Development Of Learning Model Research Articles

Advancing hydrogen storage predictions in metal-organic frameworks: A comparative study of LightGBM and random forest models with data enhancement

The escalating consumption of fossil fuels has given rise to a substantial upsurge in greenhouse gas concentrations and global temperatures, which, in turn, has triggered severe climate-related consequences. The critical imperative to reduce CO2 emissions and combat global warming has spurred extensive investigations into clean energy alternatives, with hydrogen emerging as a compelling zero-emission energy source. As a pivotal component of clean energy strategies, hydrogen requires designing compact, lightweight, and efficient storage systems. This study focuses on the development and evaluation of machine learning models for predicting the efficiency of Metal-Organic Frameworks (MOFs) in hydrogen storage, a key aspect of advancing clean energy technologies. MOFs, a class of nanoporous materials, show remarkable potential for hydrogen storage due to their high surface area and porosity. However, selecting the most suitable MOF for this application from a vast array of possible structures is a daunting task. In this context, machine learning algorithms offer an efficient alternative for predicting MOF suitability by considering their structural and chemical properties. We used ensemble learning methods, specifically Light Gradient Boosting Machine (LightGBM) and Random Forest (RF), to predict hydrogen uptake of MOFs based on a dataset of 219 experimentally tested samples. Two modeling scenarios were considered: one using the entire dataset, and the other involving strategic data pre-processing, including outlier removal and feature engineering. The results demonstrate that the measures taken to refine the dataset significantly enhance the predictive performance of the developed models, reducing prediction errors and improving overall goodness of fit. Specifically, the Mean Absolute Error (MAE) values for both the LightGBM and random forest models were reduced from 0.48 and 0.94, respectively, to 0.16 for both models, and the coefficients of determination (R2) increased substantially from 0.84 and 0.72 to 0.95, in both cases. Moreover, feature importance analysis unveiled that pressure-related features make the most significant contributions to the formation of tree ensembles during the model training process. A parametric sensitivity analysis was conducted revealing that H2 uptake in MOFs is most sensitive to changes in adsorption enthalpy, followed by surface area and temperature, while showing lower sensitivity to variations in pressure, consistent with established literature. These results underscore the pivotal role of data enhancement methods in refining machine learning models and can be instrumental in accelerating the development and optimization of MOF materials for clean energy applications.

Elevating Forestry Prediction: A Study on Machine Learning Model for Plantations Survival Rate Analysis

Abstract: This paper details the development and preliminary findings of a machine learning model designed to predict the survival rate of plantations. Drawing data from official sources, various vegetation indices were used as features for the predictive model. Initial results show potential, despite certain limitations, suggesting avenues for further enhancement and application.

Development of a long noncoding RNA-based machine learning model to predict COVID-19 in-hospital mortality

Tools for predicting COVID-19 outcomes enable personalized healthcare, potentially easing the disease burden. This collaborative study by 15 institutions across Europe aimed to develop a machine learning model for predicting the risk of in-hospital mortality post-SARS-CoV-2 infection. Blood samples and clinical data from 1286 COVID-19 patients collected from 2020 to 2023 across four cohorts in Europe and Canada were analyzed, with 2906 long non-coding RNAs profiled using targeted sequencing. From a discovery cohort combining three European cohorts and 804 patients, age and the long non-coding RNA LEF1-AS1 were identified as predictive features, yielding an AUC of 0.83 (95% CI 0.82–0.84) and a balanced accuracy of 0.78 (95% CI 0.77–0.79) with a feedforward neural network classifier. Validation in an independent Canadian cohort of 482 patients showed consistent performance. Cox regression analysis indicated that higher levels of LEF1-AS1 correlated with reduced mortality risk (age-adjusted hazard ratio 0.54, 95% CI 0.40–0.74). Quantitative PCR validated LEF1-AS1’s adaptability to be measured in hospital settings. Here, we demonstrate a promising predictive model for enhancing COVID-19 patient management.

