Least Absolute Shrinkage And Selection Operator Method Research Articles

Chest CT radiomics is feasible in evaluating bone changes in chronic kidney diseases.

Non-invasive imaging methods are still lacking for evaluating bone changes in chronic kidney diseases (CKD). To investigate the feasibility of chest CT radiomics in evaluating bone changes caused by CKD. In total, 75 patients with stage 1 CKD (CKD1) and 75 with stage 5 CKD (CKD5) were assessed using the chest CT radiomics method. Radiomics features of bone were obtained using 3D Slicer software and were then compared between CKD1 and CKD5 cases. The methods of maximum correlation minimum redundancy (mRMR) and least absolute shrinkage and selection operator (LASSO) were used to establish a prediction model to determine CKD. The receiver operating characteristic (ROC) curve was used to determine the performance of the model. Cases of CKD1 and CKD5 differed in 40 radiomics features (P <0.05). Using the mRMR and LASSO methods, five features were finally selected to establish a predication model. The area under the receiver operating characteristic curve of the model in the determination of CKD1 and CKD5 was 0.903 and 0.854, respectively, for the training and validation cohorts. Chest CT radiomics is feasible in evaluating bone changes caused by CKD.

234 Predictive Radiomics in Glioblastoma: A Machine Learning Approach for Personalized Survival Outcomes and a Novel Web Application

INTRODUCTION: Diverse characteristics of glioblastomas (GBMs) complicate survival predictions, requiring personalized approaches. METHODS: From The Cancer Imaging Archive (TCIA) database, 611 patients from the UPENN-GBM collection and 500 patients from the UCSF-PDGM collections were included in this study. UPENN-GBM collection was used as the training set and for 5-fold internal cross-validation, whereas the UCSF-PDGM collection was used for external validation. Survival at 3, 6, 12, and 24 months were the outcomes of interest. Necrotic tumor core, enhancing tumor, and whole tumor segmentation masks for T1 with contrast, T1, T2, and FLAIR images were obtained automatically with a pre-trained Swin UNETR model. From these regions of interest, radiomic features were extracted using the PyRadiomicsLibrary. The LASSO (least absolute shrinkage and selection operator) method was employed for radiomic feature selection. The ML algorithms TabPFN, XGBoost, LightGBM, Support Vector Machine (SVM), and Random Forest (RF) were utilized to predict the survival outcomes of interest. RESULTS: The highlight of our study, our web application is available for use via the following link: https://huggingface.co/spaces/MSHSNeurosurgeryResearch/Radiomics_Based_Glioblastoma_Survival_Prediction_Tool. CONCLUSIONS: This study not only provides a radiomics pipeline for predicting the GBM survival outcomes, but it also contributes to the clinical adoption of these models by developing a fully automatic and user-friendly web application which make them available to clinicians worldwide.

SH3GL2 and MMP17 as lung adenocarcinoma biomarkers: a machine-learning based approach

ObjectiveUsing bioinformatics machine learning methods, our research aims to identify the potential key genes associated with Lung adenocarcinoma (LUAD). MethodsWe obtained two gene expression profiling microarrays (GSE68571 and GSE74706) from the public Gene Expression Omnibus (GEO) database at the National Centre for Biotechnology Information (NCBI). The purpose was to identify Differentially Expressed Genes (DEGs) between the lung adenocarcinoma group and the healthy control group. The limma R package in R was utilized for this analysis. For the differential gene diagnosis of lung adenocarcinoma, we employed the least absolute shrinkage and selection operator (LASSO) regression and SVM-RFE screening crossover. To evaluate the performance, ROC curves were plotted. We performed immuno-infiltration analysis using CIBERSORT. Finally, we validated the key genes through qRT-PCR and Western-blot verification, then downregulated MMP17 gene expression, upregulated SH3GL2 gene expression, and performed CCK8 experiments. ResultsA total of 32 Differentially Expressed Genes (DEGs) were identified. Two diagnostic marker genes, SH3GL2 and MMP17, were selected by employing LASSO and SVM-RFE machine learning methods. In Lung adenocarcinoma cells, the expression of MMP17 was observed to be elevated compared to normal lung epithelial cells in the control group (P < 0.05). In contrast, a down-regulation of SH3GL2 was found in Lung adenocarcinoma cells (P < 0.05). Finally, we downregulated MMP17 and upregulated SH3GL2 gene expression, then the CCK8 showed that the proliferation of both lung cancer cells was inhibited. ConclusionSH3GL2 and MMP17 are expected to be potential biomarkers for Lung adenocarcinoma.

