Lasso Logistic Regression Research Articles

Development and validation of prediction model for fall accidents among chronic kidney disease in the community.

The population with chronic kidney disease (CKD) has significantly heightened risk of fall accidents. The aim of this study was to develop a validated risk prediction model for fall accidents among CKD in the community. Participants with CKD from the China Health and Retirement Longitudinal Study (CHARLS) were included. The study cohort underwent a random split into a training set and a validation set at a ratio of 70 to 30%. Logistic regression and LASSO regression analyses were applied to screen variables for optimal predictors in the model. A predictive model was then constructed and visually represented in a nomogram. Subsequently, the predictive performance was assessed through ROC curves, calibration curves, and decision curve analysis. A total of 911 participants were included, and the prevalence of fall accidents was 30.0% (242/911). Fall down experience, BMI, mobility, dominant handgrip, and depression were chosen as predictor factors to formulate the predictive model, visually represented in a nomogram. The AUC value of the predictive model was 0.724 (95% CI 0.679-0.769). Calibration curves and DCA indicated that the model exhibited good predictive performance. In this study, we constructed a predictive model to assess the risk of falls among individuals with CKD in the community, demonstrating good predictive capability.

A Nomogram Based on a Non-Invasive Method to Distinguish Between Gram-Positive and Gram-Negative Bacterial Infections of Liver Abscess.

The diagnosis of liver abscess (LA) caused by Gram-positive bacteria (GPB) and Gram-negative bacteria (GNB) depends on ultrasonography, but it is difficult to distinguish the overlapping features. Valuable ultrasonic (US) features were extracted to distinguish GPB-LA and GNB-LA and establish the relevant prediction model. We retrospectively analyzed seven clinical features, three laboratory indicators and 11 US features of consecutive patients with LA from April 2013 to December 2023. Patients with LA were randomly divided into training group (n=262) and validation group (n=174) according to a ratio of 6:4. Univariate logistic regression and LASSO regression were used to establish prediction models. The performance of the model was evaluated using area under the curve(AUC), calibration curves, and decision curve analysis (DCA), and subsequently validated in the validation group. A total of 436 participants (median age: 55 years; range: 42-68 years; 144 women) were evaluated, including 369 participants withGNB-LA and 67 with GPB-LA, respectively. A total of 11 predictors by LASSO regression analysis, which included gender, age, the liver background, internal gas bubble, echogenic debris, wall thickening, whether the inner wall is worm-eaten, temperature, diabetes mellitus, hepatobiliary surgery and neutrophil(NEUT). The performance of the Nomogram prediction model distinguished between GNB-LA and GPB-LA was 0.80, 95% confidence interval [CI] (0.73-0.87). In the validation group, the AUC of GNB was 0.79, 95% CI (0.69-0.89). A model for predicting the risk of GPB-LA was established to help diagnose pathogenic organism of LA earlier, which could help select sensitive antibiotics before the results of drug-sensitive culture available, thereby shorten the treatment time of patients.

Construction and Validation of an Assistant Decision-Making Model for Platelet Transfusion Refractoriness in Patients with Acute Myeloid Leukemia.

Background: Platelet transfusion refractoriness (PTR) is a complication of multiple transfusions in patients with hematological malignancies. PTR may induce a series of adverse events, such as delaying the treatment of the primary disease and life-threatening bleeding. Early prediction of PTR holds promise in facilitating prompt adjustments to treatment strategies by clinicians. Methods: We collected the clinical data of 250 patients with acute myeloid leukemia (AML). Subsequently, the patients were randomly divided into a training cohort and a validation cohort at a ratio of 7:3. The least absolute shrinkage and selection operator (LASSO) and multivariate logistic-regression methods were used to select characteristic variables. Assessment of the model was conducted through the receiver operating characteristic (ROC), calibration curve and decision curve analysis (DCA). Results: Out of 250 patients with AML, 95 individuals (38.0%) experienced PTR. Among those with positive platelet associated antibodies (PAAs), the incidence of PTR was 66.7% (30/45), while among patients positive for human leukocyte antigen(HLA)-I antibodies, the PTR incidence was 56.5% (48/85). The final predictive model incorporated risk factors such as KIT mutations, splenomegaly, the number of HLA-I antibodies, and positive PAAs. A prediction nomogram model was constructed based on these four risk factors. The LASSO-logistic regression model demonstrated excellent discrimination, calibration, and clinical decision value. Conclusion: The LASSO-logistic regression model in the study can better predict the risk of PTR. The study includes both PAAs and HLA antibodies, expanding the field of work that has not been involved in the previous prediction model of PTR.

