystems, and ensuring healthy and sustainable development.

A tree-based automated machine learning approach of the obstructed view factor of thermal radiation in nuclear pebble beds

For the nuclear pebble bed of the high temperature gas-cooled reactor (HTGR), a tree-based automated machine learning approach is developed to discuss the complicated thermal radiation behaviors. The AutoML model for calculating the obstructed view factor between any two particles in the pebble bed includes the gradient boosting regression tree model with the fine-tuned hyperparameters and the analytical base model. Using the discrete packing of the HTR-PM nuclear reactor and the leaped Halton sequence, a large dataset of the view factor under different conditions is generated to train the AutoML model. On the same platform, numerical results show that AutoML model is approximately 5 × 105 times faster than the traditional approach. The Pareto front of the AutoML model indicates that the mean squared error decreases with the model complexity until it reaches the optimal solution. Then, the trained AutoML model is applied to the engineering pebble bed. It is shown that the average radiation exchange factor near the wall is lower than that in the bulk region. In addition, the packing height is also an important parameter for evaluating the radiative behaviors. In particular, when the height of the nuclear pebble bed is less than 20 times particle diameter, it is necessary to consider the contribution of the packing height to the radiation effective thermal conductivity.

An analysis of factors influencing cognitive dysfunction among older adults in Northwest China based on logistic regression and decision tree modelling

BackgroundCognitive dysfunction is one of the leading causes of disability and dependence in older adults and is a major economic burden on the public health system. The aim of this study was to investigate the risk factors for cognitive dysfunction and their predictive value in older adults in Northwest China.MethodsA cross-sectional study was conducted using a multistage sampling method. The questionnaires were distributed through the Elderly Disability Monitoring Platform to older adults aged 60 years and above in Northwest China, who were divided into cognitive dysfunction and normal cognitive function groups. In addition to univariate analyses, logistic regression and decision tree modelling were used to construct a model to identify factors that can predict the occurrence of cognitive dysfunction in older adults.ResultsA total of 12,494 valid questionnaires were collected, including 2617 from participants in the cognitive dysfunction group and 9877 from participants in the normal cognitive function group. Univariate analysis revealed that ethnicity, BMI, age, educational attainment, marital status, type of residence, residency status, current work status, main economic source, type of chronic disease, long-term use of medication, alcohol consumption, participation in social activities, exercise status, social support, total scores on the Balanced Test Assessment, total scores on the Gait Speed Assessment total score, and activities of daily living (ADL) were significantly different between the two groups (all P < 0.05). According to logistic regression analyses, ethnicity, BMI, educational attainment, marital status, residency, main source of income, chronic diseases, annual medical examination, alcohol consumption, exercise status, total scores on the Balanced Test Assessment, and activities of daily living (ADLs) were found to influence cognitive dysfunction in older adults (all P < 0.05). In the decision tree model, the ability to perform activities of daily living was the root node, followed by total scores on the Balanced Test Assessment, marital status, educational attainment, age, annual medical examination, and ethnicity.ConclusionsTraditional risk factors (including BMI, literacy, and alcohol consumption) and potentially modifiable risk factors (including balance function, ability to care for oneself in daily life, and widowhood) have a significant impact on the increased risk of cognitive dysfunction in older adults in Northwest China. The use of decision tree models can help health care workers better assess cognitive function in older adults and develop personalized interventions. Further research could help to gain insight into the mechanisms of cognitive dysfunction and provide new avenues for prevention and intervention.

Magnetic resonance imaging-based prediction models for differentiating intraspinal schwannomas from meningiomas: classification and regression tree and random forest analysis.

