Prediction Score Research Articles

The Lyon Score: A Novel Reflux Scoring System Based on the Lyon Consensus 2.0 That Associates With Treatment Outcome From Antireflux Therapy.

We explored if a score derived from parameters from esophageal testing could increase confidence in diagnosing conclusive gastroesophageal reflux disease and in predicting outcome. A prediction score was developed using metrics based on Lyon Consensus 2.0 thresholds extracted from endoscopy and pH-impedance monitoring. The Lyon score was the sum of weighted scores derived from a logistic regression model. The outcome was response to antireflux therapy, defined as 50% reduction in global symptoms on validated questionnaires. An existing database of endoscopy-negative patients with typical reflux symptoms undergoing esophageal testing from 2 centers (Europe and the United States) constituted the developmental cohort, while 2 separate cohorts (Europe and Asia) served as validation cohorts. Receiver operating characteristics analysis determined performance of the Lyon score in predicting treatment response. In 281 developmental cohort patients (median age 53 years, 57.7% female), the Lyon score demonstrated an area under the curve (AUC) of 0.819 in predicting 50% symptom improvement ( P < 0.001) on receiver operating characteristics, with an optimal threshold of 6.25 (sensitivity 81.2%, specificity 73.4%). Of the individual components, only acid exposure time (AUC 0.799, P < 0.001), mean nocturnal baseline impedance (AUC 0.785, P < 0.001), and reflux episodes (AUC 0.764, P < 0.001) approached the Lyon score performance. The Lyon score segregated treatment response in both the European (AUC 0.908, P < 0.001) and Asian validation cohorts (AUC 0.637, P < 0.001) and outperformed the DeMeester score in sensitivity for predicting outcome in the developmental and Asian validation cohorts. The novel Lyon score segregates reflux phenotypes and identifies likelihood of symptom response from antireflux therapy.

Clinical correlates of a high cardiorespiratory risk score for very low birth weight infants

Abstract Background A pulse oximetry warning system (POWS) to analyze heart rate and oxygen saturation data and predict risk of sepsis was developed for very low birth weight (VLBW) infants. Methods We determined the clinical correlates and positive predictive value (PPV) of a high POWS score in VLBW infants. In a two-NICU retrospective study, we identified times when POWS increased above 6 (POWS spike). We selected an equal number of control times, matched for gestational and chronologic age. We reviewed records for infection and non-infection events around POWS spikes and control times. We calculated the frequencies and PPV of a POWS spike for infection or another significant event. Results We reviewed 111 POWS spike times and 111 control times. Days near POWS spikes were more likely to have clinical events than control days (77% vs 50%). A POWS spike had 52% PPV for suspected or confirmed infection and 77% for any clinically significant event. Respiratory deterioration occurred near more POWS spike times than control times (34% vs 18%). Conclusions In a retrospective cohort, infection and respiratory deterioration were common clinical correlations of a POWS spike. POWS had a high PPV for significant clinical events with or without infection. Impact There are significant gaps in understanding the best approach to implementing continuous sepsis prediction models so that clinicians can best respond to early signals of deterioration. Infection and respiratory deterioration were common clinical events identified at the time of a high predictive model score. Understanding the clinical correlates of a high-risk early warning score will inform future implementation efforts.

Model-Based Prediction for Small Domains Using Covariates: A Comparison of Four Methods

Abstract We consider methods for model-based small area estimation when the number of areas with sampled data is a small fraction of the total areas for which estimates are required. Abundant auxiliary information is available from the survey for all the sampled areas. Further, through an external source, there is information for all areas. The goal is to use auxiliary variables to predict the outcome of interest for all areas. We compare areal-level random forests and LASSO approaches to a frequentist forward variable selection approach and a Bayesian shrinkage method using a horseshoe prior. Further, to measure the uncertainty of estimates obtained from random forests and the LASSO, we propose a modification of the split conformal procedure that relaxes the assumption of exchangeable data. We show that the proposed method yields intervals with the correct coverage rate and this is confirmed through a simulation study. This work is motivated by Ghanaian data available from the sixth Ghana Living Standards Survey (GLSS) and the 2010 Population and Housing Census, in the Greater Accra Metropolitan Area (GAMA) region, which comprises eight districts that are further divided into enumeration areas (EAs). We estimate the areal mean household log consumption using both datasets. The outcome variable is measured only in the GLSS for 3 percent of all the EAs (136 out of 5019) and 174 potential covariates are available in both datasets. In the application, among the four modeling methods considered, the Bayesian shrinkage performed the best in terms of bias, mean squared error (MSE), and prediction interval coverages and scores, as assessed through a cross-validation study. We find substantial between-area variation with the estimated log consumption showing a 1.3-fold variation across the GAMA region. The western areas are the poorest while the Accra Metropolitan Area district has the richest areas.