Exploring the Application of Financial Indicator Prediction Models in Optimizing Tax Burden

Abstract As economic globalization progresses unceasingly, the financial activities of enterprises are increasingly complex, paralleled by a concomitant increase in the intricacy of tax laws and regulations. Consequently, the tax risks faced by enterprises are becoming more pronounced. To predict the financial status of enterprises, a logistic regression model is created by combining relevant financial indicators. Considering the large amount of enterprise financial sample data, a Lasso-Logistic regression model is constructed, and the ADMM algorithm optimizes the model to improve the model's prediction accuracy. Finally, after using the ROC curve to test the validity of the constructed financial index prediction model, the A-share listed companies in the advanced manufacturing industry are taken as the research object of empirical analysis to analyze the relationship between the impact of financial indexes on tax burden. The regression equation for the standardization of tax burden is tax risk = -0.02341+0.03572 VAT effective tax burden +0.15451 income tax effective tax burden +0.21118 current ratio -0.26875 total asset turnover +0.06574 current sales revenue +0.02432 current cost of goods sold +0.13681 gearing ratio + 0.23708 Fixed Assets Change Ratio - 0.26895 Asset Size. Based on the analysis results, this paper proposes three optimization strategies for tax burden.

Research on Corporate Governance and Internal Audit Wisdom Building under Financial Sharing Model Based on Logistic Modeling

Abstract In recent years, the assessment of corporate financial risk has become increasingly significant for banks. Therefore, studying models for corporate financial risk assessment holds substantial practical importance. This paper combines the Logistic model and the Lasso model based on their basic principles to construct an improved Lasso-Logistic regression model. Immediately after that, this paper selects 15 representative indexes from the four aspects of the enterprise’s profitability, solvency, operating ability, and growth ability as the indexes to respond to the company’s financial situation and extracts 4 public factors after factor analysis and analyzes them using the Lasso-Logistic regression model designed in this paper with these 4 public factors as the variables. The results show that the coefficients of public factors F1, F2, F3, and F4 are -2.9513, -1.8347, -1.9659 and -2.2714, respectively, and the coefficients of the four public factors are negative, and the classification accuracy of the Lasso-Logistic combination model in this paper is 89.46%, the misclassification rate of the first category is 6.21%, and the F1 score, RS2 score and AUC values are overall better than the two single models of Lasso and Logistic, the Lasso-Logistic model designed in this paper can well help enterprises assess their own financial risk and make targeted decisions.

An Electronic Medical Record Prediction Model to Identify Inadequate Bowel Preparation in Patients at Outpatient Colonoscopy

Background and AimsInadequate bowel preparation during colonoscopy is associated with decreased adenoma detection, increased costs, and patient procedural risks. The aim of this study was to develop a prediction model for identifying patients at high risk of inadequate bowel preparation for potential clinical integration into the electronic medical record (EMR). MethodsA retrospective study was conducted using outpatient screening/surveillance colonoscopies at the University of North Carolina from 2017 to 2022. Data were extracted from the EMRs of Epic and ProVation, including demographic, socioeconomic, and clinical variables. Logistic regression, LASSO regression, and gradient boosting machine models were evaluated and validated in a held-out testing set. ResultsThe dataset included 23,456 colonoscopies, of which 6.25% had inadequate bowel preparation. The reduced LASSO regression model demonstrated an area under the curve of 0.65 (95% CI 0.63-0.67) in the held-out testing set. The relative risk of inadequate bowel prep in the high-risk group determined by the model was 2.42 (95% CI 2.07-2.82) compared with patients identified as low risk. The model calibration in the testing set revealed that among patients categorized as having 0%-11%, 11%-22%, and 22%-33% predicted risk of inadequate prep, the respective proportions of patients with inadequate prep were 5.5%, 19.3%, and 33.3%. Using the reduced LASSO model, a rudimentary code for a potential Epic FHIR application called PrepPredict was developed. ConclusionThis study developed a prediction model for inadequate bowel preparation with the potential to integrate into the EMR for clinical use and optimize bowel preparation to improve patient care.