Development and preliminary assessment of a machine learning model to predict myocardial infarction and cardiac arrest after major operations

IntroductionAccurate prediction of complications often informs shared decision-making. Derived over 10 years ago to enhance prediction of intra/post-operative myocardial infarction and cardiac arrest (MI/CA), the Gupta score has been criticized for unreliable calibration and inclusion of a wide spectrum of unrelated operations. In the present study, we developed a novel machine learning (ML) model to estimate perioperative risk of MI/CA and compared it to the Gupta score. MethodsPatients undergoing major operations were identified from the 2016–2020 ACS-NSQIP. The Gupta score was calculated for each patient, and a novel ML model was developed to predict MI/CA using ACS NSQIP-provided data fields as covariates. Discrimination (C-statistic) and calibration (Brier score) of the ML model were compared to the existing Gupta score within the entire cohort and across operative subgroups. ResultsOf 2,473,487 patients included for analysis, 25,177 (1.0%) experienced MI/CA (55.2% MI, 39.1% CA, 5.6% MI and CA). The ML model, which was fit using a randomly selected training cohort, exhibited higher discrimination within the testing dataset compared to the Gupta score (C-statistic 0.84 vs 0.80, p < 0.001). Furthermore, the ML model had significantly better calibration in the entire cohort (Brier score 0.0097 vs 0.0100). Model performance was markedly improved among patients undergoing thoracic, aortic, peripheral vascular and foregut surgery. ConclusionsThe present ML model outperformed the Gupta score in the prognostication of MI/CA across a heterogenous range of operations. Given the growing integration of ML into healthcare, such models may be readily incorporated into clinical practice and guide benchmarking efforts.

Development of an interpretable machine learning model associated with genetic indicators to identify Yin-deficiency constitution

BackgroundTraditional Chinese Medicine (TCM) defines constitutions which are relevant to corresponding diseases among people. As one of the common constitutions, Yin-deficiency constitution influences a number of Chinese population in the disease onset. Therefore, accurate Yin-deficiency constitution identification is significant for disease prevention and treatment.MethodsIn this study, we collected participants with Yin-deficiency constitution and balanced constitution, separately. The least absolute shrinkage and selection operator (LASSO) and logistic regression were used to analyze genetic predictors. Four machine learning models for Yin-deficiency constitution classification with multiple combined genetic indicators were integrated to analyze and identify the optimal model and features. The Shapley Additive exPlanations (SHAP) interpretation was developed for model explanation.ResultsThe results showed that, NFKBIA, BCL2A1 and CCL4 were the most associated genetic indicators with Yin-deficiency constitution. Random forest with three genetic predictors including NFKBIA, BCL2A1 and CCL4 was the optimal model, area under curve (AUC): 0.937 (95% CI 0.844–1.000), sensitivity: 0.870, specificity: 0.900. The SHAP method provided an intuitive explanation of risk leading to individual predictions.ConclusionWe constructed a Yin-deficiency constitution classification model based on machine learning and explained it with the SHAP method, providing an objective Yin-deficiency constitution identification system in TCM and the guidance for clinicians.

Development of interpretable machine learning models to predict in-hospital prognosis of acute heart failure patients.

In recent years, there has been remarkable development in machine learning (ML) models, showing a trend towards high prediction performance. ML models with high prediction performance often become structurally complex and are frequently perceived as black boxes, hindering intuitive interpretation of the prediction results. We aimed to develop ML models with high prediction performance, interpretability, and superior risk stratification to predict in-hospital mortality and worsening heart failure (WHF) in patients with acute heart failure (AHF). Based on the Kyoto Congestive Heart Failure registry, which enrolled 4056 patients with AHF, we developed prediction models for in-hospital mortality and WHF using information obtained on the first day of admission (demographics, physical examination, blood test results, etc.). After excluding 16 patients who died on the first or second day of admission, the original dataset (n=4040) was split 4:1 into training (n=3232) and test datasets (n=808). Based on the training dataset, we developed three types of prediction models: (i) the classification and regression trees (CART) model; (ii) the random forest (RF) model; and (iii) the extreme gradient boosting (XGBoost) model. The performance of each model was evaluated using the test dataset, based on metrics including sensitivity, specificity, area under the receiver operating characteristic curve (AUC), Brier score, and calibration slope. For the complex structure of the XGBoost model, we performed SHapley Additive exPlanations (SHAP) analysis, classifying patients into interpretable clusters. In the original dataset, the proportion of females was 44.8% (1809/4040), and the average age was 77.9±12.0. The in-hospital mortality rate was 6.3% (255/4040) and the WHF rate was 22.3% (900/4040) in the total study population. In the in-hospital mortality prediction, the AUC for the XGBoost model was 0.816 [95% confidence interval (CI): 0.815-0.818], surpassing the AUC values for the CART model (0.683, 95% CI: 0.680-0.685) and the RF model (0.755, 95% CI: 0.753-0.757). Similarly, in the WHF prediction, the AUC for the XGBoost model was 0.766 (95% CI: 0.765-0.768), outperforming the AUC values for the CART model (0.688, 95% CI: 0.686-0.689) and the RF model (0.713, 95% CI: 0.711-0.714). In the XGBoost model, interpretable clusters were formed, and the rates of in-hospital mortality and WHF were similar among each cluster in both the training and test datasets. The XGBoost models with SHAP analysis provide high prediction performance, interpretability, and reproducible risk stratification for in-hospital mortality and WHF for patients with AHF.

