Construction and validation of a nomogram prediction model for the catheter-related thrombosis risk of central venous access devices in patients with cancer: a prospective machine learning study.

Central venous access devices (CVADs) are integral to cancer treatment. However, catheter-related thrombosis (CRT) poses a considerable risk to patient safety. It interrupts treatment; delays therapy; prolongs hospitalisation; and increases the physical, psychological and financial burden of patients. Our study aims to construct and validate a predictive model for CRT risk in patients with cancer. It offers the possibility to identify independent risk factors for CRT and prevent CRT in patients with cancer. We prospectively followed patients with cancer and CVAD at Xiangya Hospital of Central South University from January 2021 to December 2022 until catheter removal. Patients with CRT who met the criteria were taken as the case group. Two patients with cancer but without CRT diagnosed in the same month that a patient with cancer and CRT was diagnosed were selected by using a random number table to form a control group. Data from patients with CVAD placement in Qinghai University Affiliated Hospital and Hainan Provincial People's Hospital (January 2023 to June 2023) were used for the external validation of the optimal model. The incidence rate of CRT in patients with cancer was 5.02% (539/10 736). Amongst different malignant tumour types, head and neck (9.66%), haematological (6.97%) and respiratory (6.58%) tumours had the highest risks. Amongst catheter types, haemodialysis (13.91%), central venous (8.39%) and peripherally inserted central (4.68%) catheters were associated with the highest risks. A total of 500 patients with CRT and 1000 without CRT participated in model construction and were randomly assigned to the training (n = 1050) or testing (n = 450) groups. We identified 11 independent risk factors, including age, catheterisation method, catheter valve, catheter material, infection, insertion history, D-dimer concentration, operation history, anaemia, diabetes and targeted drugs. The logistic regression model had the best discriminative ability amongst the three models. It had an area under the curve (AUC) of 0.868 (0.846-0.890) for the training group. The external validation AUC was 0.708 (0.618-0.797). The calibration curve of the nomogram model was consistent with the ideal curve. Moreover, the Hosmer-Lemeshow test showed a good fit (P > 0.05) and high net benefit value for the clinical decision curve. The nomogram model constructed in this study can predict the risk of CRT in patients with cancer. It can help in the early identification and screening of patients at high risk of cancer CRT.

Stem Growth and Dehydration Responses of Mediterranean Tree Species to Atmospheric and Soil Drought.

Stem growth responses to soil and atmospheric drought are critical to forecasting the tree carbon sink strength. Yet, responses of drought-prone forests remain uncertain despite global aridification trends. Stem diameter variations at an hourly resolution were monitored in five Mediterranean tree species from a mesic and a xeric site for 6 and 12 years. Stem growth and dehydration responses to soil (REW) and atmospheric (VPD) drought were explored at different timescales. Annually, growth was determined by the number of growing days and hours. Seasonally, growth was bimodal (autumn growth ≈ 8%-18% of annual growth), varying among species and sites across the hydrometeorological space, while dehydration consistently responded to REW. Sub-daily, substantial growth occurred during daytime, with nighttime-to-daytime ratios ranging between 1.2 and 3.5 (Arbutus unedo ≈ Quercus faginea < Quercus ilex < Pinus halepensis in the mesic site, and Juniperus thurifera < P. halepensis in the xeric site). Overall, time windows favourable for growth were limited by soil (rather than atmospheric) drought, modulating annual and seasonal growth in Mediterranean species, and stems maintained non-negligible growth during daytime. These patterns contrast with observations from wetter or cooler biomes, demonstrating the growth plasticity of drought-prone species to more arid climate conditions.

Inference on the Eigenvalues of the Normalized Precision Matrix.