Due to the variations in surgical approaches and prognosis between intraspinal schwannomas and meningiomas, it is crucial to accurately differentiate between the two prior to surgery. Currently, there is limited research exploring the implementation of machine learning (ML) methods for distinguishing between these two types of tumors. This study aimed to establish a classification and regression tree (CART) model and a random forest (RF) model for distinguishing schwannomas from meningiomas. We retrospectively collected 88 schwannomas (52 males and 36 females) and 51 meningiomas (10 males and 41 females) who underwent magnetic resonance imaging (MRI) examinations prior to the surgery. Simple clinical data and MRI imaging features, including age, sex, tumor location and size, T1-weighted images (T1WI) and T2-weighted images (T2WI) signal characteristics, degree and pattern of enhancement, dural tail sign, ginkgo leaf sign, and intervertebral foramen widening (IFW), were reviewed. Finally, a CART model and RF model were established based on the aforementioned features to evaluate their effectiveness in differentiating between the two types of tumors. Meanwhile, we also compared the performance of the ML models to the radiologists. The receiver operating characteristic (ROC) curve, accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were used to evaluate the models and clinicians' discrimination performance. Our investigation reveals significant variations in ten out of 11 variables in the training group and five out of 11 variables in the test group when comparing schwannomas and meningiomas (P<0.05). Ultimately, the CART model incorporated five variables: enhancement pattern, the presence of IFW, tumor location, maximum diameter, and T2WI signal intensity (SI). The RF model combined all 11 variables. The CART model, RF model, radiologist 1, and radiologist 2 achieved an area under the curve (AUC) of 0.890, 0.956, 0.681, and 0.723 in the training group, and 0.838, 0.922, 0.580, and 0.659 in the test group, respectively. The RF prediction model exhibits more exceptional performance than an experienced radiologist in discriminating intraspinal schwannomas from meningiomas. The RF model seems to be better in discriminating the two tumors than the CART model.

Variable Selection Using Bayesian Additive Regression Trees.

Variable selection is an important statistical problem. This problem becomes more challenging when the candidate predictors are of mixed type (e.g. continuous and binary) and impact the response variable in nonlinear and/or non-additive ways. In this paper, we review existing variable selection approaches for the Bayesian additive regression trees (BART) model, a nonparametric regression model, which is flexible enough to capture the interactions between predictors and nonlinear relationships with the response. An emphasis of this review is on the ability to identify relevant predictors. We also propose two variable importance measures which can be used in a permutation-based variable selection approach, and a backward variable selection procedure for BART. We introduce these variations as a way of illustrating limitations and opportunities for improving current approaches and assess these via simulations.

The Use of Boosted Regression Trees to Predict the Occurrence and Quantity of Staphylococcus aureus in Recreational Marine Waterways

Microbial monitoring in marine recreational waterways often overlooks environmental variables associated with pathogen occurrence. This study employs a predictive boosted regression trees (BRT) model to predict Staphylococcus aureus abundance in the Tampa Bay estuary and identify related environmental variables associated with the microbial pathogen’s occurrence. We provide evidence that the BRT model’s adaptability and ability to capture complex interactions among predictors make it invaluable for research on microbial indicator research. Over 18 months, water samples from 7 recreational sites underwent microbial quantitation and S. aureus isolation, followed by genetic validation. BRT analysis of S. aureus occurrence and environmental variables revealed month, precipitation, salinity, site, temperature, and year as relevant predictors. In addition, the BRT model accurately predicted S. aureus occurrence, setting a precedent for pathogen–environment research. The approach described here is novel and informs proactive management strategies and community health initiatives in marine recreational waterways.

Big data resolving using Apache Spark for load forecasting and demand response in smart grid: a case study of Low Carbon London Project

Using recent information and communication technologies for monitoring and management initiates a revolution in the smart grid. These technologies generate massive data that can only be processed using big data tools. This paper emphasizes the role of big data in resolving load forecasting, renewable energy sources integration, and demand response as significant aspects of smart grids. Meters data from the Low Carbon London Project is investigated as a case study. Because of the immense stream of meters' readings and exogenous data added to load forecasting models, addressing the problem is in the context of big data. Descriptive analytics are developed using Spark SQL to get insights regarding household energy consumption. Spark MLlib is utilized for predictive analytics by building scalable machine learning models accommodating meters' data streams. Multivariate polynomial regression and decision tree models are preferred here based on the big data point of view and the literature that ensures they are accurate and interpretable. The results confirmed the descriptive analytics and data visualization capabilities to provide valuable insights, guide the feature selection process, and enhance load forecasting models' accuracy. Accordingly, proper evaluation of demand response programs and integration of renewable energy resources is accomplished using achieved load forecasting results.