Transcriptional survey of abiotic stress response in maize (Zea mays) in the level of gene co-expression network and differential gene correlation analysis.

Abstract. Maize may be exposed to several abiotic stresses in the field. Therefore, identifying the tolerance mechanisms of natural field stress is mandatory. Gene expression data of maize upon abiotic stress were collected, and 560 differentially expressed genes (DEGs) were identified through meta-analysis. The most significant gene ontology terms in up-regulated genes were 'response to abiotic stress' and 'chitinase activity'. 'Phosphorelay signal transduction system' was the most significant enriched biological process in down-regulated DEGs. The co-expression analysis unveiled seven modules of DEGs, with a notable positive correlation between the modules and abiotic stress. Furthermore, the statistical significance was strikingly high for the turquoise, green and yellow modules. The turquoise group played a central role in orchestrating crucial adaptations in metabolic and stress response pathways in maize when exposed to abiotic stress. Within three up-regulated modules, Zm.7361.1.A1_at, Zm.10386.1.A1_a_at and Zm.10151.1.A1_at emerged as hub genes. These genes might introduce novel candidates implicated in stress tolerance mechanisms, warranting further comprehensive investigation and research. In parallel, the R package glmnet was applied to fit a logistic LASSO regression model on the DEGs profile to select candidate genes associated with abiotic responses in maize. The identified hub genes and LASSO regression genes were validated on an independent microarray dataset. Additionally, Differential Gene Correlation Analysis (DGCA) was performed on LASSO and hub genes to investigate the gene-gene regulatory relationship. The P value of DGCA of 16 pairwise gene comparisons was lower than 0.01, indicating a gene-gene significant change in correlation between control and abiotic stress. Integrated weighted gene correlation network analysis and logistic LASSO analysis revealed Zm.11185.1.S1_at, Zm.2331.1.S1_x_at and Zm.17003.1.S1_at. Notably, these 3 genes were identified in the 16 gene-pair comparisons. This finding highlights the notable significance of these genes in the abiotic stress response. Additional research into maize stress tolerance may focus on these three genes.

Screening biomarkers for spinal cord injury using weighted gene co-expression network analysis and machine learning.

JOURNAL/nrgr/04.03/01300535-202412000-00028/figure1/v/2024-04-08T165401Z/r/image-tiff Immune changes and inflammatory responses have been identified as central events in the pathological process of spinal cord injury. They can greatly affect nerve regeneration and functional recovery. However, there is still limited understanding of the peripheral immune inflammatory response in spinal cord injury. In this study, we obtained microRNA expression profiles from the peripheral blood of patients with spinal cord injury using high-throughput sequencing. We also obtained the mRNA expression profile of spinal cord injury patients from the Gene Expression Omnibus (GEO) database (GSE151371). We identified 54 differentially expressed microRNAs and 1656 differentially expressed genes using bioinformatics approaches. Functional enrichment analysis revealed that various common immune and inflammation-related signaling pathways, such as neutrophil extracellular trap formation pathway, T cell receptor signaling pathway, and nuclear factor-κB signal pathway, were abnormally activated or inhibited in spinal cord injury patient samples. We applied an integrated strategy that combines weighted gene co-expression network analysis, LASSO logistic regression, and SVM-RFE algorithm and identified three biomarkers associated with spinal cord injury: ANO10, BST1, and ZFP36L2. We verified the expression levels and diagnostic performance of these three genes in the original training dataset and clinical samples through the receiver operating characteristic curve. Quantitative polymerase chain reaction results showed that ANO10 and BST1 mRNA levels were increased and ZFP36L2 mRNA was decreased in the peripheral blood of spinal cord injury patients. We also constructed a small RNA-mRNA interaction network using Cytoscape. Additionally, we evaluated the proportion of 22 types of immune cells in the peripheral blood of spinal cord injury patients using the CIBERSORT tool. The proportions of naïve B cells, plasma cells, monocytes, and neutrophils were increased while the proportions of memory B cells, CD8+ T cells, resting natural killer cells, resting dendritic cells, and eosinophils were markedly decreased in spinal cord injury patients increased compared with healthy subjects, and ANO10, BST1 and ZFP26L2 were closely related to the proportion of certain immune cell types. The findings from this study provide new directions for the development of treatment strategies related to immune inflammation in spinal cord injury and suggest that ANO10, BST1, and ZFP36L2 are potential biomarkers for spinal cord injury. The study was registered in the Chinese Clinical Trial Registry (registration No. ChiCTR2200066985, December 12, 2022).