Recent developments in the spectral theory of Bayesian Networks has led to a need for a developed theory of estimation and inference on the eigenvalues of the normalized precision matrix, Ω . In this paper, working under conditions where n → ∞ and p remains fixed, we provide multivariate normal asymptotic distributions of the sample eigenvalues of Ω under general conditions and under normal populations, a formula for second-order bias correction of these sample eigenvalues, and a Stein-type shrinkage estimator of the eigenvalues. Numerical simulations are performed which demonstrate under what generative conditions each estimation technique is most effective. When the largest eigenvalue of Ω is small the simulations show that the second order bias-corrected eigenvalue was considerably less biased than the sample eigenvalue, whereas the smallest eigenvalue was estimated with less bias using either the sample eigenvalue or the proposed shrinkage method.

What's behind Margin Status in Oral Cancer?

In the 2nd century AD, Galen argued that the failure to remove any single 'root' of a malignant tumor could result in a local relapse. After nearly 2 millennia, this problem appears to be even more challenging due to our increased understanding of the complexity of tumor formation and spread. Pathological analysis of tumor margins under a microscope remains the primary and only accepted method for confirming the complete tumor removal. However, this method is not an all-or-nothing test, and it can be compromised by various intrinsic and extrinsic limitations. Among the intrinsic limitations of pathological analysis we recall the pathologist handling, tissue shrinkage, the detection of minimal residual disease and the persistence of a precancerous field. Extrinsic limitations relate to surgical tools and their thermal damage, the different kinds of surgical resections and frozen sections collection. Surgeons, as well as oncologists and radiotherapists, should be well aware of and deeply understand these limitations to avoid misinterpretation of margin status, which can have serious consequences. Meanwhile, new technologies such as Narrow band imaging have shown promising results in assisting with the achievement of clear superficial resection margins. More recently, emerging techniques like Raman spectroscopy and near-infrared fluorescence have shown potential as real-time guides for surgical resection. The aim of this narrative review is to provide valuable insights into the complex process of margin analysis and underscore the importance of interdisciplinary collaboration between pathologists, surgeons, oncologists, and radiotherapists to optimize patient outcomes in oral cancer surgery.

Postmenopausal Shrinkage of Uterine Myomas: A Retrospective Study of 97 Cases Monitored Annually for 10 Years.

Introduction and aim Both patients and gynecologists are concerned about how much and how quickly myomas shrink after menopause. This study aimed to elucidate clinical findings that may be associated with postmenopausal shrinkage of uterine myomas. Materials and methods This study included 97 patients who underwent menopause by August 2012, had myoma nodules with the longest diameter between 50 mm and 160 mm, and visited our specialized myoma clinic annually for at least 10 years after menopause. They underwent transabdominal ultrasonography at least once per year. An experienced gynecologist measured the longest diameter of myoma nodules with a maximum diameter between 50 mm and 160 mm. The shrinkage rate of myoma diameters after menopause compared to premenopausal diameters was calculated each year for 10 years. The shrinkage rate of the longest diameter of the largest nodule 10 years after menopause (10-year shrinkage rate) and its relationship with clinical findings (the age at menopause, parity, body mass index {BMI}, number of nodules, MRI findings on T2-weighted image, location of the nodule, and longest diameter of the largest nodule before menopause) were analyzed. Additionally, we examined annual changes in shrinkage rate of myomas over a 10-year period after menopause (annual trend), and the relationship between annual trends and factors such as BMI and the number of nodules. Results In this examination of 10-year shrinkage rate, the group with a BMI of less than 25 showed a significantly greater shrinkage rate compared to the group with a BMI of 25 or more (25.0% vs 15.7%, p=0.023). Additionally, the group with a single nodule showed a significantly greater 10-year shrinkage rate compared to the group with four or more nodules (26.3% vs 15.2%, p=0.036). For annual trends, the rate of change in the first two years after menopause was significantly faster compared to the trend from the third to the 10th year (difference in slope: 3.888 points per year, p<0.001). When divided into two groups based on the number of nodules (one or two nodules group and three or more nodules group), the group with one or two nodules showed a significant difference in the shrinkage rate between up to two years after menopause and from the period from the third to the 10th year (difference in slope: 4.590 points per year, p<0.001). However, for the group with three or more nodules, there was no significant difference in the annual trend between the first two years after menopause and the rate from the third to the 10th year (difference in slope: 1.626 points per year, p=0.107). Conclusion BMI and the number of myoma nodules were significantly related to the 10-year shrinkage rate. Although myomas shrank significantly faster within the first two years after menopause compared to the later period, the early annual trend did not differ significantly from the trend in the later period when there were multiple nodules with a maximum diameter of 50 mm or more.