Development and validation of a risk prediction model and prediction tools for post-thrombotic syndrome in patients with lower limb deep vein thrombosis

PurposeOur research aims to compare the predictive performance of decision tree algorithms (DT) and logistic regression analysis (LR) in constructing models, and develop a Post-Thrombotic Syndrome (PTS) risk stratification tool. MethodsWe retrospectively collected and analyzed relevant case information of 618 patients diagnosed with DVT from January 2012 to December 2021 in three different tertiary hospitals in Jiangxi Province as the modeling group. Additionally, we used the case information of 212 patients diagnosed with DVT from January 2022 to January 2023 in two tertiary hospitals in Hubei Province and Guangdong Province as the validation group. We extracted electronic medical record information including general patient data, medical history, laboratory test indicators, and treatment data for analysis. We established DT and LR models and compared their predictive performance using receiver operating characteristic (ROC) curves and confusion matrices. Internal and external validations were conducted. Additionally, we utilized LR to generate nomogram charts, calibration curves, and decision curves analysis (DCA) to assess its predictive accuracy. ResultsBoth DT and LR models indicate that Year, Residence, Cancer, Varicose Vein Operation History, DM, and Chronic VTE are risk factors for PTS occurrence. In internal validation, DT outperforms LR (0.962 vs 0.925, z = 3.379, P < 0.001). However, in external validation, there is no significant difference in the area under the ROC curve between the two models (0.963 vs 0.949, z = 0.412, P = 0.680). The validation results of calibration curves and DCA demonstrate that LR exhibits good predictive accuracy and clinical effectiveness. A web-based calculator software of nomogram (https://sunxiaoxuan.shinyapps.io/dynnomapp/) was utilized to visualize the logistic regression model. ConclusionsThe combination of decision tree and logistic regression models, along with the web-based calculator software of nomogram, can assist healthcare professionals in accurately assessing the risk of PTS occurrence in individual patients with lower limb DVT.

A lasso and random forest model using flow cytometry data identifies primary myelofibrosis.

Thrombocythemia (ET), polycythemia vera (PV), primary myelofibrosis (PMF), prefibrotic/early (pre-PMF), and overt fibrotic PMF (overt PMF) are classical Philadelphia-Negative (Ph-negative) myeloproliferative neoplasms (MPNs). Differentiating between these types based on morphology and molecular markers is challenging. This study aims to clarify the application of flow cytometry in the diagnosis and differential diagnosis of classical MPNs. This study retrospectively analyzed the immunophenotypes, clinical characteristics, and laboratory findings of 211 Ph-negative MPN patients, including ET, PV, pre-PMF, overt PMF, and 47 controls. Compared to ET and PV, PMF differed in white blood cells, hemoglobin, blast cells in the peripheral blood, abnormal karyotype, and WT1 gene expression. PMF also differed from controls in CD34+ cells, granulocyte phenotype, monocyte phenotype, percentage of plasma cells, and dendritic cells. Notably, the PMF group had a significantly lower plasma cell percentage compared with other groups. A lasso and random forest model select five variables (CD34+CD19+cells and CD34+CD38- cells on CD34+cells, CD13dim+CD11b- cells in granulocytes, CD38str+CD19+/-plasma, and CD123+HLA-DR-basophils), which identify PMF with a sensitivity and specificity of 90%. Simultaneously, a classification and regression tree model was constructed using the percentage of CD34+CD38- on CD34+ cells and platelet counts to distinguish between ET and pre-PMF, with accuracies of 94.3% and 83.9%, respectively. Flow immunophenotyping aids in diagnosing PMF and differentiating between ET and PV. It also helps distinguish pre-PMF from ET and guides treatment decisions.