Comparison of LASSO and random forest models for predicting the risk of premature coronary artery disease

PurposeWith the change of lifestyle, the occurrence of coronary artery disease presents a younger trend, increasing the medical and economic burden on the family and society. To reduce the burden caused by this disease, this study applied LASSO Logistic Regression and Random Forest to establish a risk prediction model for premature coronary artery disease(PCAD) separately and compared the predictive performance of the two models.MethodsThe data are obtained from 1004 patients with coronary artery disease admitted to a third-class hospital in Liaoning Province from September 2019 to December 2021. The data from 797 patients were ultimately evaluated. The dataset of 797 patients was randomly divided into the training set (569 persons) and the validation set (228 persons) scale by 7:3. The risk prediction model was established and compared by LASSO Logistic and Random Forest.ResultThe two models in this study showed that hyperuricemia, chronic renal disease, carotid artery atherosclerosis were important predictors of premature coronary artery disease. A result of the AUC between the two models showed statistical difference (Z = 3.47, P < 0.05).ConclusionsRandom Forest has better prediction performance for PCAD and is suitable for clinical practice. It can provide an objective reference for the early screening and diagnosis of premature coronary artery disease, guide clinical decision-making and promote disease prevention.

Pyroptosis-related signatures predict immune characteristics and prognosis in IPF

The purpose of this work was to use integrated bioinformatics analysis to screen for pyroptosis-related genes (PRGs) and possible immunological phenotypes linked to the development and course of IPF. Transcriptome sequencing datasets GSE70866, GSE47460 and GSE150910 were obtained from GEO database. From the GSE70866 database, 34 PRGs with differential expression were found in IPF as compared to healthy controls. In addition, a diagnostic model containing 4 genes PRGs (CAMP, MKI67, TCEA3 and USP24) was constructed based on LASSO logistic regression. The diagnostic model showed good predictive ability to differentiate between IPF and healthy, with ROC-AUC ranging from 0.910 to 0.997 in GSE70866 and GSE150910 datasets. Moreover, based on a combined cohort of the Freiburg and the Siena cohorts from GSE70866 dataset, we identified ten PRGs that might predict prognosis for IPF. We constructed a prognostic model that included eight PRGs (CLEC5A, TREM2, MMP1, IRF2, SEZ6L2, ADORA3, NOS2, USP24) by LASSO Cox regression and validated it in the Leuven cohort. The risk model divided IPF patients from the combined cohort into high-risk and low-risk subgroups. There were significant differences between the two subgroups in terms of IPF survival and GAP stage. There is a close correlation between leukocyte migration, plasma membrane junction, and poor prognosis in a high-risk subgroup. Furthermore, a high-risk score was associated with more plasma cells, activated NK cells, monocytes, and activated mast cells. Additionally, we identified HDAC inhibitors in the cMAP database that might be therapeutic for IPF. To summarize, pyroptosis and its underlying immunological features are to blame for the onset and progression of IPF. PRG-based predictive models and drugs may offer new treatment options for IPF.

Identification of the transcriptome signatures and immune-inflammatory responses in postmenopausal osteoporosis

Postmenopausal osteoporosis is the most common type of osteoporosis in women. To date, little is known about their transcriptome signatures, although biomarkers from peripheral blood mononuclear cells are attractive for postmenopausal osteoporosis diagnoses. Here, we performed bulk RNA sequencing of 206 samples (124 postmenopausal osteoporosis and 82 normal samples) and described the clinical phenotypic characteristics of postmenopausal women. We then highlighted the gene set enrichment analyses between the extreme T-score group and the heathy control group, revealing that some immune-inflammatory responses were enhanced in postmenopausal osteoporosis, with representative pathways including the mitogen-activated protein kinase (NES = 1.6, FDR <0.11) pathway and B_CELL_RECEPTOR (NES = 1.69, FDR <0.15) pathway. Finally, we developed a combined risk prediction model based on lasso-logistic regression to predict postmenopausal osteoporosis, which combined eleven genes (PTGS2, CXCL16, NECAP1, RPS23, SSR3, CD74, IL4R, BTBD2, PIGS, LILRA2, MAP3K11) and three pieces of clinical information (age, procollagen I N-terminal propeptide, β isomer of C-terminal telopeptide of type I) and provided the best prediction ability (AUC = 0.97). Taken together, this study filled a gap in the large-scale transcriptome signature profiles and revealed the close relationship between immune-inflammatory responses and postmenopausal osteoporosis, providing a unique perspective for understanding the occurrence and development of postmenopausal osteoporosis.