Weakly Supervised Learning Method for Semantic Segmentation of Large-Scale 3D Point Cloud Based on Transformers

Abstract. Nowadays, semantic segmentation results of 3D point cloud have been widely applied in the fields of robotics, autonomous driving, and augmented reality etc. Thanks to the development of relevant deep learning models (such as PointNet), supervised training methods have become hotspot, in which two common limitations exists: inferior feature representation of 3D points and massive annotations. To improve 3D point feature, inspired by the idea of transformer, we employ a so-call LCP network that extracts better feature by investigating attentions between target 3D points and its corresponding local neighbors via local context propagation. Training transformer-based network needs amount of training samples, which itself is a labor-intensive, costly and error-prone work, therefore, this work proposes a weakly supervised framework, in particular, pseudo-labels are estimated based on the feature distances between unlabeled points and prototypes, which are calculated based on labeled data. The extensive experimental results show that, the proposed PL-LCP can yield considerable results (67.6% mIOU for indoor and 67.3% for outdoor) even if only using 1% real labels, and comparing to several state-of-the-art method using all labels, we achieve superior results in mIOU, OA for indoor (65.9%, 89.2%).

Data-Driven Strategies for Complex System Forecasts: The Role of Textual Big Data and State-Space Transformers in Decision Support

In this research, an innovative state space-based Transformer model is proposed to address the challenges of complex system prediction tasks. By integrating state space theory, the model aims to enhance the capability to capture dynamic changes in complex data, thereby improving the accuracy and robustness of prediction tasks. Extensive experimental validations were conducted on three representative tasks, including legal case judgment, legal case translation, and financial data analysis to assess the performance and application potential of the model. The experimental results demonstrate significant performance improvements of the proposed model over traditional Transformer models and other advanced variants such as Bidirectional Encoder Representation from Transformers (BERT) and Finsformer across all evaluated tasks. Specifically, in the task of legal case judgment, the proposed model exhibited a precision of 0.93, a recall of 0.90, and an accuracy of 0.91, significantly surpassing the traditional Transformer model (with precision of 0.78, recall of 0.73, accuracy of 0.76) and performances of other comparative models. In the task of legal case translation, the precision of the proposed model reached 0.95, with a recall of 0.91 and an accuracy of 0.93, also outperforming other models. Likewise, in the task of financial data analysis, the proposed model also demonstrated excellent performance, with a precision of 0.94, recall of 0.90, and accuracy of 0.92. The state space-based Transformer model proposed not only theoretically expands the research boundaries of deep learning models in complex system prediction but also validates its efficiency and broad application prospects through experiments. These achievements provide new insights and directions for future research and development of deep learning models, especially in tasks requiring the understanding and prediction of complex system dynamics.

Development and Validation of an Explainable Deep Learning Model to Predict In-Hospital Mortality for Patients With Acute Myocardial Infarction: Algorithm Development and Validation Study.