Prediction of Flooding Characteristic in Pulsed Disc and Doughnut Extraction Column Using LASSO Method

For achieving high separation efficiency and a large throughput in the nuclear fuel reprocessing industry, it is crucial to have a profound understanding of the flooding characteristics in pulsed disc and doughnut extraction columns (PDDCs). For this purpose, the least absolute shrinkage and selection operator (LASSO) method was utilized to obtain predictive equations that provide high applicability and analytical convenience. The effects of three operating conditions (dispersed-phase velocity, continuous-phase velocity, and pulse intensity) on the hydrodynamic parameters (dispersed phase holdup, slip velocity, characteristic velocity, and flooding point) were studied in a Φ50 PDDC in the kerosene-water system. The LASSO method was applied to select highly correlated features of the hydrodynamic parameters and to propose second-order prediction equations. The effectiveness of LASSO was also compared to the published correlations and traditional linear regression. The second-order-regression of LASSO produced more intuitive prediction equations with the mean relative error within 15%. The impact of each operating variable on the hydrodynamic parameters was quantitatively analyzed by calculating the partial derivatives of these prediction equations. The dispersed-phase flow rate predominantly affects the holdup within the operating conditions. Pulse intensity emerges as the primary factor affecting slip velocity, characteristic velocity, and flooding throughput.

FAST® questionnaire for identifying glaucoma patients at risk for ocular surface disease.

To evaluate the validity and reliability of the new Fast Assessment of the Ocular Surface Trouble (FAST®) questionnaire for identifying glaucoma or ocular hypertension (OHT) patients at risk of ocular surface disease (OSD). A multicenter, international, cross-sectional, epidemiological survey evaluated the most accurate interview items and ocular signs on the initial 14-item version of FAST® to develop a shorter version for routine, quick clinical use. Rasch analysis and least absolute shrinkage and selection operator (LASSO) method was used to reduce the number of items on the questionnaire. Sensitivity and specificity of FAST® were assessed with receiver operating characteristic (ROC) curves for the detection of OSD with the questionnaire and ophthalmic assessment. A total of 2308 eyes (1154 patients) were analyzed in this study by 92 ophthalmologists. The initial version of the FAST® indicated 60% of the subjects had OSD. Rasch analysis allowed removal of some clinical signs. The LASSO method allowed elimination of some items from the original questionnaire for a 9-item and a 6-item version of FAST®. For the 6-item questionnaire, the sensitivity and specificity were 71.9% and 74.3% respectively and the area under the curve was 0.815. The FAST® questionnaire is a valid and reliable tool for use in routine clinical practice and in clinical trials. The short versions of the questionnaire allow quick detection of the majority of patients with OHT or glaucoma at risk of dry eye.

Dyslipidemia and its associated factors among community adults located in Shangcheng district, Zhejiang province