Development and validation of a nomogram for premature coronary artery disease patients in Guangzhou.

Background: Data regarding risk factors for premature coronary artery disease (PCAD) is scarce given that few research focus on it. This study aimed to develop and validate a clinical nomogram for PCAD patients in Guangzhou. Methods: We recruited 108 PCAD patients (female ≤65 years old and male ≤55 years old) and 96 healthy controls from Sun Yat-sen Memorial Hospital of Sun Yat-sen University between 01/01/2021 and 31/12/2022. Twenty potentially relevant indicators of PCAD were extracted. Next, the least absolute shrinkage and selection operator (LASSO) regression analysis was used to optimize variable selection. The nomogram was developed based on the selected variables visually. Results: Independent risk factors, including body mass index (BMI), history of PCAD, glucose, Apolipoprotein A1(ApoA1), high density lipoprotein 2-cholesterol (HDL2-C), total cholesterol and triglyceride, were identified by LASSO and logistic regression analysis. The nomogram showed accurate discrimination (area under the receiver operator characteristic curve, ROC, 87.45 %, 95 % CI: 82.58 %-92.32 %). Decision curve analysis (DCA) suggested that the nomogram was clinical beneficial. HDL2, one risk factor, was isolated by a two-step discontinuous density-gradient ultracentrifugation method. And HDL2 from PCAD patients exhibited less 3H-cholesterol efflux (22.17 % vs 26.64 %, P < 0.05) and less delivery of NBD-cholesterol detecting by confocal microscope compared with healthy controls. Conclusions: In conclusion, the seven-factor nomogram can achieve a reasonable relationship with PCAD, and a large cohort were needed to enhance the credibility and effectiveness of our model in future.

3D micro-CT and O-PTIR spectroscopy bring new understanding of the influence of filler content in dental resin composites.

BACKGROUND: Dental resin composites' performance is intricately linked to their polymerisation shrinkage characteristics. This study compares polymerisation shrinkage using advanced 3D micro-computed tomography (micro-CT) and traditional 2D linear assessments. It delves into the crucial role of filler content on shrinkage and the degree of conversion in dental resin composites, providing valuable insights for the field. METHODS: Five experimental dental composite materials were prepared with increasing filler contents (55-75 wt%) and analysed using either 3D micro-CT for volumetric shrinkage or a custom-designed linometer for 2D linear shrinkage. The degree of conversion was assessed using Optical Photothermal Infrared (O-PTIR) and Fourier-Transform Infrared (FTIR) spectroscopy. Light transmittance through a 2-mm layer was evaluated using a NIST-calibrated spectrometer. Scanning Electron Microscopy (SEM) and Energy-Dispersive X-ray Spectroscopy (EDX) examined surface morphology and elemental distribution. Correlation between the investigated parameters was determined using Spearman correlation analyses. RESULTS: The study found significant differences in polymerisation-related properties among different filler content categories, with volumetric shrinkage consistently demonstrating higher mean values than linear shrinkage across most groups. Volumetric shrinkage decreased with increasing curing depth, showing no direct correlation between filler content and shrinkage levels at different curing depths. The results highlighted a strong negative correlation between filler content and degree of conversion, volumetric and linear shrinkage, as well as maximum shrinkage rate. Light transmittance showed a moderate correlation with the filler content and a weak correlation with other tested parameters. CONCLUSIONS: This study underscores the importance of considering both volumetric and linear shrinkage in the design and analysis of dental composite materials. The findings advocate optimising filler content to minimise shrinkage and enhance material performance. Integrating micro-CT and O-PTIR techniques offers novel insights into dental composites' polymerisation behaviour, providing a foundation for future research to develop materials with improved clinical outcomes.