Acute myocardial infarction (AMI) is one of the most severe cardiovascular diseases and is associated with a high risk of in-hospital mortality. However, the current deep learning models for in-hospital mortality prediction lack interpretability. This study aims to establish an explainable deep learning model to provide individualized in-hospital mortality prediction and risk factor assessment for patients with AMI. In this retrospective multicenter study, we used data for consecutive patients hospitalized with AMI from the Chongqing University Central Hospital between July 2016 and December 2022 and the Electronic Intensive Care Unit Collaborative Research Database. These patients were randomly divided into training (7668/10,955, 70%) and internal test (3287/10,955, 30%) data sets. In addition, data of patients with AMI from the Medical Information Mart for Intensive Care database were used for external validation. Deep learning models were used to predict in-hospital mortality in patients with AMI, and they were compared with linear and tree-based models. The Shapley Additive Explanations method was used to explain the model with the highest area under the receiver operating characteristic curve in both the internal test and external validation data sets to quantify and visualize the features that drive predictions. A total of 10,955 patients with AMI who were admitted to Chongqing University Central Hospital or included in the Electronic Intensive Care Unit Collaborative Research Database were randomly divided into a training data set of 7668 (70%) patients and an internal test data set of 3287 (30%) patients. A total of 9355 patients from the Medical Information Mart for Intensive Care database were included for independent external validation. In-hospital mortality occurred in 8.74% (670/7668), 8.73% (287/3287), and 9.12% (853/9355) of the patients in the training, internal test, and external validation cohorts, respectively. The Self-Attention and Intersample Attention Transformer model performed best in both the internal test data set and the external validation data set among the 9 prediction models, with the highest area under the receiver operating characteristic curve of 0.86 (95% CI 0.84-0.88) and 0.85 (95% CI 0.84-0.87), respectively. Older age, high heart rate, and low body temperature were the 3 most important predictors of increased mortality, according to the explanations of the Self-Attention and Intersample Attention Transformer model. The explainable deep learning model that we developed could provide estimates of mortality and visual contribution of the features to the prediction for a patient with AMI. The explanations suggested that older age, unstable vital signs, and metabolic disorders may increase the risk of mortality in patients with AMI.

Development and external validation of a multidimensional deep learning model to dynamically predict kidney outcomes in IgA nephropathy.

Accurately predicting kidney outcomes in IgA nephropathy is crucial for clinical decision making. Insufficient use of longitudinal data in previous studies has limited the accuracy and interpretability of prediction models for failing to reflect the chronic nature of IgA nephropathy. This study aimed at establishing a multivariable dynamic deep learning model using comprehensive longitudinal data for the prediction of kidney outcomes in IgA nephropathy. In this retrospective cohort study of 2,056 IgA nephropathy patients at 18 kidney centers, a total of 28,317 data points were collected by the sliding window method. Among them, 15,462 windows in a single center were randomly assigned to training (80%) and validation (20%) sets while 8797 windows in 18 kidney centers were assigned to an independently test set. Interpretable Multi-Variable Long Short-Term Memory (IMV-LSTM), a deep learning model, was implemented to predict kidney outcomes (kidney failure or 50% decline in kidney function) based on time-invariant variables measured at biopsy and time-variant variables measured during follow-up. Risk performance was evaluated using Kaplan-Meier analysis and the C statistic. Trajectory analysis was performed to assess the various trends of clinical variables during follow-up. The model achieved a higher C statistic (0.93; 95% CI, 0.92-0.95) on the test set than the XGBoost prediction model that we developed in a previous study using only baseline information (C statistic, 0.84; 95% CI, 0.80-0.88). Kaplan-Meier analysis showed that groups with lower predicted risks from the full model survived longer than groups with higher risks. Time-variant variables demonstrated higher importance scores than time-invariant variables. Within time-variant variables, more recent measurements showed higher importance scores. Further interpretation showed that certain trajectory groups of time-variant variables such as serum creatinine and urine protein were associated with elevated risks of adverse outcomes. In IgA nephropathy, a deep learning model can be used to accurately and dynamically predict kidney prognosis based on longitudinal data, and time-variant variables show strong ability to predict kidney outcome.

Development, evaluation and validation of machine learning models to predict hospitalizations of patients with coronary artery disease within the next 12 months

BackgroundImproved survival of patients after acute coronary syndromes, population growth, and overall life expectancy rise have led to a significant increase in the proportion of patients with stable coronary artery disease (CAD), creating a significant load on the entire healthcare system. The disease often progresses with the development of many complications while significantly increasing the likelihood of hospitalization. Developing and applying a machine learning model for predicting hospitalizations of patients with CAD to an inpatient medical facility will allow for close monitoring of high-risk patients, early preventive interventions, and optimized medical care. AimsDevelopment and external validation of personalized models for predicting the preventable hospitalizations of patients with stable CAD and its complications using ML algorithms and data of real-world clinical practice. Methods135,873 depersonalized electronic health records of 49,103 patients with stable CAD were included in the study. Anthropometric measurements, physical examination results, laboratory, instrumental, anamnestic, and socio-demographic data, widely used in routine medical practice, were considered as potential predictors, a total of 73 features. Logistic regression, decision tree-based methods including gradient boosting (AdaBoost, LightGBM, XGBoost, CatBoost) and bagging (RandomForest and ExtraTrees), discriminant analysis (LinearDiscriminant, QuadraticDiscriminant), and naive Bayes classifier were compared. External validation was performed on the data of a separate region. ResultsThe best results and stability to external validation data were shown by the CatBoost model with an AUC of 0.875 (95% CI 0.865–0.885) for the internal testing and 0.872 (95% CI 0.856–0.886) for the external validation. The best model showed good performance evaluated through AUROC, Brier score and standardized net benefit (for the target NPV threshold) for the validation dataset that was only slightly similar to the train data. ConclusionThe metrics of the best model were superior to previously published studies. The results of external validation demonstrated the relative stability of the model to new data from another region that confirms the possibility of the model’s application in real clinical practice.