A tree-based extended model to predict intention to speed for taxi drivers

The need to predict and understand drivers’ intention to violate emphasizes the importance of developing a model based on the motives behind the action. This study aims to develop a model to predict a taxi driveŕs intention to speed in an urban area using variables including psychological factors, demographic information, and exposure. Aggregate (factor-based) and disaggregate (item-based) models of input variables will also be compared. In this study, the self-reported data on Tehran taxi drivers was collected in the form of a scenario-based questionnaire which is inspired by the Theory of Planned Behaviour (TPB). The intention to commit violations was predicted and modelled using stepwise regression and decision tree regression models. The results, based on the stepwise regression model showed a higher value of fitting score for an item-based linear model (R2 = 0.695), but the final model using the tree-based regression presented a better fit for the factored model (R2 = 0.746) considering the cross-validated model with the lowest deviation criteria (RMSD = 0.26). Furthermore, in this model attitude made the greatest contribution to the prediction of intention. The results obtained from the two different modelling approaches showed that exposure and age did not have much effect on the models. A main advantage of the current study is that the variables were compared and refined in two stages of exploratory factor analysis and stepwise regression before being entered into the tree regression model. In addition, the results of the factor-based model were compared with the item-based model.

Spatial coupling relationship between architectural landscape characteristics and urban heat island in different urban functional zones

The process of urbanization exacerbates the urban heat island phenomenon. In order to reasonably identify urban functional zones, and reveal the influence of buildings on land surface temperature (LST). Guiyang, a multi-mountainous city in China, was taken as the research area, Open Street Map (OSM) and land use type data were used to identify urban functional zones. Subsequently, the correlation between urban building morphology metrics and LST was analyzed using Pearson correlation and boosted regression tree (BRT) model. The results showed that: (1) There were significant differences in LST among different urban functional zones. (2) The relative contribution value of building density (BD) to land surface temperature was high, and it significantly exacerbated the urban heat island effect across all urban functional zones, with the maximum impact on LST reaching 2.5 °C in Utility zone. (3) Mean building height (MBH), Southerly wind speed (S_Wind speed) and Landscape shape index of building (LSI_Building) had the effect of alleviating thermal environment, and the mitigation degree of land surface temperature was the highest in Residential zone and Ecological zone, reaching 0.6 °C, 0.8 °C and 2.0 °C, respectively. This study enhances our understanding of the relationship between urban functions and LST, providing valuable insights for mitigating the urban thermal environment and promoting sustainable development.

Modeling and Estimation of Reference Evapotranspiration using Machine Learning Algorithms: A Comparative Performance Analysis

Fresh, clean water is necessary for human health. Currently, the agriculture sector uses the majority of freshwater for irrigation without using planning or optimization techniques. Evapotranspiration, which may have a major impact in planning water supply management and crop yield improvement, is an element of the hydrological cycle. Accurate anticipation of reference evapotranspiration (ETO) is an intricate job due to its nonlinear behavior. Machine learning approach based model may be an intelligent tool to predict the accurate ETO. This study investigates and compares the predictive skills of three regression based supervised learning algorithms: decision tree (dtr), and random forest (rfr), and k-nearest-neighbors (knnr) along with tuning their hyper-parameters like how many neighbors there are in knnr, minimum samples in dtr at a leaf node and quantity of trees in the rfr scenario to forecast ETO. Every model's performance is quantified on four different groups of meteorological parameters. Groups are created based on close correlation of meteorological parameters with ETo. In this investigation, analysis is carried out on daily meteorological information of New Delhi, India for the periods from 2000 to 2021. The predicted results of the knnr, dtr and rfr models on four groups of meteorological inputs (twelve different models) are compared with ETO obtained from the FAO-PM56 equations. The study's conclusions show that the k-nearest-neighbors and random forest regression-based models outperform the decision tree regression models concerning performance. The finest performance noted by knnr and rfr models with r2 (coefficient of determination) of 0.99 and rmse of 0.21 and 0.22 mm/day respectively whereas dtr model noted r2 of 0.98 and rmse of 0.40 mm/day. Therefor these models may provide scientists, engineers, and farmers with more potent choices for managing water resources and scheduling irrigation.

