Clinical Prediction Model Research Articles

Developing a risk score using liquid biopsy biomarkers for selecting Immunotherapy responders and stratifying disease progression risk in metastatic melanoma patients

BackgroundDespite the high response rate to PD-1 blockade therapy in metastatic melanoma (MM) patients, a significant proportion of patients do not respond. Identifying biomarkers to predict patient response is crucial, ideally through non-invasive methods such as liquid biopsy.MethodsSoluble forms of PD1, PD-L1, LAG-3, CTLA-4, CD4, CD73, and CD74 were quantified using ELISA assay in plasma of a cohort of 110 MM patients, at baseline, to investigate possible correlations with clinical outcomes. A clinical risk prediction model was applied and validated in pilot studies.ResultsNo biomarker showed statistically significant differences between responders and non-responders. However, high number of significant correlations were observed among certain biomarkers in non-responders. Through univariate and multivariate Cox analyses, we identified sPD-L1, sCTLA-4, sCD73, and sCD74 as independent biomarkers predicting progression-free survival and overall survival. According to ROC analysis we discovered that, except for sCD73, values of sPD-L1, sCTLA-4, and sCD74 lower than the cut-off predicted lower disease progression and reduced mortality. A comprehensive risk score for predicting progression-free survival was developed by incorporating the values ​​of the two identified independent factors, sCTLA-4 and sCD74, which significantly improved the accuracy of outcome prediction. Pilot validations highlighted the potential use of the risk score in treatment-naive individuals and long responders.ConclusionIn summary, risk score based on circulating sCTLA-4 and sCD74 reflects the response to immune checkpoint inhibitor (ICI) therapy in MM patients. If confirmed, through further validation, these findings could assist in recommending therapy to patients likely to experience a long-lasting response.

Mortality Risk Prediction in Patients With Antimelanoma Differentiation-Associated, Gene 5 Antibody-Positive, Dermatomyositis-Associated Interstitial Lung Disease: Algorithm Development and Validation.

Patients with antimelanoma differentiation-associated gene 5 antibody-positive dermatomyositis-associated interstitial lung disease (anti-MDA5+DM-ILD) are susceptible to rapidly progressive interstitial lung disease (RP-ILD) and have a high risk of mortality. There is an urgent need for a reliable prediction model, accessible via an easy-to-use web-based tool, to evaluate the risk of death. This study aimed to develop and validate a risk prediction model of 3-month mortality using machine learning (ML) in a large multicenter cohort of patients with anti-MDA5+DM-ILD in China. In total, 609 consecutive patients with anti-MDA5+DM-ILD were retrospectively enrolled from 6 hospitals across China. Patient demographics and laboratory and clinical parameters were collected on admission. The primary endpoint was 3-month mortality due to all causes. Six ML algorithms (Extreme Gradient Boosting [XGBoost], logistic regression (LR), Light Gradient Boosting Machine [LightGBM], random forest [RF], support vector machine [SVM], and k-nearest neighbor [KNN]) were applied to construct and evaluate the model. After applying inclusion and exclusion criteria, 509 (83.6%) of the 609 patients were included in our study, divided into a training cohort (n=203, 39.9%), an internal validation cohort (n=51, 10%), and 2 external validation cohorts (n=92, 18.1%, and n=163, 32%). ML identified 8 important variables as critical for model construction: RP-ILD, erythrocyte sedimentation rate (ESR), serum albumin (ALB) level, age, C-reactive protein (CRP) level, aspartate aminotransferase (AST) level, lactate dehydrogenase (LDH) level, and the neutrophil-to-lymphocyte ratio (NLR). LR was chosen as the best algorithm for model construction, and the model demonstrated excellent performance, with an area under the receiver operating characteristic (ROC) curve (AUC) of 0.866, a sensitivity of 84.8%, and a specificity of 84.4% on the validation data set and an AUC of 0.90, a sensitivity of 85.0%, and a specificity of 83.9% on the training data set. Calibration curves and decision curve analysis (DCA) confirmed the model's accuracy and clinical applicability. Moreover, the model showed strong predictive performance in the external validation cohorts (cohort 1: AUC=0.836, 95% CI 0.754-0.916; cohort 2: AUC=0.915, 95% CI 0.871-0.959), indicating good generalizability. This model was integrated into a web-based tool to predict the 3-month mortality for patients with anti-MDA5+DM-ILD. We successfully developed a robust clinical prediction model and an accompanying web tool to estimate the 3-month mortality risk for patients with anti-MDA5+DM-ILD.