Dyslipidemia is highly prevalent and an important modifiable risk factor of cardiovascular disease in China. However, there is little information on the dyslipidemia in Shangcheng district, eastern China. Therefore, this study aims to investigate the prevalence and associated factors of dyslipidemia among community adults in this area. A community based cross-sectional study was conducted from August 1 to November 30, 2020. The study utilized a multi-stage probability sampling method to enroll permanent residents (those who have resided in this region for 6 months or more) who were 18 years old or above. Firstly, five streets were selected randomly, and then two communities were randomly selected from each of the chosen streets, finally, systematic sampling at the household level was conducted. All participants were interviewed by trained investigators and underwent anthropometric and biochemical measurements using standard criteria. LASSO (least absolute shrinkage and selection operator) and multivariate binary logistic regression were employed to identify the factors associated with dyslipidemia. In total, 3153 participants were enrolled into this study, resulting in a response rate of 93.28%. 33 subjects were excluded because of incomplete data. Finally, 3120 participants with a mean age of 55.26 (SD = 17.97) years were included into analysis. The overall prevalence of dyslipidemia was 35.96%. 21 variables were screened to multivariate binary logistic regression through the implementation of LASSO method. The multivariate binary logistic regression analysis revealed that individuals aged 40–49 [adjusted odds ratio (aOR) = 2.197, 95% confidence interval (CI) 1.445–3.341], 50–59 (aOR = 3.213, 95% CI 2.121–4.868), 60–69 (aOR = 4.777, 95% CI 3.169–7.201), and 70 and above (aOR = 5.067, 95% CI 3.301–7.777), with an educational level of junior middle school (aOR = 1.503, 95% CI 1.013–2.229), with an educational level of senior middle school (aOR = 1.731, 95% CI 1.25–2.397), with an educational level of under graduate and above (aOR = 2.125, 95% CI 1.46–3.095), without hypertension (aOR = 0.627, 95% CI 0.517–0.76), without diabetes (aOR = 0.625, 95% CI 0.498–0.785), obesity (aOR = 1.887, 95% CI 1.13–3.154), frequent smoking (aOR = 1.727, 95% CI 1.293–2.308), frequent drinking (aOR = 0.738, 95% CI 0.556–0.981), without family history of CVD (aOR = 0.505, 95% CI 0.342–0.744), and daily seafood intakes between 42.87 and 71.43 g (aOR = 1.31, 95% CI 1.05–1.634) were significantly associated with dyslipidemia. Gender-stratified analyses showed that aged 70 and above (aOR = 2.127, 95% CI 1.195–3.785), without hypertension (aOR = 0.643, 95% CI 0.484–0.854), without diabetes (aOR = 0.603, 95% CI 0.436–0.834), without CVD (aOR = 0.494, 95% CI 0.309–0.791), without stroke (aOR = 1.767, 95% CI 1.036–3.012), frequent smoking (aOR = 1.951, 95% CI 1.415–2.691), former smoking (aOR = 1.703, 95% CI 1.16–2.502) were significantly associated with dyslipidemia in male. Aged 40–49 (aOR = 3.51, 95% CI 1.789–6.887), 50–59 (aOR = 7.03, 95% CI 3.584–13.791), 60–69 (aOR = 15.728, 95% CI 8.005–30.9), and 70 and above (aOR = 12.929, 95% CI 6.449–25.921), with an educational level of senior middle school (aOR = 1.926, 95% CI 1.288–2.881), with an educational level of under graduate and above (aOR = 2.91, 95% CI 1.75–4.837), without hypertension (aOR = 0.592, 95% CI 0.45–0.779), without diabetes (aOR = 0.619, 95% CI 0.443–0.865), without family history of CVD (aOR = 0.429, 95% CI 0.251–0.733), without family history of cancer (aOR = 0.542, 95% CI 0.316–0.929), daily vegetables intakes between 251 and 500 g (aOR = 0.734, 95% CI 0.545–0.99), daily seafood intakes between 42.87 and 71.43 g (aOR = 1.421, 95% CI 1.04–1.942) were significantly associated with dyslipidemia in female. In the age-stratified analyses, it was found that without hypertension (aOR = 0.522, 95% CI 0.375–0.727) or diabetes (aOR = 0.445, 95% CI 0.267–0.744), obesity (aOR = 2.956, 95% CI 1.258–6.942), frequent smoking (aOR = 1.826, 95% CI 1.196–2.787), showed a significant association with dyslipidemia in individuals aged younger than 60 years. Female (aOR = 1.764, 95% CI 1.316–2.366), with an educational level of junior middle school (aOR = 1.793, 95% CI 1.169–2.749), with an educational level of senior middle school (aOR = 2.002, 95% CI 1.406–2.849), with an educational level of under graduate and above (aOR = 2.849, 95% CI 1.791–4.532), without hypertension (aOR = 0.604, 95% CI 0.477–0.764), without diabetes (aOR = 0.63, 95% CI 0.486–0.818), without CVD (aOR = 0.66, 95% CI 0.473–0.921), frequent smoking (aOR = 1.513, 95% CI 1.02–2.245), former smoking (aOR = 1.647, 95% CI 1.089–2.491), without family history of CVD (aOR = 0.406, 95% CI 0.239–0.692), daily seafood intakes between 42.87 and 71.43 g (aOR = 1.376, 95% CI 1.018–1.859) were significantly associated with dyslipidemia among participants aged 60 and above. Dyslipidemia is a prevalent condition observed among adults residing in Shangcheng district. Risk factors such as gender, age, education, hypertension, diabetes, cardiovascular disease, stroke, obesity, smoking, drinking, family history of cardiovascular disease, family history of cancer, daily vegetables intakes, daily seafood intakes were associated with dyslipidemia and varied across population of different gender and age groups. Enhancing education and promoting self-awareness regarding the necessity of behavior modification and regular medication intake would be beneficial in reducing the occurrence of dyslipidemia among adults in the Shangcheng district.