Emerging technologies for the evaluation of spatio-temporal polymerisation changes in flowable vs. sculptable dental resin-based composites.

BACKGROUND: This study presents a novel multi-technique approach that integrates micro-CT and optical photothermal infrared spectroscopy (O-PTIR) to evaluate polymerisation differences, so-called spatio-temporal polymerisation properties, between flowable and sculptable dental resin-based composites. METHODS: Ten commercially available dental composites were investigated, including flowable and sculptable counterparts from the same manufacturer. Eight parameters were evaluated: short-term polymerisation characteristics (degree of conversion after 5 min, maximum polymerisation rate, time to reach maximum polymerisation rate) was measured using ATR-FTIR with real-time monitoring; changes in the degree of conversion with depth were evaluated with O-PTIR, 3D visualisation of shrinkage patterns, overall volumetric shrinkage, depth-specific shrinkage, and porosity were measured using micro-CT; surface morphology with detailed measurements of elemental composition was characterised using SEM/EDX; light transmittance was analysed with a NIST-referenced spectrometer. RESULTS: The study found that the increase in filler weight and volume ratio reduced the degree of conversion and polymerisation shrinkage, while moderately influencing the maximum polymerisation rates. The time to reach maximum polymerisation rates and light transmittance were not dependent on the filler amount. O-PTIR assessed a depth-dependent decrease in the degree of conversion for both composite types, with flowable composites generally showing a greater decrease in the degree of conversion than sculptable composites, except for bulk-fill composites. Micro-CT scans showed significantly higher flowable shrinkage values than their sculptable counterparts, highlighting the performance differences between the two types of composites. CONCLUSIONS: The findings of this study have practical implications for the selection and use of dental composites. Flowable composites, despite their higher degrees of conversion and polymerisation rates, also exhibit higher volumetric shrinkage, which can be detrimental for clinical applications. The new measurement methods used in this study provide a comprehensive overview of the polymerisation behaviour of commercially available dental composites, offering valuable insights for material optimisation.

Apoptosis-Cell Cycle-Autophagy Molecular Mechanisms Network in Heterogeneous Aggressive Phenotype Prostate Hyperplasia Primary Cell Cultures Have a Prognostic Role.

Our study highlights the apoptosis, cell cycle, DNA ploidy, and autophagy molecular mechanisms network to identify prostate pathogenesis and its prognostic role. Caspase 3/7 expressions, cell cycle, adhesion glycoproteins, autophagy, nuclear shrinkage, and oxidative stress by flow-cytometry analysis are used to study the BPH microenvironment's heterogeneity. A high late apoptosis expression by caspases 3/7 activity represents an unfavorable prognostic biomarker, a dependent predictor factor for cell adhesion, growth inhibition by arrest in the G2/M phase, and oxidative stress processes network. The heterogeneous aggressive phenotype prostate adenoma primary cell cultures present a high S-phase category (>12%), with an increased risk of death or recurrence due to aneuploid status presence, representing an unfavorable prognostic biomarker, a dependent predictor factor for caspase 3/7 activity (late apoptosis and necrosis), and cell growth inhibition (G2/M arrest)-linked mechanisms. Increased integrin levels in heterogenous BPH cultures suggest epithelial-mesenchymal transition (EMT) that maintains an aggressive phenotype by escaping cell apoptosis, leading to the cell proliferation necessary in prostate cancer (PCa) development. As predictor biomarkers, the biological mechanisms network involved in apoptosis, the cell cycle, and autophagy help to establish patient prognostic survival or target cancer therapy development.