Data Visualization and Machine Learning Trend Analysis Forecasting Based on Roblox Stock Data

The stock market, as a highly complex, unstable and dynamically changing system, is affected by a wide range of factors, such as political, economic, social and technological, in terms of price changes. Therefore, stock price prediction is a challenging problem, but also a problem of great importance. In order to provide a more accurate and reliable method for stock closing price prediction, this paper aims to reduce investment risk and improve investment returns by comparing the prediction effects of different regression models for investors to make investment strategies and decisions. In this paper, Open, High, Low, and Adj Close are used as input parameters, Close is used as the target parameter, the final closing price of the stock is determined from the various stock data, the training set, validation set, and test set are divided in accordance with 6:2:2, and the predictive effect of the model is evaluated using the parameters of MSE, RMSE, MAE, MAPE, and R2. After the comparison of multiple machine learning algorithms, XGBoost regression model and Random Forest Regression model perform the best in the prediction of stock closing price, and their MSEs are 0.061 and 0.072 respectively; followed by LightGBM regression model and decision tree regression model, the MSEs of the two models are 0.113 and 0.299 respectively, but the prediction results of BP neural network regression are poor. By comparing the prediction results of different regression models, it can provide investors with a more accurate and reliable method to predict the closing price of stocks, help investors better formulate investment strategies and decisions, reduce investment risks and improve investment returns. This is of great significance to both individual and institutional investors.

Responses of the coral reef cryptobiome to environmental gradients in the Red Sea.

An essential component of the coral reef animal diversity is the species hidden in crevices within the reef matrix, referred to as the cryptobiome. These organisms play an important role in nutrient cycling and provide an abundant food source for higher trophic levels, yet they have been largely overlooked. Here, we analyzed the distribution patterns of the mobile cryptobiome (>2000 μm) along the latitudinal gradient of the Saudi Arabian coast of the Red Sea. Analysis was conducted based on 54 Autonomous Reef Monitoring Structures. We retrieved a total of 5273 organisms, from which 2583 DNA sequences from the mitochondrially encoded cytochrome c oxidase I were generated through sanger sequencing. We found that the cryptobiome community is variable over short geographical distances within the basin. Regression tree models identified sea surface temperature (SST), percentage cover of hard coral and turf algae as determinant for the number of operational taxonomic units present per Autonomous Reef Monitoring Structures (ARMS). Our results also show that the community structure of the cryptobiome is associated with the energy available (measured as photosynthetic active radiation), sea surface temperature, and nearby reef habitat characteristics (namely hard corals, turf and macroalgae). Given that temperature and reef benthic characteristics affect the cryptobiome, current scenarios of intensive climate change are likely to modify this fundamental biological component of coral reef functioning. However, the trajectory of change is unknow and can be site specific, as for example, diversity is expected to increase above SST of 28.5°C, and with decreasing hard coral and turf cover. This study provides a baseline of the cryptobenthic community prior to major coastal developments in the Red Sea to be used for future biodiversity studies and monitoring projects. It can also contribute to better understand patterns of reef biodiversity in a period where Marine Protected Areas are being discussed in the region.