Time-varying and multi-scale analysis of copper price influencing factors based on LASSO and EMD methods

In this paper, we select 75 indicators to conduct a comprehensive analysis of the factors influencing the copper price along six dimensions: inventory, supply, demand, the macroeconomy, finance, and geopolitics. Facing the high-dimensionality problem, we use the least absolute shrinkage and selection operator (LASSO) regression model to select variables to measure the contribution of each category of factors. Furthermore, we identify the time-varying nature of the relationship among factors with rolling windows. Then, we decompose copper prices into different scales of fluctuations by means of empirical mode decomposition (EMD) and investigate the driving factors at each scale. The results show that financial and geopolitical factors have played an important role in copper pricing in recent years. The long-run fluctuation trend of copper prices is mainly determined by fundamental factors, while financial and geopolitical factors have a more direct impact on short-term fluctuations.

Predicting and analysing of the unfavourable outcomes of early applicated percutaneous endoscopic interlaminar discectomy for lumbar disc herniation: development and validation based nomogram.

To predict and analyse the unfavourable outcomes of early applicated percutaneous endoscopic interlaminar discectomy for lumbar disc herniation. Information of 426 patients treated by early applicated percutaneous endoscopic interlaminar discectomy (PEID) for lumbar disc herniation (LDH) at our hospital from June 2017 to October 2021 in addition to 17 other features was collected. The risk factors were selected by the least absolute shrinkage and selection operator method (LASSO) regression. Then, a prediction model (nomogram) was established to predict the unfavourable outcomes of using the risk factors selected from LASSO regression. Bootstrap (n = 1000) was used to perform the internal validation, and the performance of the model was assessed by the concordance index (C-index), receiver operating characteristic curve, and calibration curve. The decision curve analysis (DCA) and clinical impact curve (CIC) were used to assess the clinical utility of the model, respectively. Finally, 53 of 426 patients showed unfavourable outcomes. Five potential factors, Modic change, Calcification, Lumbar epidural steroid injection preoperative, Articular process hyperplasia and cohesion, and Laminoplasty technique, were selected according to the LASSO regression, that identified the predictors to establish nomogram model. Meanwhile, the C-index of the prediction nomogram was 0.847, the area under the receiver operating characteristic curve value was 0.847, and the interval bootstrapping (n = 1000) validation C-index was 0.809. The model has good practicability for clinics according to the DCA and CIC. This nomogram model has good predictive performance and clinical practicability, which could provide a certain basis for predicting unfavourable outcomes of early applicated PEID for LDH.