A Precise Prediction of the Chemical and Thermal Shrinkage during Curing of an Epoxy Resin.

A precise prediction of the cure-induced shrinkage of an epoxy resin is performed using a finite element simulation procedure for the material behaviour. A series of experiments investigating the cure shrinkage of the resin system has shown a variation in the measured cure-induced strains. The observed variation results from the thermal history during the pre-cure. A proposed complex thermal expansion model and a conventional chemical shrinkage model are utilised to predict the cure shrinkage observed with finite element simulations. The thermal expansion model is fitted to measured data and considers material effects such as the glass transition temperature and the evolution of the expansion with the degree of cure. The simulations accurately capture the exothermal heat release from the resin and the cure-induced strains across various temperature profiles. The simulations follow the experimentally observed behaviour. The simulation predictions achieve good accuracy with 2-6% discrepancy compared with the experimentally measured shrinkage over a wide range of cure profiles. Demonstrating that the proposed complex thermal expansion model affects the potential to minimise the shrinkage of the studied epoxy resin. A recommendation of material parameters necessary to accurately determine cure shrinkage is listed. These parameters are required to predict cure shrinkage, allow for possible minimisation, and optimise cure profiles for the investigated resin system. Furthermore, in a study where the resin movement is restrained and therefore able to build up residual stresses, these parameters can describe the cure contribution of the residual stresses in a component.

Comparison of Hybrid Machine Learning Approaches for Surrogate Modeling Part Shrinkage in Injection Molding.

Machine learning (ML) methods present a valuable opportunity for modeling the non-linear behavior of the injection molding process. They have the potential to predict how various process and material parameters affect the quality of the resulting parts. However, the dynamic nature of the injection molding process and the challenges associated with collecting process data remain significant obstacles for the application of ML methods. To address this, within this study, hybrid approaches are compared that combine process data with additional process knowledge, such as constitutive equations and high-fidelity numerical simulations. The hybrid modeling approaches include feature learning, fine-tuning, delta-modeling, preprocessing, and using physical constraints, as well as combinations of the individual approaches. To train and validate the hybrid models, both the experimental and simulated shrinkage data of an injection-molded part are utilized. While all hybrid approaches outperform the purely data-based model, the fine-tuning approach yields the best result in the simulation setting. The combination of calibrating a physical model (feature learning) and incorporating it implicitly into the training process (physical constraints) outperforms the other approaches in the experimental setting.

Mechanical and Drying Shrinkage Performance Study of Ultra-High-Performance Concrete Prepared from Titanium Slag under Different Curing Conditions.

This research investigates the effects of various curing regimes, the incorporation of titanium slag, and the utilization of quartz sand on the strength properties and shrinkage behavior of ultra-high-performance concrete (UHPC). By using low-heat silicate cement to prepare UHPC, this study conducted standard curing and steam curing, and comprehensively analyzed the macro and micro performance of UHPC under different curing conditions. The findings indicate that the application of steam curing markedly enhances the mechanical attributes of UHPC while efficiently decreasing its drying shrinkage. In the comparative tests, we found that the compressive strength of concrete that had undergone 2 days of steam curing was 9.15% higher than that of concrete cured for 28 days under standard conditions. In addition, under the same curing conditions, titanium slag sand had higher mechanical properties than quartz sand. Under standard curing conditions, the 28-day compressive strength of UHPC using titaniferous slag aggregate was 12.64% higher than that of UHPC using standard sand. Through the data analysis of XRD, TG, and MIP, we found that the content of Ca(OH)2 in the hydration products after steam curing was reduced compared to the standard curing conditions, and the pore structure had been optimized. The UHPC prepared with titanium slag sand has greater advantages in mechanical properties and drying shrinkage, and has a smaller pore structure than the UHPC prepared with quartz sand. Moreover, the use of titanium slag sand offers ecological and economic benefits, making it a more sustainable and cost-effective option for high-performance construction applications.

