The rate of unplanned hospital readmissions following total hip arthroplasty (THA) varies from 3 to 10%, representing a major economic burden. However, it is unknown if specific factors are ...associated with different types of complications (ie, medical or orthopaedic-related) that lead to readmissions. Therefore, this study aimed to: (1) determine the overall, medical-related, and orthopaedic-related 90-day readmission rate; and (2) develop a predictive model for risk factors affecting overall, medical-related, and orthopaedic-related 90-day readmissions following THA.
A prospective cohort of primary unilateral THAs performed at a large tertiary academic center in the United States from 2016 to 2020 was included (n = 8,893 patients) using a validated institutional data collection system. Orthopaedic-related readmissions were specific complications affecting the prosthesis, joint, and surgical wound. Medical readmissions were due to any other cause requiring medical management. Multivariable logistic regression models were used to investigate associations between prespecified risk factors and 90-day readmissions, as well as medical and orthopaedic-related readmissions independently.
Overall, the rate of 90-day readmissions was 5.6%. Medical readmissions (4.2%) were found to be more prevalent than orthopaedic-related readmissions (1.4%). The area under the curve for the 90-day readmission model was 0.71 (95% confidence interval: 0.69 to 0.74). Factors significantly associated with medical-related readmissions were advanced age, Black race, education, Charlson Comorbidity Index, surgical approach, opioid overdose risk score, and nonhome discharge. In contrast, risk factors linked to orthopaedic-related readmissions encompassed body mass index, patient-reported outcome measure phenotype, nonosteoarthritis indication, opioid overdose risk, and nonhome discharge.
Of the overall 90-day readmissions following primary THA, 75% were due to medical-related complications. Our successful predictive model for complication-specific 90-day readmissions highlights how different risk factors may disproportionately influence medical versus orthopaedic-related readmissions, suggesting that patient-specific, tailored preventive measures could reduce postoperative readmissions in the current value-based health care setting.
Remote sensing based on Remote Piloted Aircraft Systems (RPAS) has proved valuable for monitoring agronomic parameters in precision agriculture. This research aimed to develop predictive models based ...on machine learning to estimate indirect nitrogen levels (Narea) and grain yield in irrigated rice. During the five phenological stages of cultivation, a Sequoia® camera aboard the Phantom 4® Pro platform acquired the multispectral images. In addition to the spectral bands, 11 vegetation indices were taken as predictors of the response variables (Narea and grain yield). Spearman's correlation coefficient (p) analyzed the ideal monitoring window and selected the model variables. The Multi-Layer Perceptron (MLP) algorithm adjusted the predictive models that had their performance evaluated in training and testing. The results obtained by the Spearman correlation indicate that the ideal window for monitoring rice by RPAS, for both response variables, occurs at the beginning of the reproduction phase (R1). MLP generated a more accurate model for Narea, demonstrated by Pearson's correlation between predicted and observed values (0.82 and 0.71) and mean absolute error (MAE) of 9.47 and 10.89. Grain yield models show good MLP at all stages and excellent accuracy. In this way, our results reinforce the excellent efficiency of the combination of remote sensing via RPAS and machine learning in applications aimed at precision agriculture, serving as a useful tool for managing production and evaluating grain yield in irrigated rice fields.
•Extensive database of building configurations using building information modeling (BIM).•Incorporating cooling load as a major factor in building design.•Performance comparison of five different ML ...algorithms.•Online interactive graphical user interface.•The CatBoost model outperformed other ML models.•SHAP analysis for the impact of different features on the predictions.
Since the cooling systems used in buildings in hot climates account for a significant portion of the energy consumption, it is very important for both economy and environment to accurately predict the cooling load and consider it in building designs. This study aimed to maximize energy efficiency by appropriately selecting the features of a building that affect its cooling load. To this end, data-driven, accurate, and accessible tools were developed that enable the prediction of the cooling load of a building by practitioners. The study involves simulating the energy consumption of a mid-rise, double-story terrace house in Malaysia using building information modeling (BIM) and estimating the cooling load using ensemble machine learning models and genetic programming. Categorical Boosting (CatBoost), eXtreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), and Random Forest (RF) models have been developed and made available as an online interactive graphical user interface on the Streamlit platform. Furthermore, the symbolic regression technique has been utilized to obtain a closed-form equation that predicts the cooling load. The dataset used for training the predictive models comprised 94,310 data points with 10 input variables and the cooling load as the output variable. Performance metrics such as the coefficient of determination (R2), root mean squared error (RMSE), and mean absolute error (MAE) were used to measure the predictive model performances. The results of the machine learning models indicated successful prediction, with the CatBoost model achieving the highest score (R2 = 0.9990) among the four ensemble models and the predictive equation. The SHAP analysis determined the aspect ratio of the building as the most impactful feature of the building.