Utilizing MV-FLOW™ and multidimensional ultrasound characteristics for prognosticating FET outcomes in RIF patients: Study Protocol for a cross-sectional study.

Recurrent implantation failure (RIF) is a common issue in frozen-thawed embryo transfer (FET). Prior to transfer, uterine endometrial receptivity of FET patients can be assessed using multimodal transvaginal ultrasound indicators to predict the success rate of the current FET cycle. Endometrial blood flow is a crucial element in evaluating endometrial receptivity. MV-FLOW™ is an advanced two-dimensional superb microvascular imaging technology that can detect and display blood flow in micro-vessels. The data for this study were obtained from an ongoing cross-sectional study comprising 323 RIF patients and 323 first implantation (FI) patients, who underwent transvaginal ultrasound before FET. We collected basic clinical data and multimodal ultrasound data from these patients as predictive features, with clinical pregnancy as the predictive label, for model training. Based on the above, this study aims to establish and validate a clinical prediction model for FET outcomes using support vector classification (SVC) algorithms, based on MV-FLOW™ and multidimensional transvaginal ultrasound imaging features. The objective is to determine the predictive role of multimodal transvaginal ultrasound in embryo transfer outcomes and provide evidence for the clinical application of MV-FLOW™. Trial registration: Trial Registration: ChiCTR2400086401.

Advancing treatment response prediction in first-episode psychosis: integrating clinical and electroencephalography features.

Prompt diagnosis and intervention are crucial for first-episode psychosis (FEP) outcomes, but predicting the response to antipsychotics remains challenging. We studied whether adding electroencephalography (EEG) characteristics improves clinical prediction models for treatment response and whether EEG-based predictors are influenced by initial treatment. We included 115 antipsychotic-naïve patients with FEP. Positive and Negative Syndrome Scale (PANSS) and sociodemographic items were included as clinical features. Additionally, we analyzed resting-state EEG data (n = 45) for (relative) power, functional connectivity, and network organization. Treatment response, measured as change in PANSS positive subscale scores (∆PANSS+), was predicted using a random forest regression model. We analyzed whether the most predictive EEG characteristics were influenced after treatment. The clinical model explained 12% variance in symptom reduction in the training set and 32% in the validation set. Including EEG variables in the model led to a nonsignificant increase of 2% (total 34%) explained variance in symptom reduction. High hallucination symptom scores and a more hierarchical organization of alpha band networks (tree hierarchy) were associated with ∆PANSS+ reduction. The tree hierarchy in the alpha band decreased after medication. EEG source analysis revealed that this change was driven by alterations in the degree and centrality of frontal and parietal nodes in the functional brain network. Both clinical and EEG characteristics can inform treatment response prediction in patients with FEP, but the combined model may not be beneficial over a clinical model. Nevertheless, adding a more objective marker such as EEG could be valuable in selected cases.

Elevated serum levels of GPX4, NDUFS4, PRDX5, and TXNRD2 as predictive biomarkers for castration resistance in prostate cancer patients: an exploratory study.

Prostate cancer (PCa) is a heterogeneous disease affecting over 14% of the male population worldwide. Although patients often respond positively to initial treatments within the first 2-3 years, many eventually develop a more lethal form of the disease known as castration-resistant PCa (CRPC). At present, no biomarkers that predict the onset of CRPC are available. This study aims to provide insights into the diagnosis and prediction of CRPC emergence. Protein expression dynamics were analysed in drug (androgen receptor inhibitor)-tolerant persister (DTP) and drug withdrawal cells using proteomics to identify potential biomarkers. These biomarkers were subsequently validated using a mouse model, 180-paired carcinoma/benign tissues, and 482 serum samples. Five machine learning algorithms were employed to build clinical prediction models, wherein the SHapley Additive exPlanation (SHAP) framework was used to interpret the best-performing model. Moreover, three regression models were developed to determine the Time from initial PCa diagnosis to CRPC development (TPC) in patients. We identified that the protein expression levels of GPX4, NDUFS4, PRDX5, and TXNRD2 were significantly upregulated in PCa patients, particularly in those with CRPC. Among the tested machine learning models, the random forest and extreme gradient boosting models performed best on tissue and serum cohorts, achieving AUCs of 0.958 and 0.988, respectively. In addition, a significant inverse correlation was observed between TPC and serum levels of these four biomarkers. This correlation was formulated in three regression models, which achieved the smallest mean absolute error of 1.903 on independent datasets for predicting CRPC emergence. Our study provides new insights into the role of DTP cells in CRPC development. The quad protein panel identified in our study, along with the post hoc and intrinsically explainable prediction models, may serve as a convenient and real-time prognostic tool, addressing the current lack of clinical biomarkers for CRPC.