Influence of Ambient Relative Humidity on the Shrinkage Strain of Engineered Geopolymer Composites Based on Orthogonal Experimental Design.

With the aim to systematically analyze the ambient relative humidity on the shrinkage strain of Engineered Geopolymer Composites (EGCs), this paper studied four variables (fly ash to ground granulated blast furnace slag mass ratio, alkali content, water-binder ratio, and fiber volume content) though orthogonal experimental design and three different relative humidity values (30%, 60%, and 100% RH). The results indicated that, for EGC specimens under 30% RH and 60% RH, the decrease in slag content and increase in alkali content both resulted in greater drying shrinkage. The addition of fibers effectively reduced the shrinkage strain, while a minor impact on shrinkage was presented by the W/B ratio. The first and second key factors affecting the drying shrinkage strain were the FA/GGBS ratio and the alkali content. The optimal ratio of FA/GGBS, alkali content, and fiber volume fraction were 0/100, 4%, and 1.5%, respectively. Dring shrinkage strain was decreased with the increase in ambient relative humidity. Compared with the shrinkage strain under 30% RH, the reduction in shrinkage strain under 60% RH and 100%RH was up to 46.1% and 107.5%, respectively. At last, a relationship between shrinkage strain and curing age under 30% and 60% RH was established with a fitting degree from 0.9492 to 0.9987, while no clear relationship was presented under 100% RH. The results in this paper provide a practical method for solving the shrinkage problem of EGCs.

Shrinkage and Creep Properties of Low-Carbon Hybrid Cement.

Hybrid cements combine clinker with large amount of supplementary cementitious materials while utilizing hydration and alkali activation processes. This paper summarizes shrinkage and creep properties of industrially produced H-cement, containing only 20% of Portland clinker. In comparison with a reference cement CEM II/B-S 32.5 R, autogenous shrinkage is smaller after 7 days, and drying shrinkage is similar at similar times. A different capillary system of H-cement leads to faster water mass loss during drying. Basic and total creep of concrete remains in the standard deviation of B4 and EC2 creep models. The results demonstrate that shrinkage and creep properties of concrete made from H-cement have similar behavior as conventional structural concrete or high-volume fly ash concrete.

Investigating the Heterogeneity of "Study Twins".

Meta-analyses are commonly performed based on random-effects models, while in certain cases one might also argue in favor of a common-effect model. One such case may be given by the example of two "study twins" that are performed according to a common (or at least very similar) protocol. Here we investigate the particular case of meta-analysis of a pair of studies, for example, summarizing the results of two confirmatory clinical trials in phase III of a clinical development program. Thereby, we focus on the question of to what extent homogeneity or heterogeneity may be discernible and include an empirical investigation of published ("twin") pairs of studies. A pair of estimates from two studies only provide very little evidence of homogeneity or heterogeneity of effects, and ad hoc decision criteria may often be misleading.

Construction and validation of a prediction model for inguinal lymph node metastasis of penile malignancy.