Development and comparison of machine learning models for in-vitro drug permeation prediction from microneedle patch

The field of machine learning (ML) is advancing to a larger extent and finding its applications across numerous fields. ML has the potential to optimize the development process of microneedle patch by predicting the drug release pattern prior to its fabrication and production. The early predictions could not only assist the in-vitro and in-vivo experimentation of drug release but also conserve materials, reduce cost, and save time. In this work, we have used a dataset gleaned from the literature to train and evaluate different ML models, such as stacking regressor, artificial neural network (ANN) model, and voting regressor model. In this study, models were developed to improve prediction accuracy of the in-vitro drug release amount from the hydrogel-type microneedle patch and the in-vitro drug permeation amount through the micropores created by solid microneedles on the skin. We compared the performance of these models using various metrics, including R-squared score (R2 score), root mean squared error (RMSE), and mean absolute error (MAE). Voting regressor model performed better with drug permeation percentage as an outcome feature having RMSE value of 3.24. In comparison, stacking regressor have a RMSE value of 16.54, and ANN model has shown a RMSE value of 14. The value of permeation amount calculated from the predicted percentage is found to be more accurate with RMSE of 654.94 than direct amount prediction, having a RMSE of 669.69. All our models have performed far better than the previously developed model before this research, which had a RMSE of 4447.23. We then optimized voting regressor model’s hyperparameter and cross validated its performance. Furthermore, it was deployed in a webapp using Flask framework, showing a way to develop an application to allow other users to easily predict drug permeation amount from the microneedle patch at a particular time period. This project demonstrates the potential of ML to facilitate the development of microneedle patch and other drug delivery systems.

Development and validation of machine learning models to predict unplanned hospitalizations of patients with diabetes within the next 12 months

BACKGROUND: The incidence of diabetes mellitus (DM) both in the Russian Federation and in the world has been steadily increasing for several decades. Stable population growth and current epidemiological characteristics of DM lead to enormous economic costs and significant social losses throughout the world. The disease often progresses with the development of specific complications, while significantly increasing the likelihood of hospitalization. The creation and inference of a machine learning model for predicting hospitalizations of patients with DM to an inpatient medical facility will make it possible to personalize the provision of medical care and optimize the load on the entire healthcare system.AIM: Development and validation of models for predicting unplanned hospitalizations of patients with diabetes due to the disease itself and its complications using machine learning algorithms and data from real clinical practice.MATERIALS AND METHODS: 170,141 depersonalized electronic health records of 23,742 diabetic patients were included in the study. Anamnestic, constitutional, clinical, instrumental and laboratory data, widely used in routine medical practice, were considered as potential predictors, a total of 33 signs. Logistic regression (LR), gradient boosting methods (LightGBM, XGBoost, CatBoost), decision tree-based methods (RandomForest and ExtraTrees), and a neural network-based algorithm (Multi-layer Perceptron) were compared. External validation was performed on the data of the separate region of Russian Federation.RESULTS: The best results and stability to external validation data were shown by the LightGBM model with an AUC of 0.818 (95% CI 0.802–0.834) in internal testing and 0.802 (95% CI 0.773–0.832) in external validation.CONCLUSION: The metrics of the best model were superior to previously published studies. The results of external validation showed the relative stability of the model to new data from another region, that reflects the possibility of the model’s application in real clinical practice.

Development and Validation of a Novel Machine Learning Model to predict the Survival of Patients with Gastrointestinal Neuroendocrine Neoplasms.