Personalized frailty risk assessment in long-term survivors of colorectal cancer.

65 Background: Frailty, a pathologic form of aging, is associated with reduced quality of life and increased risk of death. Its incidence increases with age. Frailty is a concern for cancer survivors, especially since this population is surviving longer and increasing in size. There is a need to predict which patients are at risk of frailty to tailor preventative measures, particularly in early-stage colon and rectal cancer survivors, the largest group in gastrointestinal cancer survivors. Methods: This was a retrospective cohort study of individuals aged 66 and older in the SEER-Medicare linked database, diagnosed with stage I-III colon or rectal cancer between 2003-2012. Patients included in the study received definitive surgical treatment and survived for at least 5 years after diagnosis. Frailty was assessed using administrative claims codes with the Kim frailty index. Patients already frail at year 5 were excluded from the analysis. An increase in frailty score, indicating the onset of frailty or worsening to moderate or severe frailty, occurring within 5-10 years following cancer diagnosis was the primary outcome. Predictors of frailty were identified using restricted mean survival time (RMST) regression, with results less than 1 indicating a shorter time to frailty, and significant factors were used to develop a clinical prediction model and stratify patients into risk tertiles. Results: At 10 years or end of available follow-up, 58% of the patients had developed new onset or worsening frailty. There were no significant differences in RMST by patient race, sex, cancer stage and grade. Receipt of systemic therapy 4-5 years after diagnosis, having an ostomy present in years 4-5, advancing age, comorbidities, and living in an area with a greater proportion of residents below the federal poverty line were all associated with shorter time to frailty; the largest effects were seen with advancing age and comorbidities (Table). Our clinical prediction model incorporated age>85 and having multiple comorbidities for both cohorts and ostomy in years 4-5 for the colon cohort. Conclusions: Our population level analysis provided a clinical prediction model for frailty 5-10 years after cancer diagnosis in colon and rectal cancer survivors and may help inform long-term survivorship management. RMST for selected variables. Selected Variables RMST for Colon Cohort (95% CI) RMST for Rectal Cohort (95% CI) Age 85-89 (ref: age 71-74) 0.76 (0.73, 0.79) 0.79 (0.74, 0.84) Age 90+ (ref: age 71-74) 0.67 (0.64, 0.70) 0.74 (0.66, 0.82) 20-100% area-level poverty (ref: 0-<5%) 0.95 (0.92, 0.98) 0.96 (0.91, 1.02) Receiving systemic therapy in years 4-5 0.89 (0.85, 0.92) Not selected Ostomy in years 4-5 0.80 (0.69, 0.92) 0.97 (0.92, 1.02) Elixhauser Comorbidity Index (ECI) 1-2 (reference: ECI 0) 0.77 (0.75, 0.79) 0.79 (0.76, 0.82) ECI 3 (ref: ECI 0) 0.60 (0.57, 0.64) 0.67 (0.59, 0.76)

Development and validation of clinical prediction model for functional independence measure following stroke rehabilitation.