Background: Penile squamous cell carcinoma is a relatively rare malignancy among male malignancies, there are more than 30,000 new cases and more than 10,000 deaths of penile cancer annually. In patients with penile malignancy, inguinal lymph node metastasis (ILNM) significantly reduces patient survival. Thus, we identified the risk factors for ILNM in penile malignancies, aiming to develop a precise prediction model. Methods: We retrospectively analyzed 112 male patients with penile cancer. All subjects underwent penile surgery and inguinal lymphadenectomy at the same time, and postoperative pathology confirmed ILNM. Fisher's exact test, t-test, and Wilcoxon rank sum test were used to assess differences in demographic information and clinical features between the two groups, followed by logical least absolute shrinkage and selection operator (LASSO) regression analysis to determine risk factors of ILNM. The prediction model was constructed using nomogram. Results: LASSO regression revealed that age [ß=-0.005, odds ratio (OR) =0.995], smoking history (ß=-0.006, OR =0.994) and interleukin 2 (IL-2) level (ß=-0.0112, OR =0.989) were protective against ILNM. However, lymph node diameter (ß=0.3117, OR =1.366), T-stage (ß=0.1254, OR =1.134), fibrinogen (ß=0.0377, OR =1.038), IL-4 level (ß=0.004, OR =1.001), and neutrophil-to-lymphocyte ratio (ß=0.0355, OR =1.034) were risk factors for developing ILNM. When assessing the risk of metastasis, it is crucial to balance these factors. The aforementioned characteristics were utilized to establish the predictive model, which demonstrated a good predictive ability with an area under the curve (AUC) value of 0.81. Moreover, internal leave-one-way cross-validation was used to construct a nomogram showing consistency, with an AUC of 0.75. Conclusions: The diagnosis of ILNM in penile malignant tumors can be predicted through clinicopathological features, biochemical tests, and prediction models based on tumor markers.

Tumor enhancement and shrinkage pattern in dynamic contrast-enhanced magnetic resonance imaging for predicting pathologic complete response after human epidermal growth factor receptor 2 (HER2)-targeted therapy in breast cancer.

Background: Targeted therapy with neoadjuvant chemotherapy for patients with human epidermal growth factor receptor 2 (HER2)-positive breast cancer has increased the rates of pathological complete response (pCR) and breast preservation surgery and improved the overall disease-free survival rate. This study aimed to determine whether tumor enhancement and shrinkage patterns in dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) can predict the efficacy of targeted therapy in patients with HER2-positive breast cancer and differentiate pCR from non-pCR. Methods: The data of 64 patients with HER2-positive breast cancer who received targeted therapy prior to surgery were retrospectively collected. All patients had complete postoperative pathological data. The pretreatment evaluation of the tumor enhancement pattern and the shrinkage pattern after two treatment cycles were assessed. The difference in the enhancement and shrinkage patterns between the pCR and non-pCR groups was evaluated via the χ2 test. Logistic regression analysis was used to assess the value of enhancement and shrinkage patterns for predicting pCR in patients with HER2-positive breast cancer. Results: There were statistically significant differences in tumor size, estrogen receptor (ER) status, lymph node metastasis, enhancement pattern, and shrinkage pattern between the pCR and non-pCR cases. Patients with a tumor size ≤20 mm were likely to achieve pCR. ER status, lymph node metastasis, and enhancement and shrinkage patterns each had good precision for predicting pCR, and the combination of enhancement and shrinkage patterns had the highest prediction accuracy. Multivariate logistic regression analysis indicated that only enhancement pattern had a significant predictive value. Conclusions: Among patients with HER2-positive breast cancer, those with tumor size ≤20 mm, ER-negative status, no lymph node metastases, and mass enhancement and concentric shrinkage patterns are more likely to achieve pCR. Mass enhancement combined with concentric shrinkage had the highest accuracy in predicting pCR, indicating that preoperative imaging may be useful for guiding clinical decisions regarding targeted treatments.

A nomogram and heat map based on LASSO-Cox regression for predicting the risk of early-stage severe fever with thrombocytopenia syndrome patients developing into critical illness at 7-day and 14-day.