Abstract WP94: Forecasting Outcomes After Mechanical Thrombectomy: Predictive Modeling of Neurological Improvement in the SELECT Database

Introduction: Outcome prediction and prognostication are at the heart of determining next steps in stroke care. We aim to evaluate two variable selection strategies to provide optimal prediction of functional independence among post-EVT patients at admission, day 1, and time of discharge using data from the SELECT prospective cohort study. Methods: Prediction strategies were created using baseline demographics, clinical presentation and imaging characteristics for the admission model, procedure characteristics and outcomes, clinical status, and baseline factors for the day 1 model, and follow-up infarct volume, cerebral edema, clinical status, and previously mentioned factors for the discharge model. Model selection used backwards stepwise regression and least absolute shrinkage and selection operator (LASSO) strategy. Patients with unavailable data were excluded. Calibration and internal validation was performed using bootstrapping for 50 repetitions. Results: 149 EVT patients with all available data points were included. Functional independence (mRS 0-2) was observed in 89 (60%) patients. LASSO and stepwise selection methods selected different variables at baseline, with LASSO providing reasonably better model performance over stepwise model selection. Both stepwise and LASSO regression methods selected baseline ASPECTS and NIHSS at day 1 for day 1 model, and baseline ASPECTS, NIHSS at day 1 and NIHSS at day 7 for discharge model. Day 1 and discharge models demonstrated significantly better predictive performance in bootstrap validation over baseline models [Table]. Conclusions: Clinical symptoms in the post-acute phase were the strongest predictors of functional independence at 90-day follow-up and were selected over procedural factors or cerebral edema. Prediction of functional independence at baseline involved several clinical and imaging variables with lower predictive performance than post-acute phase models.

Exploring factors influencing farmers’ health self-assessment in China based on the LASSO method

As the main force and practice subject of rural revitalisation, farmers' health is intricately linked to agricultural production and the rural economy. This study utilizes open data from the 2015 China Nutrition and Health Survey and employs the Least Absolute Shrinkage and Selection Operator (LASSO) method to explore the factors influencing farmers' self-assessment of health. The findings reveal that education level, proactive nutrition knowledge seeking, healthy dietary preferences and habits, and the use of clean cooking fuel positively impact farmers' health self-assessment. Conversely, age, history of illness or injury, and participation in medical insurance negatively affect their self-assessment. Furthermore, factors influencing farmers' health self-assessment exhibit heterogeneity across regions. Our findings suggest that promoting health education, disseminating nutritional dietary knowledge, and enhancing rural household infrastructure play an important role in improving farmers' self-evaluation of health. Therefore, policymakers should design more targeted health interventions and infrastructure improvement plans based on farmers' self-assessment of health and the level of regional economic development.

Anoikis related genes may be novel markers associated with prognosis for ovarian cancer

The aim of this study was to determine the prognostic significance of anoikis related genes (ARGs) in ovarian cancer (OC) and to develop a prognostic signature based on ARG expression. We analyzed cohorts of OC patients and used nonnegative matrix factorization (NMF) for clustering. Single-sample gene-set enrichment analysis (ssGSEA) was employed to quantify immune infiltration. Survival analyses were performed using the Kaplan–Meier method, and differences in survival were determined using the log-rank test. The extent of anoikis modification was quantified using a risk score generated from ARG expression. The analysis of single-cell sequencing data was performed by the Tumor Immune Single Cell Hub (TISCH). Our analyses revealed two distinct patterns of anoikis modification. The risk score was used to evaluate the anoikis modification patterns in individual tumors. Three hub-genes were screened using the LASSO (Least Absolute Shrinkage and Selection Operator) method and patients were classified into different risk groups based on their individual score and the median score. The low-risk subtype was characterized by decreased expression of hub-genes and better overall survival. The risk score, along with patient age and gender, were considered to identify the prognostic signature, which was visualized using a nomogram. Our findings suggest that ARGs may play a novel role in the prognosis of OC. Based on ARG expression, we have developed a prognostic signature for OC that can aid in patient stratification and treatment decision-making. Further studies are needed to validate these results and to explore the underlying mechanisms.

Staging liver fibrosis: comparison of radiomics model and fusion model based on multiparametric MRI in patients with chronic liver disease.