To develop and internally validate a clinical prediction model that includes balance ability and nutritional indices for the motor-functional independence measure (M-FIM) at 90 days post-stroke stroke. This retrospective, single-center study included 566 patients with stroke undergoing rehabilitation at our rehabilitation hospital. The primary outcome was the M-FIM score of >61 at 3 months post-strokes onset. Stepwise conditional forward selection was first used to identify predictors for the achievement of M-FIM>61 at 90 days post-stroke, from 25 potential predictors at admission. The selected predictors were dichotomized with cut-off values to establish scoring systems, resulting in the B-ADL model, which includes postural balance (B), albumin level, age, arm function (A), days since stroke onset (D), and level of activities of daily living (ADL) (L). For internal validation, we corrected the optimism of the area under the curve of receiver operating characteristic curve (AUROC) induced by overfitting the original data using the bootstrap validation method. Calibration capacity was assessed using a calibration plot. We developed a clinical model to predict the M-FIM at 90 days post-stroke onset. The AUROC of the B-ADL model was 0.92 (sensitivity, 93.7%; specificity, 89.7%). The B-ADL model showed high accuracy with an AUROC of 0.970 in the internal validation. The scoring system in the validation cohort had a cut-off value of 5.5/12 points to predict the achievement of M-FIM>61 (AUROC: 0.950; 95% CI 0.930-0.970). The B-ADL model accurately predicted M-FIM >61 at 90 days post-stroke on the day of admission to the recovery rehabilitation ward. The B-ADL model is useful for optimizing rehabilitation programs and resource allocation, allowing for targeted interventions after stroke.

Harnessing temporal patterns in administrative patient data to predict risk of emergency hospital admission.

Unplanned hospital admissions are associated with worse patient outcomes and cause strain on health systems worldwide. Primary care electronic health records (EHRs) have successfully been used to create prediction models for emergency hospitalisation, but these approaches require a broad range of diagnostic, physiological, and laboratory values. In this study, we aimed to capture temporal patterns of patient activity from EHR data and evaluate their effectiveness in predicting emergency hospital admissions compared with conventional methods. In this retrospective observational study, we used the Secure Anonymised Information Linkage databank to extract temporal patterns of primary care activity from undifferentiated electronic health record timestamp data for 1·37 million patients in Wales aged 18-80 years with at least one recorded Read code between the years 2016 and 2018. Using Gaussian mixture modelling we grouped patients into distinct temporal clusters, performed a three-stage validation of our approach and calculated the risk of emergency hospital admission for each temporal cluster group. Finally, these temporal clusters were combined with five administrative variables and incorporated into four emergency hospital admission prediction models (logistic regression, naive Bayes, XGBoost, and multilayer perceptron [MLP]) and compared with a more traditional, but data-intensive, modelling technique. The primary outcome was emergency hospital admission as the next health-care event. Six distinct temporal cluster patterns of primary care EHR activity were identified, associated with varying risks of future emergency hospital admission risk. These patterns were visually interpretable, repeatable at a population-level, and clinically plausible. The best emergency hospital admission prediction model (MLP) achieved an area under the receiver operating characteristic (AUROC) of 0·82 and precision of 0·94 in regional cohorts. In external validation in regional cohorts, similar model performance was observed (AUROC 0·82 and precision 0·92). This model also matched the performance of a more complex model (extended feature model) requiring 33 clinical parameters (AUROC 0·82 vs 0·83; precision 0·94 vs 0·90) for the same task on the same dataset. We developed a novel machine learning pipeline that extracts interpretable temporal patterns from simple representations of EHR data and can be incorporated into emergency hospital admission predictors. This framework might enable more rapid development of parsimonious clinical prediction models. UKRI CDT in AI for Healthcare, UKRI Turing AI Fellowship, NIHR Imperial Biomedical Research Centre, and Research Capability Funding.

Prediction model establishment of prognosis factors for distant metastasis of hepatocellular carcinoma based on the SEER database.