Establishment and validation of a prediction model for gestational diabetes.

To develop a visual prediction model for gestational diabetes (GD) in pregnant women and to establish an effective and practical tool for clinical application. To establish a prediction model, the modelling set included 1756 women enrolled in the Zunyi birth cohort, the internal validation set included 1234 enrolled women, and pregnant women in the Wuhan cohort were included in the external validation set. We established a demographic-lifestyle factor model (DLFM) and a demographic-lifestyle-environmental pollution factor model (DLEFM) based on whether the women were exposed to environmental pollutants. The least absolute shrinkage and selection lasso-logistic regression analyses were used to identify the independent predictors of GD and construct a nomogram for predicting its occurrence. The DLEFM regression analysis showed that a family history of diabetes (odd ratio [OR] 2.28; 95% confidence interval [CI] 1.05-4.71), a history of GDin pregnant women (OR 4.22; 95% CI 1.89-9.41), being overweight or obese before pregnancy (OR 1.71; 95% CI 1.27-2.29), a history of hypertension (OR 2.61; 95% CI 1.41-4.72), sedentary time (h/day) (OR 1.16; 95% CI 1.08-1.24), monobenzyl phthalate (OR 1.95; 95% CI 1.45-2.67) and Q4 mono-ethyl phthalate concentration (OR 1.85; 95% CI 1.26-2.73) were independent predictors. The area under the receiver operating curves for the internal validation of the DLEFM and the DLFM constructed using these seven factors was 0.827 and 0.783, respectively. The calibration curve of the DLEFM was close to the diagonal line. The DLEFM was thus the more optimal model, and the one which we chose. A nomogram based on preconception factors was constructed to predict the occurrence of GD in the second and third trimesters. It provided an effective tool for the early prediction and timely management of GD.

Development and validation of primary graft dysfunction predictive algorithm for lung transplant candidates

PurposePrimary graft dysfunction (PGD) is the leading cause of early morbidity and mortality after lung transplantation. Accurate prediction of PGD risk could inform donor approaches and peri-operative care planning. We sought to develop a clinically useful, generalizable PGD prediction model to aid in transplant decision making. MethodsWe derived a predictive model in a prospective cohort study of subjects from 2012-2018, followed by a single-center external validation. We used regularized (lasso) logistic regression to evaluate the predictive ability of clinically-available PGD predictors and developed a user interface for clinical application. Using decision curve analysis (DCA), we quantified the net benefit of the model across a range of PGD risk thresholds and assessed model calibration and discrimination. ResultsThe PGD predictive model included distance from donor hospital to recipient transplant center, recipient age, predicted total lung capacity, lung allocation score (LAS), body mass index, pulmonary artery mean pressure, sex, and indication for transplant; donor age, sex, mechanism of death, and donor smoking status; and interaction terms for LAS and donor distance. The interface allows for real-time assessment of PGD risk for any donor/recipient combination. The model offers decision making net benefit in the PGD risk range of 10-75% in the derivation centers and 2-10% in the validation cohort, a range incorporating the incidence in that cohort. ConclusionWe developed a clinically useful PGD predictive algorithm across a range of PGD risk thresholds to support transplant decision making, post-transplant care, and enrich samples for PGD treatment trials.