SIMAP500: A novel risk score to identify recipients at higher risk of hepatocellular carcinoma recurrence following liver transplantation.

Recurrence of hepatocellular carcinoma (HCC) following liver transplantation (LT) has a devastating influence on recipients' survival; however, the risk of recurrence is not routinely stratified. Risk stratification is vital with a long LT waiting time, as that could influence the recurrence despite strict listing criteria. This study aims to identify predictors of recurrence and develop a novel risk prediction score to forecast HCC recurrence following LT. A retrospective review of LT for HCC recipients at University Hospitals Birmingham between July 2011 and February 2020. Univariate and multivariate analyses were performed to identify recurrence predictors, based on which the novel SIMAP500 (satellite nodules, increase in size, microvascular invasion, AFP > 500, poor differentiation) risk score was proposed. 234 LTs for HCC were performed with a median follow-up of 5.3 years. Recurrence developed in 25 patients (10.7%). On univariate analyses, RETREAT score > 3, α-fetoprotein (AFP) at listing 100-500 and > 500, bridging, increased tumour size between imaging at the listing time and explant histology, increase in the size of viable tumour between listing and explant, presence of satellite nodules, micro- and macrovascular invasion on explant and poor differentiation of tumours were significantly associated with recurrence, based on which, the SIMAP500 risk score is proposed. The SIMAP500 demonstrated an excellent predictive ability (c-index = 0.803) and outperformed the RETREAT score (c-index = 0.73). SIMAP500 is indicative of the time to disease recurrence. SIMAP500 risk score identifies the LT recipients at risk of HCC recurrence. Risk stratification allows patient-centric post-transplant surveillance programs. Further validation of the score is recommended.

Acute COVID-19 severity markers predict post-COVID new-onset psychiatric disorders: A 2-year cohort study of 34,489 patients.

New-onset psychiatric disorders are frequent after COVID-19. We aim to determine whether acute COVID-19 severity markers can predict post-COVID new-onset psychiatric disorders. We conducted an electronic health records (EHR) cohort study of patients hospitalized for COVID-19 and without any known history of psychiatric disorders. Patients were included between January 2020 and September 2022 in one of the 36 university hospitals of the Assistance Publique - Hôpitaux de Paris. Acute COVID-19 clinical and biological severity markers were recorded during hospitalization for COVID-19. Psychiatric ICD-10 diagnoses were recorded up to 2 years and 9 months after hospitalization for COVID-19. Predictors of post-COVID new-onset psychiatric disorders were identified based on Cox regression models and sensitivity analyses. Predictive scores were built and tested in age- and sex-stratified populations. A total 34,489 patients hospitalized for COVID-19 were included; 3717 patients (10.8%) had at least one post-COVID new-onset psychiatric disorder. Hospital stay >7 days (HR = 1.72, 95%CI [1.59-1.86], p < 0.001), acute delirium (HR = 1.49, 95%CI [1.28-1.74], p < 0.001), elevated monocyte count (HR = 1.14, 95%CI [1.06-1.23], p < 0.001) and elevated plasma CRP (HR = 0.92, 95%CI [0.86-0.99], p = 0.04) independently predicted post-COVID new-onset psychiatric disorders. Sensitivity analyses confirmed hospital stay >7 days, acute delirium, and elevated monocyte count as predictors. Predictive scores based on these variables had good 12-month positive predictive values, up to 7.5 times more accurate than random in women < 65 years. In conclusion, hospital stay >7 days, acute delirium, and elevated monocyte count during acute COVID-19 predict post-COVID new-onset psychiatric disorders.

Waste management and water quality evaluation prediction in urban environments through advanced robust hybrid machine learning algorithms

Water quality management is a crucial aspect of environmental protection, requiring the monitoring and regulation of effluent discharges into surface water bodies. This research introduces a novel approach to predicting Water Quality Evaluation (WQE) through a unique hybrid model, ABC-DWKNN-ICA, which integrates the Distance-weighted K-Nearest Neighbors (DWKNN) algorithm with the Artificial Bee Colony (ABC), Firefly Algorithm (FA), Imperialist Competitive Algorithm (ICA), and Gravitational Search Algorithm (GSA). Utilizing a comprehensive dataset of 1106 data points from Telangana, India, spanning 2018–2020, the study examines a range of water quality parameters, including Ground Water Level (GWL), Potential of Hydrogen (PH), Electrical Conductivity (EC), and others. The ABC-DWKNN-ICA model demonstrates exceptional performance in terms of Recall, Precision, Accuracy, and F1 Score for WQE prediction, distinguishing itself with enhanced feature selection, improved classification accuracy, robustness to noise and outliers, reduced dimensionality, and scalability to large datasets. This hybrid model represents a significant advancement over existing approaches, including traditional Hybrid Machine Learning (HML) algorithms such as ABC-DWKNN, FA-DWKNN, ICA-DWKNN, and GSA-DWKNN. By focusing on the capabilities of ABC-DWKNN-ICA rather than comparing all HML algorithms, the research highlights its superior effectiveness in water quality prediction, with performance metrics of 0.83 for Recall, 0.86 for Precision, 0.91 for Accuracy, and 0.86 for F1 Score. This study thus fills a critical research gap by demonstrating the model's value in environmental data analysis and offering promising prospects for more effective management of water resources. Additionally, feature selection, dimensionality reduction, enhanced accuracy, noise handling, and imbalanced dataset management are key advantages of the proposed model.