Distant metastasis in hepatocellular carcinoma (HCC) is an important indicator of poor patient prognosis. Identifying patients who are at high risk of metastasis early on is essential for creating personalized treatment plans, yet currently, there is a scarcity of effective predictive tools. To investigate the effects of different factors on distant metastasis in HCC patients and to establish a clinical prediction model for predicting distant metastasis in HCC patients. Our study retrospectively examined 22,318 patients diagnosed with confirmed HCC from the SEER database. Prognostic factors for developing distant metastases in HCC patients were identified by univariate and multivariate logistic regression analyses. Utilizing data from a multivariate logistic regression analysis, we created a nomogram. Its predictive precision was evaluated by analyzing the calibration curve, the area under the curve (AUC) of the receiver operating characteristic curve, decision curve assessment (DCA), and Kaplan-Meier (KM) curve analysis of overall survival. Finally,the nomogram was visualized with an online calculator. We identified six independent prognostic factors: ethnicity, marital status, tumor size, survival time, surgery, and radiotherapy. The nomogram constructed from these six factors showed good calibration, discrimination, and clinical application value after calibration curve analysis, receiver operating characteristic curve analysis and DCA curve analysis. Besides, KaplanMeier survival curves also demonstrated that this nomogram had predictive accuracy. In this research, a nomogram model was created to accurately predict distant metastasis risk in patients with HCC. This study provides guidance for optimizing individual therapies and making better clinical decisions.

Development of an Ultrasound-Based Radiomics Nomogram for Preoperative Prediction of HER-2 Status in Invasive Breast Cancer.

This study aimed to create a radiomics nomogram using grayscale ultrasound (US) to predict human epidermal growth factor receptor 2 (HER-2) expression status preoperatively in invasive breast cancer (IBC) patients. The study population was randomly divided into a training dataset (360 patients, 99 HER-2-positive) and a validation dataset (155 patients, 42 HER-2-positive). Clinical data, including US features, were collected. Radiomics features were extracted from grayscale US images, followed by feature selection to establish a radiomics score (Radscore) model. Univariate and multivariate logistic regression analyses identified independent risk factors for the clinical and radiomics nomogram models. Model performance was evaluated using receiver operating characteristic curves, calibration curves, decision curve analysis, net reclassification improvement, and integrated discrimination improvement. 16 radiomics features were selected for the Radscore model. Tumor margin and calcification emerged as significant preoperative risk factors for HER-2 status, forming the basis of a clinical prediction model. The integrated radiomics nomogram, combining tumor margin, calcification, and Radscore, demonstrated strong discrimination with area under the curvevalues of 0.810 in the training dataset and 0.807 in the validation dataset. The US-based radiomics nomogram shows substantial promise for preoperatively predicting HER-2 status in IBC patients.

Application and progress of nomograms in gastric cancer

Gastric cancer, as one of the malignant tumors with a significant disease burden globally, emphasizes the importance of early diagnosis and personalized treatment for improving patient prognosis. In recent years, clinical prediction models (CPMs) have played a crucial role in predicting disease risks, assisting medical decision-making, and evaluating clinical prognosis and benefits as tools for risk–benefit assessment. Nomograms, as an important visualization form of clinical prediction models, have been increasingly applied in tumor-related research. Numerous studies have constructed multiple nomogram models by integrating clinical, pathological, laboratory, imaging data, and genetic characteristics, providing an accurate and effective tool for predicting the risk of gastric cancer, early diagnosis, treatment response assessment, and prognosis analysis. This article aims to review the current clinical applications and research progress of nomograms in gastric cancer, with the goal of providing robust references and theoretical support for clinical practice.

Left atrial assessment improves prediction of incident atrial fibrillation beyond CHARGE-AF risk score

Abstract Background Atrial fibrillation (AF) is associated with substantial mortality and morbidity, even when it is asymptomatic and/or remains undetected. Clinical predictive models such as Charge-AF are not robust enough to be used in the clinical practice. We investigated the benefits of adding echocardiographic risk variables to Charge-AF risk model for its potential use in targeted screening for AF. Methods This study included 1108 consecutive patients aged 60±14 years old without a previous diagnosis of AF referred for echocardiographic assessment. Left atrial and ventricular functions were measured using strain analysis. Five-year predicted risk for AF was estimated, using Charge-AF risk model. The incidence of AF was ascertained from resting and exercise stress electrocardiograms and Holter recordings. Sequential Cox models were used to identify echocardiographic variables for predicting the incidence of new onset AF. Results The overall incident rate of AF was 3.1%, 5.5% and 7.2% at 1 year, 3 year and 5 year, respectively. Left atrial volume index (LAVi) predicted the incidence with a hazard ratio of 1.06 (p<0.001), that was independent of and increment to Charge-AF risk estimates (figure A). The addition of E/e’ (hazard ratio=1.064, p=0.001), left atrial reservoir strain (hazard ratio=0.98, p=0.034), left atrial booster strain (hazard ratio=0.97, p=0.048) to the baseline model including Charge-AF risk estimates also showed a significant, but less incremental benefit for AF prediction, compared with LAVi. When all these echocardiographic variables were added to a stepwise nested Cox model, only LAVi was significantly and independently predictive of AF incidence beyond Charge-AF risk estimates. This finding was confirmed by Kaplan-Meier curves showing that an increase in LAVi was progressively associated with increased risk for AF (figure B). When the cumulative AF incidence was analyzed separately between patients with higher (predicted risk >= 1.5%/year) and lower (predicted risk < 1.5%/year) Charge-AF risk estimates, LAVi was able to sub-stratify only patients with higher Charge-AF risk estimate (figures C and D). Conclusion LAVi provided incremental value in predicting incident AF in hospital patients. The inclusion of LAVi to the diagnostic algorithm may help guide screening and further monitoring for AF risk in this population.