Association of Impedance Aggregometry-Measured Platelet Aggregation With Thromboembolic Events in Patients Who Undergo Carotid Endarterectomy: A Pilot Study

ObjectivesThe aim of the current study was to assess the relationship between thrombin receptor activator peptide 6 (TRAP test), adenosine-5′-diphosphate (ADP test) and arachidonic acid (ASPI test) and stroke/TIA, using the multiple electrode aggregometry (Multiplate®) in patients undergoing CEA. DesignA retrospective study. SettingVascular surgery operating rooms of university hospital. ParticipantsOne hundred and thirty-one out of 474 patients undergoing CEA between November 2020 and October 2022. InterventionsNone. Measurements & Main ResultsA pre-operative blood sample of all enrolled patients was analyzed using the multiplate analyzer. Receiver operating characteristics curves were generated to test the ability of TRAP, ADP and ASPI in discriminating perioperative thromboembolic stroke/TIA. A logistic LASSO regression model was used to identify factors independently associated with stroke/TIA. Eight patients suffered of perioperative stroke/TIA. Although all the platelet functional assays showed an excellent predictive performance an ADP value exciding 72 U showed the highest specificity (87%) and sensitivity (68%) in discriminating patients who had perioperative thromboembolic stroke/TIA with a negative predictive value of 99% and a positive predictive value of 15%. After LASSO regression an ADP> 72 U and need of shunt during CEA were the only two variables independently associated with perioperative stroke/TIA. ConclusionSince the ADP test was independently associated with perioperative stroke/TIA, the assessment of platelet reactivity using Multiplate® may offer potential utility in monitoring patients undergoing CEA.

Predictors of clinical worsening during a discontinuation trial of serotonin reuptake inhibitors for obsessive compulsive disorder

ObjectiveTo explore predictors and moderators of clinical worsening during a double-blind trial in which patients with obsessive-compulsive disorder (OCD) were randomized to either continue or discontinue their Serotonin Reuptake Inhibitor (SRI) medication after achieving wellness from the addition of exposure and response prevention (EX/RP) therapy. MethodThe data came from a double-blind discontinuation trial that included N = 101 participants, 35 of whom were removed from the study due to clinical worsening. We first used LASSO logistic regression to identify which of the 34 potential baseline variables of interest (including demographics, diagnoses, other relevant clinical constructs, and specific genotypes), might moderate or predict this clinical worsening. Then logistic regression was used to examine which of these identified variables were significantly related to later clinical worsening. We verified the validity of our final prediction model using k-fold cross-validation. ResultsThere was one significant predictor of clinical worsening: In both groups, those with more past diagnoses had a greater likelihood of clinical worsening (p = .015). There were several moderators. Rates of clinical worsening were higher in the Discontinuation group compared to the Continuation group for participants who were taking a shorter half-life SRI (p = .044), were female (p = .022), had higher baseline levels of maladaptive metacognitions (p < .001), had fewer sleep problems at baseline (p = .001), and/or had more years of education (p < .001). ConclusionsOur results identified several factors that may predict the development of clinical worsening in OCD patients discontinuing SRI medication following successful EX/RP treatment.

Development and Validation of a Risk-Prediction Nomogram for Preoperative Blood Type and Antibody Testing in Spinal Fusion Surgery.

With advancements in minimally invasive techniques, the use of spinal fusion surgery is rapidly increasing and transfusion rates are decreasing. Routine preoperative ABO/Rh blood type and antibody screening (T&S) laboratory tests may not be appropriate for all spinal fusion patients. Herein, we constructed a nomogram to assess patient transfusion risk based on various risk factors in patients undergoing spinal fusion surgery, so that preoperative T&S testing can be selectively scheduled in appropriate patients to reduce healthcare and patient costs. Patients who underwent spinal fusion surgery between 01/2020 and 03/2023 were retrospectively examined and classified into the training (n = 3533, 70%) and validation (n = 1515, 30%) datasets. LASSO and multivariable logistic regression were used to analyze risk factors for blood transfusion. Nomogram predictive model was built according to the independent predictors and mode predictive power was validated using consistency index (C-index), Hosmer-Lemeshow (HL) test, calibration curve analysis and area under the curve (AUC) for receiver operating characteristic (ROC) curve. Bootstrap resampling was used for internal validation. Decision curve analysis (DCA) was applied to evaluate the model's performance in the clinic. Being female, age, BMI, admission route, critical patient, operative time, heart failure, end-stage renal disease or chronic kidney disease (ESRD or CKD), anemia, and coagulation defect were predictors of blood transfusion for spinal fusion. A prediction nomogram was developed according to a multivariate model with good discriminatory power (C-index = 0.887); Bootstrap resampling internal validation C-index was 0.883. Calibration curves showed strong matching between the predicted and actual probabilities of the training and validation sets. HL tests for the training and validation sets had p-values of 0.327 and 0.179, respectively, indicating good calibration. When applied to the training set, the following parameters were found: AUC: 0.895, 95% CI: 0.871-0.919, sensitivity 78.2%, specificity 86.7%, positive predictive value 29.4% and negative predictive value 98.2%. If the model were applied in the training set, 2911 T&S tests (82.4%) would be eliminated, equaling a RMB349,320 cost reduction. The AUC in the internal validation was: 0.879, 95% CI: 0.839-0.927, sensitivity 75.2%, specificity 88.8%, positive predictive value 34.3%, negative predictive value 97.9%, would eliminate 1276 T&S tests (84.2%), saving RMB 153,120. The DCA curve indicated good clinical application value. The nomogram based on 10 independent factors can help healthcare professionals predict the risk of transfusion for patients undergoing spinal fusion surgery to target preoperative T&S testing to appropriate patients and reduce healthcare costs.