To develop and compare radiomics model and fusion model based on multiple MR parameters for staging liver fibrosis in patients with chronic liver disease. Patients with chronic liver disease who underwent multiparametric abdominal MRI were included in this retrospective study. Multiparametric MR images were imported into 3D-Slicer software for drawing bounding boxes on MR images. By using a 3D-Slicer extension of SlicerRadiomics, radiomics features were extracted from these MR images. The z-score normalization method was used for post-processing radiomics features. The least absolute shrinkage and selection operator method (LASSO) was performed for selecting significant radiomics features. The logistic regression analysis was used for building the radiomics model. A fusion model was built by integrating serum fibrosis biomarkers of aspartate transaminase-to-platelet ratio index (APRI) and the fibrosis-4 index (FIB-4) with radiomics signatures. In the training cohort, AUCs of radiomics and fusion model were 0.707-0.842 and 0.718-0.854 for differentiating different groups. In the testing cohort, AUCs were 0.514-0.724 and 0.609-0.728. For the training cohort, there was no significantdifference of AUCs between radiomics and fusion model (p > 0.05). For the testing cohort, AUCs of fusion model were higher than those of radiomics model in differentiating F1-3 vs. F4 and F1-2 vs. F4 (p = 0.011 & 0.042). Radiomics model and fusion model based on multiparametric MRI exhibited the feasibility for staging liver fibrosis in patients with CLD, and APRI and FIB-4 could improve the diagnostic performance of radiomics model in differentiatingF1-3 vs. F4 and F1-2 vs. F4.

Chest CT Radiomics is Feasible in Evaluating Muscle Change in Diabetes Patients.

Non-invasive imaging methods are still lacking for the evaluation of muscle changes in diabetes. To investigate the feasibility of muscle CT radiomics in evaluating muscle changes in diabetes. 60 diabetics and 60 health controls (HC) were assessed with the method of muscle CT radiomics. 93 CT images of radiomics features of the pectoralis major muscle (PMM) were obtained by using the software 3D Slicer and were then compared between diabetics and HC cases. The least absolute shrinkage and selection operator (LASSO) regression method was used to establish a prediction model. The receiver operating characteristic (ROC) curve was used to determine the performance of the model. Diabetics and HC cases differed in 19 radiomics features (P<0.05). By using the LASSO method, 6 features were finally selected. The AUC of the model in the discrimination of diabetics and HC were 0.92 and 0.90, respectively, for the training cohort and validation cohort. Muscle CT radiomics is feasible in evaluating muscle changes in diabetes.

Radiomics analysis of pancreas based on dual-energy computed tomography for the detection of type 2 diabetes mellitus.

To utilize radiomics analysis on dual-energy CT images of the pancreas to establish a quantitative imaging biomarker for type 2 diabetes mellitus. In this retrospective study, 78 participants (45 with type 2 diabetes mellitus, 33 without) underwent a dual energy CT exam. Pancreas regions were segmented automatically using a deep learning algorithm. From these regions, radiomics features were extracted. Additionally, 24 clinical features were collected for each patient. Both radiomics and clinical features were then selected using the least absolute shrinkage and selection operator (LASSO) technique and then build classifies with random forest (RF), support vector machines (SVM) and Logistic. Three models were built: one using radiomics features, one using clinical features, and a combined model. Seven radiomic features were selected from the segmented pancreas regions, while eight clinical features were chosen from a pool of 24 using the LASSO method. These features were used to build a combined model, and its performance was evaluated using five-fold cross-validation. The best classifier type is Logistic and the reported area under the curve (AUC) values on the test dataset were 0.887 (0.73-1), 0.881 (0.715-1), and 0.922 (0.804-1) for the respective models. Radiomics analysis of the pancreas on dual-energy CT images offers potential as a quantitative imaging biomarker in the detection of type 2 diabetes mellitus.