P-321. Risk Factors for Methicillin-Resistant Staphylococcus aureus Carriers in the Intensive Care Unit: A Single-Center, Retrospective Cohort Study in Japan

Abstract Background Methicillin-resistant Staphylococcus aureus (MRSA) is one of the most common pathogens in the intensive care unit (ICU). Active surveillance cultures (ASCs) for MRSA often performed in ICUs may not be optimal in ICU settings with low prevalence of MRSA. This study aims to determine the risk factors of MRSA carriage in the ICU and develop a clinical predictive model to optimize the screening process. Methods All patients screened for MRSA by nasal ASC at the time of ICU admission between April 2015 and August 2022 were retrospectively included and divided into MRSA-positive and MRSA-negative groups. Patients’ characteristics were evaluated to determine the prevalence of MRSA and the risk factors. Cost analysis was conducted based on the risk factors identified by our analysis. Results Of the 3,927 ICU patients included, 133 (3.4%) were MRSA-positive. Multivariate analyses showed that risk factors for MRSA carriage were age ≥60 years (odds ratio [OR]: 1.60), history of hospitalization within a year (OR: 1.51), admission under departments dealing with skin and soft tissues (OR: 3.00), and ICD-10 codes classification I (OR: 3.44). Screening patients based on at least one of the risk factors exhibited high sensitivity (93.2%) to identifying MRSA carriage and could reduce ASC overall costs by 86.9%. Conclusion This study suggests that universal ASCs to detect MRSA may not be optimal in ICU settings with low prevalence of MRSA. Targeted screening based on risk factors may reduce the volume and cost of MRSA screening. Prospective studies are warranted to confirm these findings. Disclosures All Authors: No reported disclosures

Corrigendum: A clinical practical model for preoperative prediction of visual outcome for pituitary adenoma patients in a retrospective and prospective study

Corrigendum: A clinical practical model for preoperative prediction of visual outcome for pituitary adenoma patients in a retrospective and prospective study

Refining MR-guided thermal ablation for HCC within the Milan criteria: a decade of clinical outcomes and predictive modeling at a single institution

BackgroundThe appropriateness of ablation for liver cancer patients meeting the Milan criteria remains controversial.PurposeThis study aims to evaluate the long-term outcomes of MR-guided thermal ablation for HCC patients meeting the Milan criteria and develop a nomogram for predicting survival rates.MethodsA retrospective analysis was conducted from January 2009 to December 2021 at a single institution. Patients underwent MR-guided thermal ablation. Factors influencing progression-free survival (PFS) and overall survival (OS) were identified using univariate and multivariate Cox regression and stepwise regression. A nomogram was developed for survival prediction, followed by risk stratification and internal validation. Adverse events (AEs) were also analyzed.ResultsA total of 181 patients were included, with a mean follow-up of 73.8 ± 31.7 months. The cumulative local tumor progression rates at 1, 3, and 5 years were 0.80%, 1.27%, and 1.86%, respectively. The 1-, 3-, and 5-year PFS rates were 81.8%, 57.4%, and 38.1%, and OS rates were 98.3%, 87.8%, and 62.9%. Poorer outcomes were associated with age ≤ 60 years, tumor size > 2 cm, multiple tumors, cirrhosis, proximity to major vessels, and narrow ablation margins (P < 0.05). The nomogram accurately predicted 3- and 5-year survival, and internal validation confirmed the results. AEs occurred in 33.7% of patients, with pain being the most common.ConclusionMR-guided ablation is effective for HCC patients within the Milan criteria, especially for those with smaller tumors and better liver function. The nomogram and risk stratification model are valuable tools for predicting patient outcomes and guiding treatment.