Predicting progression to Alzheimer’s disease dementia using cognitive measures: findings from the Essex Memory Clinic

AbstractBackgroundTimely diagnosis of Alzheimer’s disease (AD) is important for patients, carers and healthcare systems. It is therefore imperative to identify measures with the greatest power to predict future AD dementia. In contrast to research settings, expensive biomarkers are not always available in memory clinics. It is therefore vital to identify predictive measures which can be used in ordinary clinical settings.MethodThis study utilized clinical assessment data from 328 patients assessed in the Essex Memory Clinic between 2002 and 2019. Only patients with a diagnosis of subjective cognitive decline (SCD) or mild cognitive impairment (MCI) at baseline and a minimum of two follow‐up visits were included in analyses. Lasso logistic regression (using AD at visit 3 as the outcome) and Lasso Cox proportional hazards models (using time to AD at visit 3 and beyond as outcome) were fitted in two stages. The first stage included baseline cognitive scores only, while the second stage also included change scores for these measures (derived by subtracting scores at baseline from those at the first follow‐up visit).ResultIn the first stage models, lower Mini‐Mental State Examination (MMSE) and Logical Memory Test (LMT) delayed recall scores, and longer Trail‐Making Test part A (TMT‐A) time were associated with greater odds (logistic regression) and hazards (Cox proportional hazards) of progression to AD. In addition, in the Cox model (only), lower baseline Cambridge Cognitive Examination‐Revised (CAMCOG‐R) score was associated with greater hazard of AD. In the second stage models, worsening CAMCOG‐R and LMT (delayed recall) scores were independently associated with both the odds/hazards of AD.ConclusionIn this memory clinic cohort, the cognitive scores most predictive of progression to AD were the delayed LMT, TMT‐A, CAMCOG‐R and MMSE. These results are in‐keeping with earlier work conducted in research cohorts. The inclusion of cognitive change scores over one year follow‐up improved predictive accuracy versus baseline scores alone.

#57 : A Prediction Model of Low Birthweight in Singleton Pregnancies Reduced from Dichorionic Twin Pregnancies

Background and Aims: The incidence of twin pregnancies has risen over the last few decades, mainly attributed to assisted reproduction techniques (ART). Fetal reduction (FR) can significantly reduce low birthweight (LBW) risk in twins. However, the LBW risk is still higher in singletons reduced from twins than for primary singletons. Many factors, such as maternal factors, FR timing and approach may affect LBW risks after FR and no relevant prediction models have been reported to date. The main objective of this study is to develop a nomogram to predict LBW risk in singleton pregnancies reduced from dichorionic (DC) twins. Method: We retrospectively reviewed and analyzed the data on women with (DC) twin pregnancies who underwent FR in Women’s Hospital, Zhejiang university, School of medicine in July 2005 to August 2021. Logistic LASSO regression was used to identify the most relevant variables associated with LBW. The nomogram was constructed and receiver operating characteristic (ROC) curve, Hosmer-Lemeshow test, and decision clinical analysis were used for model performances assessment and visualization. The model was evaluated by cohorts produced by 500 resampling bootstrap analysis to testify the stability. Results: A total of 471 patients were enrolled in the analysis. At last, seven independent predictive factors for LBW were identified and integrated to construct the nomograms, including maternal height, nulliparous, conception mode, reasons for FR, gestational age at FR, gestational diabetes, pregnancy hypertensive disease. The AUC of our prediction model was 0.793, which was validated by internal (AUC 0.762) confirmation with bootstrap analysis. The nomogram had well-fitted calibration curves. Decision curve analysis demonstrated that the nomogram was clinically useful. Conclusion: We made the first attempt to develop a reliable predictive nomogram for the risk of LBW in DC twin pregnancies reduced to singletons, providing a useful guide for clinicians and patients in making appropriate FR decisions.