Development and Validation of a Predictive Risk Score for Blood Transfusion in Patients Undergoing Curative-Intent Surgery for Intrahepatic Cholangiocarcinoma.

Among patients undergoing liver resection for intrahepatic cholangiocarcinoma (ICC), perioperative bleeding requiring blood transfusion is a common complication, yet preoperative identification of patients at risk for transfusion remains challenging. The objective of this study was to develop a preoperative risk score for blood transfusion requirement during surgery for ICC. Patients undergoing curative-intent liver surgery for ICC (1990-2020) were identified from a multi-institutional database. A predictive model was developed and validated. An easy-to-use risk calculator was made available online. Among 1420 patients, 300 (21.1%) received an intraoperative transfusion. Independent predictors of transfusion included severe preoperative anemia (OR = 1.65, 95% CI 1.10-2.47), T2 category or higher (OR = 2.00, 95% CI 1.36-3.02), positive lymph nodes (OR = 1.75, 95% CI 1.32-2.32) and major resection (OR = 2.56, 95%CI 1.85-3.58). Receipt of blood transfusion significantly correlated with worse outcomes. The model showed good discriminative ability in both training (AUC = 0.68, 95% CI 0.66-0.72) and bootstrapping validation (C-index = 0.67, 95% CI 0.65-0.70) cohorts. An online risk calculator of blood transfusion requirement was developed (https://catalano-giovanni.shinyapps.io/TransfusionRisk). Intraoperative blood transfusion was significantly associated with poor postoperative outcomes among patients undergoing surgery for ICC. The identification of patients at high risk of transfusion could improve perioperative patient care and blood resources allocation.

Abstract B052: Single-cell and spatial analysis of the immune landscape unveils a subset of potentially tumor-reactive T cells in patients with localized PDAC

Abstract Even in its localized stages, pancreatic ductal adenocarcinoma (PDAC) is virtually incurable, characterized by early dissemination and disease recurrence. The liver is the most common site of metastasis, and in most cases hepatic recurrences drive survival outcomes. This underscores the critical need to target and prevent liver metastases. We leveraged a novel biobanking strategy to perform paired single-cell RNA (scRNA-seq) and TCR sequencing (scTCR-seq) to profile multisite immunity in patients with localized PDAC across the tumor, peripheral blood, and premetastatic liver. This analysis of three patients included three tumor, three blood, and nine liver (three per patient from separate segments) samples, and generated 85,748 cells for downstream analysis. We evaluated populations of tumor-expanded T cells as potential indicators of antitumoral immunity, and utilized validated computational approaches to identify tumor-reactive T cells based on key markers of dysfunction and exhaustion. Gene set enrichment analysis was performed on tumor infiltrating lymphocytes (TILs) using signatures specific to tumor-reactive CD4 and CD8 T cells (“NeoTCR4” and “NeoTCR8,” respectively), and scored these using the AUCell package in R. Overall, 153 CD4 and 22 CD8 T cells fit candidate NeoTCR4 and NeoTCR8 signature states, representing 5.6% and 0.8% of the total TIL population respectively. These nominated T cells expressed key marker genes associated with tumor specificity such as CXCL13, TIGIT, PDCD1, ENTPD1, and TOX. Thirteen of these predicted tumor-reactive T cell clones identified within the tumor were also detected within our liver samples, suggestive of a potential antitumoral immune response in the premetastatic liver space. We next performed 10x Xenium in situ spatial transcriptomic analysis of primary tumor, normal adjacent pancreas, and premetastatic liver from two of these patients, using tissue microarrays to incorporate replicate 2mm core biopsies from each of these sites. A custom designed gene panel, guided by our scRNA-seq dataset, targeted tumor-specific mutations (e.g. KRAS G12V, G12D, G12R, and Q61) and immune related genes with a specific focus on markers of T cell activity and exhaustion. Leveraging these latter marker genes, which represent a subset of the NeoTCR4 and NeoTCR8 signatures, we identified a population of exhausted TILs within the tumor microenvironment, representing potentially tumor-reactive T cells whose interactions can now be explored at spatial resolution. Future directions are aimed at experimentally validating the tumor reactivity of the predicted neoantigen-specific TILs within our dataset and characterizing the immune neighborhoods that exist in both the tumor and liver of PDAC patients. This work establishes exciting opportunities for novel T-cell therapy in the treatment of PDAC, and we have already implemented these experimental findings in a forthcoming clinical trial investigating the applicability of this predictive signature score for the generation of TCR-engineered lymphocyte therapy in patients with PDAC. Citation Format: Elishama N Kanu, Ashley A Fletcher, Sri Krishna, Frank J Lowery, Haotian Zhuang, Ethan S Agritelley, Jiayin Bao, Austin M Eckhoff, Karrie Comatas, Tao Wang, Bin-Jin Hwang, Michael E Lidsky, Sabino Zani, Dan G Blazer III, Peter J Allen, Nicholas D Klemen, Zhicheng Ji, Daniel P Nussbaum, Erika J Crosby. Single-cell and spatial analysis of the immune landscape unveils a subset of potentially tumor-reactive T cells in patients with localized PDAC [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Advances in Pancreatic Cancer Research; 2024 Sep 15-18; Boston, MA. Philadelphia (PA): AACR; Cancer Res 2024;84(17 Suppl_2):Abstract nr B052.