Exploring the role of multimodal [18F]F-PSMA-1007 PET/CT and multiparametric MRI data in predicting ISUP grading of primary prostate cancer.

The study explores the role of multimodal imaging techniques, such as [18F]F-PSMA-1007 PET/CT and multiparametric MRI (mpMRI), in predicting the ISUP (International Society of Urological Pathology) grading of prostate cancer. The goal is to enhance diagnostic accuracy and improve clinical decision-making by integrating these advanced imaging modalities with clinical variables. In particular, the study investigates the application of few-shot learning to address the challenge of limited data in prostate cancer imaging, which is often a common issue in medical research. This study conducted a retrospective analysis of 341 prostate cancer patients enrolled between 2019 and 2023, with data collected from five imaging modalities: [18F]F-PSMA-1007 PET, CT, Diffusion Weighted Imaging (DWI), T2 Weighted Imaging (T2WI), and Apparent Diffusion Coefficient (ADC). The study compared the performance of five single-modality data sets, PET/CT dual-modality fusion data, mpMRI tri-modality fusion data, and five-modality fusion data within deep learning networks, analyzing how different modalities impact the accuracy of ISUP grading prediction. To address the issue of limited data, a few-shot deep learning network was employed, enabling training and cross-validation with only a small set of labeled samples. Additionally, the results were compared with those from preoperative biopsies and clinical prediction models to further assess the reliability of the experimental findings. The experimental results demonstrate that the multimodal model (combining [18F]F-PSMA-1007 PET/CT and multiparametric MRI) significantly outperforms other models in predicting ISUP grading of prostate cancer. Meanwhile, both the PET/CT dual-modality and mpMRI tri-modality models outperform the single-modality model, with comparable performance between the two multimodal models. Furthermore, the experimental data confirm that the few-shot learning network introduced in this study provides reliable predictions, even with limited data. This study highlights the potential of applying multimodal imaging techniques (such as PET/CT and mpMRI) in predicting ISUP grading of prostate cancer. The findings suggest that this integrated approach can enhance the accuracy of prostate cancer diagnosis and contribute to more personalized treatment planning. Furthermore, incorporating few-shot learning into the model development process allows for more robust predictions despite limited data, making this approach highly valuable in clinical settings with sparse data.

Development and validation of a novel risk-predicted model for early sepsis-associated acute kidney injury in critically ill patients: a retrospective cohort study

ObjectivesThis study aimed to develop a prediction model for the detection of early sepsis-associated acute kidney injury (SA-AKI), which is defined as AKI diagnosed within 48 hours of a sepsis...

Developing a clinical prediction model to modify empirical antibiotics for non-typhoidal Salmonella bloodstream infection in children under-five in the Democratic Republic of Congo

BackgroundNon-typhoidal Salmonella (NTS) frequently cause bloodstream infection in children under-five in sub-Saharan Africa, particularly in malaria-endemic areas. Due to increasing drug resistance, NTS are often not covered by standard-of-care empirical antibiotics for severe febrile illness. We developed a clinical prediction model to orient the choice of empirical antibiotics (standard-of-care versus alternative antibiotics) for children admitted to hospital in settings with high proportions of drug-resistant NTS.MethodsData were collected during a prospective cohort study in children (> 28 days—< 5 years) admitted with severe febrile illness to Kisantu district hospital, DR Congo. The outcome variable was blood culture confirmed NTS bloodstream infection; the comparison group were children without NTS bloodstream infection. Predictors were selected a priori based on systematic literature review. The prediction model was developed with multivariable logistic regression; a simplified scoring system was derived. Internal validation to estimate optimism-corrected performance was performed using bootstrapping and net benefits were calculated to evaluate clinical usefulness.ResultsNTS bloodstream infection was diagnosed in 12.7% (295/2327) of enrolled children. The area under the curve was 0.79 (95%CI: 0.76–0.82) for the prediction model, and 0.78 (0.85–0.80) for the scoring system. The estimated calibration slopes were 0.95 (model) and 0.91 (scoring system). At a decision threshold of 20% NTS risk, the prediction model and scoring system had 57% and 53% sensitivity, and 85% specificity. The net benefit for decisions thresholds < 30% ranged from 2.4 to 3.9 per 100 children.ConclusionThe model predicts NTS bloodstream infection and can support the choice of empiric antibiotics to include coverage of drug-resistant NTS, in particular for decision thresholds < 30%. External validation studies are needed to investigate generalizability.Trial registrationDeNTS study, clinicaltrials.gov: NCT04473768 (registration 16/07/2020) and TreNTS study, clinicaltrials.gov: NCT04850677 (registration 20/04/2021).