Well-calibrated models for personalized prognostication of patients with gastrointestinal neuroendocrine neoplasms (GINENs) are limited. This study aimed to develop and validate a machine-learning model to predict the survival of patients with GINENs. oblique random survival forest (ORSF) model, Cox proportional hazard risk model, Cox model with least absolute shrinkage and selection operator penalization, CoxBoost, Survival Gradient Boosting Machine, Extreme Gradient Boosting survival regression, DeepHit, DeepSurv, DNNSurv, Logistic-Hazard model, and PC-Hazard model were compared. We further tuned hyperparameters and selected variables for the best-performing ORSF. Then, the final ORSF model was validated. 43444 patients with GINENs were included. The median (interquartile range) survival time was 53 (19-102) months. The ORSF model performed best, in which age, histology, M stage, tumor size, primary tumor site, sex, tumor number, surgery, lymph nodes removed, N stage, race, and grade were ranked as important variables. However, chemotherapy and radiotherapy were not necessary for the ORSF model. The ORSF model had an overall C-index of 0.86 (95% confidence interval, 0.85-0.87). The area under the receiver operation curves at 1-, 3-, 5-, and 10-year were 0.91, 0.89, 0.87, and 0.80, respectively. The decision curve analysis showed superior clinical usefulness of the ORSF model than the American Joint Committee on Cancer Stage. A nomogram and an online tool were given. The machine-learning ORSF model could precisely predict the survival of patients with GINENs, with the ability to identify patients at high risk for death and probably guide clinical practice.

Development of a Talking Stick Learning Model Assisted by Question Boxes on Students' Social Sciences Learning Outcomes

Education is business man For build personality in accordance with values in society or as effort help participant educate For develop and improve knowledge, skills, values, attitudes and patterns Act useful behavior​ for life. Formulation problem is there is enhancement results learn social studies students class V against development of the Talking Stick learning model assisted by Question Box, How Smoothness implementation of the Talking Stick learning model assisted by Question Box results learn social studies students class V SDN Kaleroang, Kec. South Bungku, Regency Morowali, What constraint in development of the Talking Stick learning model with the help of Question Box results learn social studies students class V SDN Kaleroang, Kec. South Bungku, Regency Morowali. Research purposes For know constraint in development of the Talking Stick learning model with the help of Question Box results learn social studies students class V SDN Kaleroang, Kec. South Bungku, Regency Morowali. Study This use type study development ( Research and Development). Research and development methods or in Language English Research and Development is method research used​ For produce product certain, and test effectiveness product the. Research result A pre-test was used For know results Study beginning student before given treatment use Development of the Talking Stick learning model assisted by Question Box.

Development and validation of a machine learning model to improve precision prediction for irrational prescriptions in orthopedic perioperative patients

ABSTRACT Objective Our objective was to develop a machine learning model capable of predicting irrational medical prescriptions precisely within orthopedic perioperative patients. Methods A dataset comprising 3047 instances of suspected irrational medication prescriptions was collected from a sample of 1318 orthopedic perioperative patients from April 2019 to March 2022. Four machine learning models were employed to forecast irrational prescriptions, following which, the performance of each model was meticulously assessed. Subsequently, a thorough variable importance analysis was conducted on the model that performed the best predictive capabilities. Thereafter, the efficacy of integrating this optimal model into the existing audit prescription process was rigorously evaluated. Results Of the models utilized in this study, the RF model yielded the highest AUC of 92%, whereas the NB model presented the lowest AUC of 68%. Also, the RF model boasted the most robust performance in terms of PPV, reaching 82.4%, and NPV, reaching 86.6%. The ANN and the XGBoost model were neck and neck, with the ANN slightly edging out with a higher PPV of 95.9%, while the XGBoost model boasted an impressive NPV of 98.2%. The RF model singled out the following five factors as the most influential in predicting irrational prescriptions: the type of drug, the type of surgery, the number of comorbidities, the date of surgery after hospitalization, as well as the associated hospital and drug costs. Conclusion The RF model showcased significantly high level of proficiency in predicting irrational prescriptions among orthopedic perioperative patients, outperforming other models by a considerable margin. It effectively enhanced the efficiency of pharmacist interventions, displaying outstanding performance in assisting pharmacists to intervene with irrational prescriptions.