Development and validation of a nomogram for 30-day readmission after hip fracture surgery in geriatric patients.

30-day readmission after hip fracture surgery in the elderly is common and costly. A predictive tool to identify high-risk patients could significantly improve outcomes. This study aims to develop and validate a risk nomogram for 30-day readmission after hip fracture surgery in geriatric patients. We retrospectively analyzed 1,249 geriatric hip fracture patients (≥60 years) undergoing surgery at Dandong Central Hospital from October 2011 to October 2023. Using a 7:3 ratio, patients were randomly divided into training (n=877) and validation (n=372) sets. Independent risk factors for 30-day readmission were identified using LASSO regression and logistic regression in the training set. A nomogram was constructed using the identified predictors. Finally, the C-index, ROC curve, calibration curve, and decision curve analysis were used to validate the model in the training and validation sets respectively. The nomogram was developed based on the 8 predictors of age, prior stroke, chronic liver disease, treatment, uric acid (UA), total protein (TP), albumin (ALB), and pneumonia that were found to be independently associated with 30-day readmission. The nomogram showed good discrimination with a C-index of 0.88 in the training set and 0.84 in the validation set. Calibration curves exhibited good agreement between predicted and observed outcomes. Decision curve analysis demonstrated clinical utility. We developed and validated a nomogram incorporating eight clinical variables to accurately predict the individualized risk of 30-day readmission after hip fracture surgery in elderly patients. The model demonstrated favorable discrimination, calibration, and clinical utility. It can help to identify high-risk patients needing additional interventions to prevent avoidable hospital readmissions.

Tau‐PET to predict conversion from mild cognitive impairment to dementia; a head‐to‐head comparison against amyloid‐PET and MRI

AbstractBackgroundIn contrast to amyloid‐PET and MRI, tau‐PET imaging has not been integrated into clinical practice even though it shows a close relationship with clinical symptoms. With the present study, we aim to assess the performance of tau‐PET to predict conversion to dementia (all‐cause and AD) in individuals with mild cognitive impairment (MCI), and compare it with amyloid‐PET and MRI.MethodWe selected participants from an ongoing multicenter study (n = 408, Table 1). The following ROIs were used to define quantified thresholds for positivity using an independent cognitively unimpaired reference group: a temporal meta‐region for tau‐PET and an AD‐susceptible temporoparietal region for MRI‐based cortical thickness. For amyloid‐PET, we used the standard Centiloid mask and a published threshold of 27 Centiloids to determine positivity. We additionally performed visual reads for all imaging modalities. We used Cox proportional hazard, logistic regression, and receiver operating curve (ROC) analyses to assess the performance of amyloid‐PET (A), tau‐PET (T), and MRI (N) to predict progression from MCI to dementia over a mean follow‐up period of 2 years.ResultHazard ratios indicated that tau‐PET and amyloid‐PET positivity (both when assessed quantitatively and using visual reads) conveyed a greater risk of progression to dementia than MRI, and that a combination of A+ and T+ (or N+) conveyed a greater risk than A‐T‐N‐ or isolated A+ or T+ (Figure 1). ROC analyses revealed that only tau‐PET performed better than demographics alone when predicting all‐cause dementia while both amyloid‐ and tau‐PET performed better than demographics and MRI when predicting AD dementia. LASSO logistic regression revealed that the optimal model to predict conversion from MCI to all‐cause dementia included a quantitative MRI marker and tau‐PET visual read. For optimal prediction of AD dementia specifically, the quantitative measure of amyloid‐PET was added in addition to the quantitative MRI marker and the tau‐PET visual read (Table 2).ConclusionAmong individuals with MCI, tau‐PET performs best when predicting dementia, and can be optimized when combined with MRI. For AD dementia, tau‐PET and amyloid‐PET are the best‐performing modalities, but an optimal prediction is again achieved when combined with MRI.