Comparative Analysis of Ridge, LASSO, and Elastic Net Regularization Approaches in Handling Multicollinearity for Infant Mortality Data in South Sulawesi

Infant mortality rate is a crucial indicator for assessing the health and infant care quality in a region. In the effort to reduce infant mortality rates, regression analysis serves as a tool to identify influential factors. However, regression analysis often encounters the challenge of multicollinearity, which involves high correlation among predictor variables. To address this issue, various regularization techniques can be applied, such as ridge regression, least absolute shrinkage and selection operator (LASSO), and elastic net. Ridge regression aims to control coefficient variance, while LASSO directs some coefficients to zero, functioning as variable selection. Elastic net combines the strengths of both methods by merging ridge and LASSO regularization. The objective of this research is to evaluate the performance of ridge regression, elastic net, and LASSO methods in handling multicollinearity issues, utilizing infant mortality rate data in South Sulawesi Province. The results indicate that the elastic net method outperforms both Ridge and LASSO methods. The best-performing model is obtained through elastic net with a coefficient of determination value of 60.81%, whereas ridge and LASSO methods yield coefficient of determination values of 54.11% and 58.18%, respectively. This demonstrates that the application of the elastic net method is capable of producing more accurate results in modeling the variables within the analysis of infant mortality rate data compared to other methods.

Predicting the risk of diabetic retinopathy using explainable machine learning algorithms

Background and objectiveDiabetic retinopathy (DR) is a global health concern among diabetic patients. The objective of this study was to propose an explainable machine learning (ML)-based system for predicting the risk of DR. Materials and methodsThis study utilized publicly available cross-sectional data in a Chinese cohort of 6374 respondents. We employed boruta and least absolute shrinkage and selection operator (LASSO) based feature selection methods to identify the common predictors of DR. Using the identified predictors, we trained and optimized four widly applicable models (artificial neural network, support vector machine, random forest, and extreme gradient boosting (XGBoost) to predict patients with DR. Moreover, shapely additive explanation (SHAP) was adopted to show the contribution of each predictor of DR in the prediction. ResultsCombining Boruta and LASSO method revealed that community, TCTG, HDLC, BUN, FPG, HbAlc, weight, and duration were the most important predictors of DR. The XGBoost-based model outperformed the other models, with an accuracy of 90.01%, precision of 91.80%, recall of 97.91%, F1 score of 94.86%, and AUC of 0.850. Moreover, SHAP method showed that HbA1c, community, FPG, TCTG, duration, and UA1b were the influencing predictors of DR. ConclusionThe proposed integrating system will be helpful as a tool for selecting significant predictors, which can predict patients who are at high risk of DR at an early stage in China.

Identification of angiogenesis-related genes in diabetic foot ulcer using machine learning algorithms

BackgroundDiabetic foot ulcers (DFUs) are among the most prevalent and dangerous complications of diabetes. Angiogenesis is pivotal for wound healing; however, its role in the chronic wound healing process in DFU requires further investigation. We aimed to investigate the pathogenic processes of angiogenesis in DFU from a molecular biology standpoint and to offer insightful information about DFU prevention and therapy. MethodsDifferential gene and weighted gene co-expression network analyses (WGCNA) were employed to screen for genes related to DFU using the downloaded and collated GSES147890 datasets. With the goal of identifying hub genes, an interaction among proteins (PPI) network was constructed, and enrichment analysis was carried out. Utilizing a variety of machine learning techniques, including Boruta, Support Vector Machine Recursive Feature Elimination (SVM-RFE), and Least Absolute Shrinkage and Selection Operator (LASSO), we were able to determine which hub genes most strongly correspond to DFU. This allowed us to create an ideally suited DFU forecasting model that was validated via an external dataset. Finally, by merging 36 angiogenesis-related genes (ARGs) and machine learning models, we identified the genes involved in DFU-related angiogenesis. ResultsBy merging 260 genes located in the green module and 59 differentially expressed genes (DEGs), 35 candidate genes highly associated with DFU were found for more investigation. 35 candidate genes were enriched in epidermal growth factor receptor binding, nuclear division regulation, fluid shear stress, atherosclerosis, and negative regulation of chromosomal structure for the enrichment study. Fifteen hub genes were found with the aid of the CytoHubba plug. The LASSO method scored better in terms of prediction performance (GSE134341) (LASSO:0.89, SVM:0.65, Boruta:0.66) based on the validation of the external datasets. We identified thrombomodulin (THBD) as a key target gene that potentially regulates angiogenesis during DFU development. Based on the external validation dataset (GSE80178 and GSE29221), receiver operating characteristic (ROC) curves with higher efficiency were generated to confirm the potential of THBD as a biomarker of angiogenesis in DFU. Furthermore supporting this finding were the results of Western blot and real-time quantitative polymerase chain reaction (RT-qPCR), which showed decreased THBD expression in human umbilical vein endothelial cells (HUVECs) cultivated under high glucose. ConclusionsThe findings implicate that THBD may influence DFU progression as a potential target for regulating angiogenesis, providing a valuable direction for future studies.