Abstract PO2-01-14: Development of Artificial Intelligence-based Machine Learning Models for Predicting Survival In Hormone-Receptor-Positive/HER2-Negative Early Breast Cancer undergoing Neoadjuvant Chemotherapy

Abstract Background. In hormone-receptor positive (HoR)/HER2 negative early breast cancer (BC) multiple efforts have been made to predict disease recurrence and survival. Machine learning techniques have been used but few studies have looked into their applicability in predicting survival on the basis of clinico-pathological characteristics. The aim of this study was to evaluate a random survival forest model to predict the prognosis of this specific BC subtype. Methods. In this multicenter, retrospective study, patients who fulfilled the following inclusion criteria were included: (1) diagnosis of pathologically confirmed HoR-positive/HER2-negative invasive BC; (2) early or locally advanced, stage I-II-III (3) patients receiving neoadjuvant anthracycline and/or taxane-based chemotherapy, concurrently or sequentially; (4) patients undergoing surgery for primary BC. Survival endpoints were disease-free survival (DFS) and overall survival (OS). A random survival forest algorithm was used to develop the predictive model. Ten-fold cross-validation was performed. The C-index and the continuous rank probability score (CRPS) were used to evaluate the discrimination of the predictive model, and a ROC curve was used to evaluate model precision. A cut point analysis based on maximally selected rank statistics was conducted to evaluate the best cut-off in the out-of-bag (OOB) mortality that could maximize DFS and OS prediction. Variable importance was assessed using Breiman-Cutler permutation importance. Results. Overall, 572 patients with HoR-positive/HER2-negative early BC were included. At univariate analysis age, T stage, N stage, grading, ER, PR and Ki67 were found to be significantly associated with DFS and ER, PgR, HER2, pCR ypT and ypN retained statistical significance at multivariate analysis. ER, pCR, pathological T and N stage were found to be significantly associated with OS at univariate and multivariate analysis. The following variables were included in the final model: menopausal status, age, histology, grade, clinical T/N, ER/PgR, Ki67 and HER2, pathological complete response (pCR), ypT and ypN. For DFS, the cross-validated C-index was 0.68 (95% CI 0.63-0.73) and the OOB CRPS was 0.15, with a OOB performance error of 0.31. The AUC calculated at 60 months was 0.91 (95% CI 0.88-0.94). Out-of bag RSF-based risk scores for individual patients were calculated and an optimal cut-off of 22.58 was identified. The HR between high risk and low risk group was 3.08 (95% CI 2.16-4.39, p&amp;lt; 0.001). For OS, the cross-validated C-index was 0.66 (95% CI 0.60-0.71) and the OOB CRPS was 0.17, with a OOB performance error of 0.35. The AUC calculated at 60 months 0.91 (95% CI 0.88-0.95). An optimal cut-off of 21.54 was identified and the HR between low risk and high risk group was 2.50 (95% CI 1.73-3.61, p&amp;lt; 0.001). For DFS, the most important variables were cN (9.00), ypN (7.35), ER (7.21), ypT (5.38), PgR (4.11), pCR (4.01) and age (2.82) while for OS the most important variables were cN (7.88), ER (5.37), PgR (3.61), age (3.37), pCR (2.80), ypT (2.64) and ypN (2.60). Discussion. We analyzed the performance of RSF in the prediction of DFS and OS based on the contribution of clinico-pathological features commonly available at the baseline of the NCT and post-surgery. By selecting patients with HoR-positive HER2-negative disease, we were able to show which clinico-pathological features within a ML model have greater predictive importance. The model could be integrated with image-based tumor and radiology features to improve its predictive accuracy. This retrospective multicenter study suggests that the combination of easily accessible clinico-pathological features within a ML model may reliably predict DFS and OS in the context of HoR-positive/HER2-negative BCs. Citation Format: Luca Mastrantoni, Giovanna Garufi, Noemi Maliziola, Elena Di Monte, Giorgia Arcuri, Valentina Frescura, Angelachiara Rotondi, Giulia Giordano, Luisa Carbognin, Alessandra Fabi, Ida Paris, Gianluca Franceschini, Armando Orlandi, Antonella Palazzo, Giovanni Scambia, Giampaolo Tortora, Emilio Bria. Development of Artificial Intelligence-based Machine Learning Models for Predicting Survival In Hormone-Receptor-Positive/HER2-Negative Early Breast Cancer undergoing Neoadjuvant Chemotherapy [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO2-01-14.

Comprehensive analysis of clinical images contributions for melanoma classification using convolutional neural networks.