A recent wave of research has attempted to define fairness quantitatively. In particular, this work has explored what fairness might mean in the context of decisions based on the predictions of ...statistical and machine learning models. The rapid growth of this new field has led to wildly inconsistent motivations, terminology, and notation, presenting a serious challenge for cataloging and comparing definitions. This article attempts to bring much-needed order. First, we explicate the various choices and assumptions made-often implicitly-to justify the use of prediction-based decision-making. Next, we show how such choices and assumptions can raise fairness concerns and we present a notationally consistent catalog of fairness definitions from the literature. In doing so, we offer a concise reference for thinking through the choices, assumptions, and fairness considerations of prediction-based decision-making.
Coronavirus disease 2019 (COVID-19) is sweeping the globe. Despite multiple case-series, actionable knowledge to tailor decision-making proactively is missing.
Can a statistical model accurately ...predict infection with COVID-19?
We developed a prospective registry of all patients tested for COVID-19 in Cleveland Clinic to create individualized risk prediction models. We focus here on the likelihood of a positive nasal or oropharyngeal COVID-19 test. A least absolute shrinkage and selection operator logistic regression algorithm was constructed that removed variables that were not contributing to the model’s cross-validated concordance index. After external validation in a temporally and geographically distinct cohort, the statistical prediction model was illustrated as a nomogram and deployed in an online risk calculator.
In the development cohort, 11,672 patients fulfilled study criteria, including 818 patients (7.0%) who tested positive for COVID-19; in the validation cohort, 2295 patients fulfilled criteria, including 290 patients who tested positive for COVID-19. Male, African American, older patients, and those with known COVID-19 exposure were at higher risk of being positive for COVID-19. Risk was reduced in those who had pneumococcal polysaccharide or influenza vaccine or who were on melatonin, paroxetine, or carvedilol. Our model had favorable discrimination (c-statistic = 0.863 in the development cohort and 0.840 in the validation cohort) and calibration. We present sensitivity, specificity, negative predictive value, and positive predictive value at different prediction cutoff points. The calculator is freely available at https://riskcalc.org/COVID19.
Prediction of a COVID-19 positive test is possible and could help direct health-care resources. We demonstrate relevance of age, race, sex, and socioeconomic characteristics in COVID-19 susceptibility and suggest a potential modifying role of certain common vaccinations and drugs that have been identified in drug-repurposing studies.
Predictive models are central to many scientific disciplines and vital for informing management in a rapidly changing world. However, limited understanding of the accuracy and precision of models ...transferred to novel conditions (their ‘transferability’) undermines confidence in their predictions. Here, 50 experts identified priority knowledge gaps which, if filled, will most improve model transfers. These are summarized into six technical and six fundamental challenges, which underlie the combined need to intensify research on the determinants of ecological predictability, including species traits and data quality, and develop best practices for transferring models. Of high importance is the identification of a widely applicable set of transferability metrics, with appropriate tools to quantify the sources and impacts of prediction uncertainty under novel conditions.
Models transferred to novel conditions could provide predictions in data-poor scenarios, contributing to more informed management decisions.
The determinants of ecological predictability are, however, still insufficiently understood.
Predictions from transferred ecological models are affected by species’ traits, sampling biases, biotic interactions, nonstationarity, and the degree of environmental dissimilarity between reference and target systems.
We synthesize six technical and six fundamental challenges that, if resolved, will catalyze practical and conceptual advances in model transfers.
We propose that the most immediate obstacle to improving understanding lies in the absence of a widely applicable set of metrics for assessing transferability, and that encouraging the development of models grounded in well-established mechanisms offers the most immediate way of improving transferability.