Machine Learning Model for Risk Stratification of Papillary Thyroid Carcinoma Based on Radiopathomics.

This study aims to develop a radiopathomics model based on preoperative ultrasoundand fine-needle aspiration cytology (FNAC) images to enable accurate, non-invasive preoperative risk stratification for patients with papillary thyroid carcinoma (PTC). The model seeks to enhance clinical decision-making by optimizing preoperative treatment strategies. A retrospective analysis was conducted on data from PTC patients who underwent thyroidectomy between October 2022 and May 2024 across six centers. Based on lymph node dissection outcomes, patients were categorized into high-risk and low-risk groups. Initially, a clinical predictive model was established based on the maximum diameter of the thyroid nodules. Radiomics features were extracted from preoperative two-dimensional ultrasound images, and pathomics features were extracted from 400x magnification H&E-stained tumor cell images from FNAC. The most predictive radiomics and pathomics features were identified through univariate analysis, Pearson correlation analysis and LASSO algorithm. The most valuable radiopathomics features were then selected by combining these predictive features. Finally, machine learningwith the XGBoost algorithm was employed to construct radiomics, pathomics, and radiopathomics models. The performance of the models was evaluated using the area under the curve (AUC), decision curve analysis, accuracy, specificity, sensitivity, positive predictive value, and negative predictive value. A total of 688 PTC patients were included, with 344 classified as intermediate/high-risk and 344 as low-risk. The multimodal radiopathomics model demonstrated excellent predictive performance, with AUCs of 0.886 (95% CI: 0.829-0.924) and 0.828 (95% CI: 0.751-0.879) in two external validation cohorts, significantly outperforming the clinical model (AUCs of 0.662 and 0.601), radiomics model (AUCs of 0.702 and 0.697), and pathomics model (AUCs of 0.741 and 0.712). The radiopathomics model exhibits significant advantages in accurately predicting preoperative risk stratification in PTC patients. Its application is expected to reduce unnecessary lymph node dissection surgeries, optimize treatment strategies, and improve therapeutic outcomes.

The Harms of Class Imbalance Corrections for Machine Learning Based Prediction Models: A Simulation Study.

Risk prediction models are increasingly used in healthcare to aid in clinical decision-making. In most clinical contexts, model calibration (i.e., assessing the reliability of risk estimates) is critical. Data available for model development are often not perfectly balanced with the modeled outcome (i.e., individuals with vs. without the event of interest are not equally prevalentin the data). It is common for researchers to correct for class imbalance, yet, the effect of such imbalance corrections on the calibration of machine learning models is largely unknown. We studied the effect of imbalance corrections on model calibration for a variety of machine learning algorithms. Using extensive Monte Carlo simulations we compared the out-of-sample predictive performance of models developed with an imbalance correction to those developed without a correction for class imbalance across different data-generating scenarios (varying sample size, the number of predictors, and event fraction). Our findings were illustrated in a case study using MIMIC-III data. In all simulation scenarios, prediction models developed without a correction for class imbalance consistently had equal or better calibration performance than prediction models developed with a correction for class imbalance. The miscalibration introduced by correcting for class imbalance was characterized by an over-estimation of risk and was not always able to be corrected with re-calibration. Correcting for class imbalance is not always necessary and may even be harmful to clinical prediction models which aim to produce reliable risk estimates on an individual basis.