Radiomics Signature Based on Support Vector Machines for the Prediction of Pathological Complete Response to Neoadjuvant Chemoradiotherapy in Locally Advanced Rectal Cancer

The objective of this study was to evaluate the discriminative capabilities of radiomics signatures derived from three distinct machine learning algorithms and to identify a robust radiomics signature capable of predicting pathological complete response (pCR) after neoadjuvant chemoradiotherapy in patients diagnosed with locally advanced rectal cancer (LARC). In a retrospective study, 211 LARC patients were consecutively enrolled and divided into a training cohort (n = 148) and a validation cohort (n = 63). From pretreatment contrast-enhanced planning CT images, a total of 851 radiomics features were extracted. Feature selection and radiomics score (Radscore) construction were performed using three different machine learning methods: least absolute shrinkage and selection operator (LASSO), random forest (RF) and support vector machine (SVM). The SVM-derived Radscore demonstrated a strong correlation with the pCR status, yielding area under the receiver operating characteristic curves (AUCs) of 0.880 and 0.830 in the training and validation cohorts, respectively, outperforming the RF and LASSO methods. Based on this, a nomogram was developed by combining the SVM-based Radscore with clinical indicators to predict pCR after neoadjuvant chemoradiotherapy. The nomogram exhibited superior predictive power, achieving AUCs of 0.910 and 0.866 in the training and validation cohorts, respectively. Calibration curves and decision curve analyses confirmed its appropriateness. The SVM-based Radscore demonstrated promising performance in predicting pCR for LARC patients. The machine learning-driven nomogram, which integrates the Radscore and clinical indicators, represents a valuable tool for predicting pCR in LARC patients.

Forecasting stock closing prices with an application to airline company data

Forecasting stock market movements is a challenging task from the practitioners’ point of view. We explore how model selection via the least absolute shrinkage and selection operator (LASSO) approach can be better used to forecast stock closing prices using real-world datasets of daily stock closing prices of three major international airlines. Combining the LASSO method with multiple external data sources in our model leads to a robust and efficient method to predict stock behavior. We also compare our approach with ridge, tree, and support vector machine regressions, as well as neural network approaches to model the data. We include lags of each external variable and response variable in the model, resulting in a total of 870 predictor variables. The empirical results indicate that the LASSO-fitted model is the most effective when compared to other approaches we consider. The results show that the closing price of an airline stock is affected by its closing price for the previous days and those of other types of airlines and is significantly correlated with the Shanghai Composite Index for the previous day and 3 days prior. Other influencing factors include the positive impact of the Shanghai Composite Index daily share volume, the negative impact of loan interest rates, the amount of highway passenger and railway freight turnover, etc.

Combined Method for Informative Feature Selection for Speech Pathology Detection

The task of detecting vocal abnormalities is characterized by a small amount of available data for training, as a consequence of which classification systems that use low-dimensional data are the most relevant. We propose to use LASSO (least absolute shrinkage and selection operator) and BSS (backward stepwise selection) methods together to select the most significant features for the detection of vocal pathologies, in particular amyotrophic lateral sclerosis. Features based on fine-frequency cepstral coefficients, traditionally used in speech signal processing, and features based on discrete estimation of the autoregressive spectrum envelope are used. Spectral features based on the autoregressive process envelope spectrum are extracted using the generative method, which involves calculating a discrete Fourier transform of the report sequence generated using the autoregressive model of the input voice signal. The sequence is generated by the autoregressive model so as to account for the periodic nature of the Fourier transform. This improves the accuracy of the spectrum estimation and reduces the spectral leakage effect. Using LASSO in conjunction with BSS allowed us to improve the classification efficiency using a smaller number of features as compared to using the LASSO method alone.