The advance rate (AR) of a tunnel boring machine (TBM) under hard rock conditions is a key parameter in the successful implementation of tunneling engineering. In this study, we improved the accuracy ...of prediction models by employing a hybrid model of extreme gradient boosting (XGBoost) with Bayesian optimization (BO) to model the TBM AR. To develop the proposed models, 1286 sets of data were collected from the Peng Selangor Raw Water Transfer tunnel project in Malaysia. The database consists of rock mass and intact rock features, including rock mass rating, rock quality designation, weathered zone, uniaxial compressive strength, and Brazilian tensile strength. Machine specifications, including revolution per minute and thrust force, were considered to predict the TBM AR. The accuracies of the predictive models were examined using the root mean squares error (RMSE) and the coefficient of determination (R2) between the observed and predicted yield by employing a five-fold cross-validation procedure. Results showed that the BO algorithm can capture better hyper-parameters for the XGBoost prediction model than can the default XGBoost model. The robustness and generalization of the BO-XGBoost model yielded prominent results with RMSE and R2 values of 0.0967 and 0.9806 (for the testing phase), respectively. The results demonstrated the merits of the proposed BO-XGBoost model. In addition, variable importance through mutual information tests was applied to interpret the XGBoost model and demonstrated that machine parameters have the greatest impact as compared to rock mass and material properties.
Black persons bear a disproportionate burden of peripheral artery disease (PAD) and experience higher rates of endovascular revascularization failure (ERF) when compared with non-Hispanic White ...persons. We aimed to identify predictors of ERF in Black persons using predictive modeling.
This retrospective study included all persons identifying as Black who underwent an initial endovascular revascularization procedure for PAD between 2011 and 2018 at a midwestern tertiary care center. Three predictive models were developed using (1) logistic regression, (2) penalized logistic regression (least absolute shrinkage and selection operator LASSO), and (3) random forest (RF). Predictive performance was evaluated under repeated cross-validation.
Of the 163 individuals included in the study, 113 (63.1%) experienced ERF at 1 y. Those with ERF had significant differences in symptom status (P < 0.001), lesion location (P < 0.001), diabetes status (P = 0.037), and annual procedural volume of the attending surgeon (P < 0.001). Logistic regression and LASSO models identified tissue loss, smoking, femoro-popliteal lesion location, and diabetes control as risk factors for ERF. The RF model identified annual procedural volume, age, PAD symptoms, number of comorbidities, and lesion location as most predictive variables. LASSO and RF models were more sensitive than logistic regression but less specific, although all three methods had an overall accuracy of ≥75%.
Black persons undergoing endovascular revascularization for PAD are at high risk of ERF, necessitating need for targeted intervention. Predictive models may be clinically useful for identifying high-risk patients, although individual predictors of ERF varied by model. Further exploration into these models may improve limb salvage for this population.
Antimicrobial resistance (AMR) continues to pose a grave threat to public health. The increase in the burden of AMR is fueled by the indiscriminate use of antimicrobial agents in agriculture. The ...objective of this study was to develop a genome-based machine learning model to predict AMR in Salmonella isolated from chicken meat. Genomic information on 205 Salmonella isolates from chicken meat was combined with data on the AMR phenotype of these isolates to amoxicillin-clavulanic acid, ampicillin, ceftiofur, ceftriaxone, sulfisoxazole, streptomycin, tetracycline, and cefoxitin. Four machine learning algorithms i.e., logit boost, random forest, support vector machine, and extreme gradient boosting were trained on this data to build models. The best-performing model for each antimicrobial was used to predict the AMR phenotypes of a new set of 200 Salmonella isolates also from chicken meat, and the predictions were compared to AMR phenotype predictions from ResFinder. The machine learning models showed high sensitivity (≥0.833), specificity (≥0.833), and balanced accuracy (≥0.866), across all the antimicrobials. The models predicted resistance prevalences ranging from 1% (ceftriaxone) to 65.5% (streptomycin). When the AMR phenotype predictions of the machine learning models were compared to predictions from ResFinder, the predictions from this study were accurate (>95%).
•Genome-based machine learning models accurately predicted AMR.•Combining AMR phenotype and WGS data is useful for AMR prediction studies.•Genome-based ML identified genes that are typically not used as markers of AMR.•This approach could aid in the identification of novel AMR genes.
There is increased interest in the use of prediction models to guide clinical decision-making in orthopedics. Prediction models are typically evaluated in terms of their accuracy: discrimination ...(area-under-the-curve AUC or concordance index) and calibration (a plot of predicted vs. observed risk). But it can be hard to know how high an AUC has to be in order to be “high enough” to warrant use of a prediction model, or how much miscalibration would be disqualifying. Decision curve analysis was developed as a method to determine whether use of a prediction model in the clinic to inform decision-making would do more good than harm. Here we give a brief introduction to decision curve analysis, explaining the critical concepts of net benefit and threshold probability. We briefly review some prediction models reported in the orthopedic literature, demonstrating how use of decision curves has allowed conclusions as to the clinical value of a prediction model. Conversely, papers without decision curves were unable to address questions of clinical value. We recommend increased use of decision curve analysis to evaluate prediction models in the orthopedics literature.