A new approach from public behavioral attitudes and perceptions towards microplastics: Influencing factors, and policy proposals

This research paper addresses the urgent environmental concern of microplastic (MP) emissions, focusing on the behavioral attitudes and perceptions of the general populace in Shyamnagar Upazila, Bangladesh. Against the backdrop of escalating MP pollution globally, this study investigates the level of awareness and the factors influencing public engagement in mitigating MP prevalence. Leveraging survey data from 350 respondents, the ordered logistic regression (OLR) and boosted regression tree (BRT) models are employed for comprehensive data analysis. The findings expose a concerning lack of awareness about MPs, as only 12% of respondents possessed prior knowledge, and a notable 63% remained uninformed about MP pollution. The OLR model reveals a positive correlation between heightened awareness of MPs and an increased willingness to take action. Gender differences become evident, with women exhibiting greater willingness than men to mitigate MP emissions, and environmental practitioners displaying heightened motivation. The BRT model underscores construction materials and industrial pollution as the primary influential factors amplifying MP pollution. These insights not only illuminate the existing scenario but also provide a basis for fostering favorable behavioral attitudes and perceptions to mitigate the prevalence of MPs within the coastal milieu.

Demonstrating the utility of Instrumented Gait Analysis in the treatment of children with cerebral palsy.

Instrumented gait analysis (IGA) has been around for a long time but has never been shown to be useful for improving patient outcomes. In this study we demonstrate the potential utility of IGA by showing that machine learning models are better able to estimate treatment outcomes when they include both IGA and clinical (CLI) features compared to when they include CLI features alone. We carried out a retrospective analysis of data from ambulatory children diagnosed with cerebral palsy who were seen at least twice at our gait analysis center. Individuals underwent a variety of treatments (including no treatment) between sequential gait analyses. We fit Bayesian Additive Regression Tree (BART) models that estimated outcomes for mean stance foot progression to demonstrate the approach. We built two models: one using CLI features only, and one using CLI and IGA features. We then compared the models' performance in detail. We performed similar, but less detailed, analyses for a number of other outcomes. All results were based on independent test data from a 70%/30% training/testing split. The IGA model was more accurate than the CLI model for mean stance-phase foot progression outcomes (RMSEIGA = 11∘, RMSECLI = 13∘) and explained more than 1.5 × as much of the variance (R2IGA = .45, R2CLI = .28). The IGA model outperformed the CLI model for every level of treatment complexity, as measured by number of simultaneous surgeries. The IGA model also exhibited superior performance for estimating outcomes of mean stance-phase knee flexion, mean stance-phase ankle dorsiflexion, maximum swing-phase knee flexion, gait deviation index (GDI), and dimensionless speed. The results show that IGA has the potential to be useful in the treatment planning process for ambulatory children diagnosed with cerebral palsy. We propose that the results of machine learning outcome estimators-including estimates of uncertainty-become the primary IGA tool utilized in the clinical process, complementing the standard medical practice of conducting a through patient history and physical exam, eliciting patient goals, reviewing relevant imaging data, and so on.

Predicting post-treatment symptom severity for adults receiving psychological therapy in routine care for generalised anxiety disorder: a machine learning approach

Approximately half of generalised anxiety disorder (GAD) patients do not recover from first-line treatments, and no validated prediction models exist to inform individuals or clinicians of potential treatment benefits. This study aimed to develop and validate an accurate and explainable prediction model of post-treatment GAD symptom severity. Data from adults receiving treatment for GAD in eight Improving Access to Psychological Therapies (IAPT) services (n=15,859) were separated into training, validation and holdout datasets. Thirteen machine learning algorithms were compared using 10-fold cross-validation, against two simple clinically relevant comparison models. The best-performing model was tested on the holdout dataset and model-specific explainability measures identified the most important predictors. A Bayesian Additive Regression Trees model out-performed all comparison models (MSE=16.54 [95 % CI=15.58; 17.51]; MAE=3.19; R²=0.33, including a single predictor linear regression model: MSE=20.70 [95 % CI=19.58; 21.82]; MAE=3.94; R²=0.14). The five most important predictors were: PHQ-9 anhedonia, GAD-7 annoyance/irritability, restlessness and fear items, then the referral-assessment waiting time. The best-performing model accurately predicted post-treatment GAD symptom severity using only pre-treatment data, outperforming comparison models that approximated clinical judgement and remaining within the GAD-7 error of measurement and minimal clinically important differences. This model could inform treatment decision-making and provide desired information to clinicians and patients receiving treatment for GAD.