Timely diagnosis plays a critical role in determining melanoma prognosis, prompting the development of deep learning models to aid clinicians. Questions persist regarding the efficacy of clinical images alone or in conjunction with dermoscopy images for model training. This study aims to compare the classification performance for melanoma of three types of CNN models: those trained on clinical images, dermoscopy images, and a combination of paired clinical and dermoscopy images from the same lesion. We divided 914 image pairs into training, validation, and test sets. Models were built using pre-trained Inception-ResNetV2 convolutional layers for feature extraction, followed by binary classification. Training comprised 20 models per CNN type using sets of random hyperparameters. Best models were chosen based on validation AUC-ROC. Significant AUC-ROC differences were found between clinical versus dermoscopy models (0.661vs. 0.869, p<0.001) and clinical versus clinical + dermoscopy models (0.661vs. 0.822, p=0.001). Significant sensitivity differences were found between clinical and dermoscopy models (0.513vs. 0.799, p=0.01), dermoscopy versus clinical + dermoscopy models (0.799vs. 1.000, p=0.02), and clinical versus clinical + dermoscopy models (0.513vs. 1.000, p<0.001). Significant specificity differences were found between dermoscopy versus clinical + dermoscopy models (0.800vs. 0.288, p<0.001) and clinical versus clinical + dermoscopy models (0.650vs. 0.288, p<0.001). CNN models trained on dermoscopy images outperformed those relying solely on clinical images under our study conditions. The potential advantages of incorporating paired clinical and dermoscopy images for CNN-based melanoma classification appear less clear based on our findings.

Evaluating the efficacy of univariate LSTM approach for COVID-19 data prediction in Indonesia

The coronavirus disease 2019 (COVID-19) pandemic, originating in 2020, has emerged as a critical global issue due to its rapid and widespread transmission. Indonesia, among the affected nations, has taken measures to address the situation, including the development of a deep learning model for predicting future COVID-19 infection and spread. This predictive tool serves as a valuable reference for the government and stakeholders, aiding them in making informed decisions and implementing appropriate measures to contain the virus. The deep learning model employs the long short-term memory (LSTM) algorithm, chosen for its ability to recognize temporal patterns in the country’s COVID-19 data. The model creation process involves data collection, preprocessing, model architecture planning, modeling, training, and evaluation. Two LSTM models were developed: a univariate and a multivariate model. Following thorough training and evaluation, the univariate model emerged as the superior choice, boasting evaluation metrics of 16.72 for mean absolute percentage error (MAPE) and 66.36 for root mean squared error (RMSE). This model was then deployed on a publicly accessible website, presenting visualizations of past COVID-19 data and predictions of future cases through line graphs. This user-friendly platform enables the public to access and analyze the data easily.

Comprehensive quantitative radiogenomic evaluation reveals novel radiomic subtypes with distinct immune pattern in glioma

BackgroundAccurate classification of gliomas is critical to the selection of immunotherapy, and MRI contains a large number of radiomic features that may suggest some prognostic relevant signals. We aimed to predict new subtypes of gliomas using radiomic features and characterize their survival, immune, genomic profiles and drug response. MethodsWe initially obtained 341 images of 36 patients from the CPTAC dataset for the development of deep learning models. Further 1812 images of 111 patients from TCGA_GBM and 152 images of 53 patients from TCGA_LGG were collected for testing and validation. A deep learning method based on Mask R-CNN was developed to identify new subtypes of glioma patients and compare the survival status, immune infiltration patterns, genomic signatures, specific drugs, and predictive models of different subtypes. Results200 glioma patients (mean age, 33 years ± 19 [standard deviation]) were enrolled. The accuracy of the deep learning model for identifying tumor regions achieved 88.3% (98/ 111) in the test set and 83% (44/53) in the validation set. The sample was divided into two subtypes based on radiomic features showed different prognostic outcomes (hazard ratio, 2.70). According to the results of the immune infiltration analysis, the subtype with a poorer prognosis was defined as the immunosilencing radiomic (ISR) subtype (n=43), and the other subtype was the immunoactivated radiomic (IAR) subtype (n=53). Subtype-specific genomic signatures distinguished celllines into ISR celllines (n=9) and control celllines (n=13), and identified eight ISR-specific drugs, four of which were validated by the OCTAD database. Three machine learning-based classifiers showed that radiomic and genomic co-features better predicted the radiomic subtypes of gliomas. ConclusionsThese findings provide insights into how radiogenomic could identify specific subtypes that predict prognosis, immune and drug sensitivity in a non-invasive manner.