Severe fever with thrombocytopenia syndrome (SFTS) represents an emerging infectious disease characterized by a substantial mortality risk. Early identification of patients is crucial for effective risk assessment and timely interventions. In the present study, least absolute shrinkage and selection operator (LASSO)-Cox regression analysis was conducted to identify key risk factors associated with progression to critical illness at 7-day and 14-day. A nomogram was constructed and subsequently assessed for its predictive accuracy through evaluation and validation processes. The risk stratification of patients was performed using X-tile software. The performance of this risk stratification system was assessed using the Kaplan-Meier method. Additionally, a heat map was generated to visualize the results of these analyses. A total of 262 SFTS patients were included in this study, and four predictive factors were included in the nomogram, namely viral copies, aspartate aminotransferase (AST) level, C-reactive protein (CRP), and neurological symptoms. The AUCs for 7-day and 14-day were 0.802 [95% confidence interval (CI): 0.707-0.897] and 0.859 (95% CI: 0.794-0.925), respectively. The nomogram demonstrated good discrimination among low, moderate, and high-risk groups. The heat map effectively illustrated the relationships between risk groups and predictive factors, providing valuable insights with high predictive and practical significance.

Post-Estimation Shrinkage in Full and Selected Linear Regression Models in Low-Dimensional Data Revisited.

The fit of a regression model to new data is often worse due to overfitting. Analysts use variable selection techniques to develop parsimonious regression models, which may introduce bias into regression estimates. Shrinkage methods have been proposed to mitigate overfitting and reduce bias in estimates. Post-estimation shrinkage is an alternative to penalized methods. This study evaluates effectiveness of post-estimation shrinkage in improving prediction performance of full and selected models. Through a simulation study, results were compared with ordinary least squares (OLS) and ridge in full models, and best subset selection (BSS) and lasso in selected models. We focused on prediction errors and the number of selected variables. Additionally, we proposed a modified version of the parameter-wise shrinkage (PWS) approach named non-negative PWS (NPWS) to address weaknesses of PWS. Results showed that no method was superior in all scenarios. In full models, NPWS outperformed global shrinkage, whereas PWS was inferior to OLS. In low correlation with moderate-to-high signal-to-noise ratio (SNR), NPWS outperformed ridge, but ridge performed best in small sample sizes, high correlation, and low SNR. In selected models, all post-estimation shrinkage performed similarly, with global shrinkage slightly inferior. Lasso outperformed BSS and post-estimation shrinkage in small sample sizes, low SNR, and high correlation but was inferior when the opposite was true. Our study suggests that, with sufficient information, NPWS is more effective than global shrinkage in improving prediction accuracy of models. However, in high correlation, small sample sizes, and low SNR, penalized methods generally outperform post-estimation shrinkage methods.

Comparison of Machine Learning Algorithms and Nomogram Construction for Diabetic Retinopathy Prediction in Type 2 Diabetes Mellitus Patients.

INTRODUCTION: The aim of this study was to compare various machine learning algorithms for constructing a diabetic retinopathy (DR) prediction model among type 2 diabetes mellitus (DM) patients and to develop a nomogram based on the best model. METHODS: This cross-sectional study included DM patients receiving routine DR screening. Patients were randomly divided into training (244) and validation (105) sets. Least absolute shrinkage and selection operator regression was used for the selection of clinical characteristics. Six machine learning algorithms were compared: decision tree (DT), k-nearest neighbours (KNN), logistic regression model (LM), random forest (RF), support vector machine (SVM), and XGBoost (XGB). Model performance was assessed via receiver-operating characteristic (ROC), calibration, and decision curve analyses (DCAs). A nomogram was then developed on the basis of the best model. RESULTS: Compared with the five other machine learning algorithms (DT, KNN, RF, SVM, and XGB), the LM demonstrated the highest area under the ROC curve (AUC, 0.894) and recall (0.92) in the validation set. Additionally, the calibration curves and DCA results were relatively favourable. Disease duration, DPN, insulin dosage, urinary protein, and ALB were included in the LM. The nomogram exhibited robust discrimination (AUC: 0.856 in the training set and 0.868 in the validation set), calibration, and clinical applicability across the two datasets after 1,000 bootstraps. CONCLUSION: Among the six different machine learning algorithms, the LM algorithm demonstrated the best performance. A logistic regression-based nomogram for predicting DR in type 2 DM patients was established. This nomogram may serve as a valuable tool for DR detection, facilitating timely treatment.