Prognostic clinical and biological markers for amyotrophic lateral sclerosis disease progression: validation and implications for clinical trial design and analysis

With increasing recognition of the value of incorporating prognostic markers into amyotrophic lateral sclerosis (ALS) trial design and analysis plans, there is a pressing need to understand which among the prevailing clinical and biochemical markers have real value, and how they can be optimally used. A subset of patients with ALS recruited through the multi-center Phenotype-Genotype-Biomarker study (clinicaltrials.gov: NCT02327845) was identified as "trial-like" based on meeting common trial eligibility criteria. Clinical phenotyping was performed by evaluators trained in relevant assessments. Serum neurofilament light (NfL) and phosphorylated neurofilament heavy (pNfH), urinary p75ECD, plasma microRNA-181, and an array of biochemical and clinical measures were evaluated for their prognostic value. Associations with functional progression were estimated by random-slopes mixed models of ALS functional rating scale-revised (ALSFRS-R) score. Associations with survival were estimated by log-rank test and Cox proportional hazards regression. Potential sample size savings from adjusting for given biomarkers in a hypothetical trial were estimated. Baseline serum NfL is a powerful prognostic biomarker, predicting survival and ALSFRS-R rate of decline. Serum NfL <40pg/mL and >100pg/mL correspond to future ALSFRS-R slopes of ∼0.5 and ∼1.5 points/month, respectively. Serum NfL also adds value to the best available clinical predictors, encapsulated by the European Network to Cure ALS (ENCALS) predictor score. In models of functional decline, the addition of NfL yields ∼25% sample size saving above those achieved by inclusion of either clinical predictors or ENCALS score alone. The prognostic value of serum pNfH, urinary p75ECD, and plasma miR-181ab is more limited. Among the multitude of biomarkers considered, only blood NfL adds value to the ENCALS prediction model and should be incorporated into analysis plans for all ongoing and future ALS trials. Defined thresholds of NfL might also be used in trial design, for enrichment or stratified randomisation, to improve trial efficiency. NIH (U01-NS107027, U54-NS092091). ALSA (16-TACL-242).

Development of a Scoring System for Predicting the Difficulty of Bile Duct Cannulation and Selecting the Appropriate Cannulation Method.

Selective biliary cannulation in endoscopic retrograde cholangiopancreatography can be challenging due to factors like papillary morphology. Various patterns indicate cannulation difficulty, but the combinations causing difficulty and the optimal cannulation method for each scenario are unclear. This study aimed to identify cannulation difficulty patterns and develop a predictive scoring system for selecting the appropriate cannulation method. We retrospectively compared 776 patients with naïve papilla, dividing them into conventional contrast cannulation (N = 510) and salvage technique (N = 266) groups. The salvage group included patients using pancreatic duct guidewire placement and/or wire-guided cannulation due to difficulties with the contrast method. Papillary morphology (Haraldsson's classification), periampullary diverticulum (PAD), and scope operability were analyzed using multiple regression to identify risk factors for cannulation difficulties. Factors were scored based on hazard ratios to access combinations causing difficulties. The salvage group had more older patients and higher frequencies of type 2 (small), type 3 (protruding or pendulous), type 4 (creased or ridged) papillae, PAD, and poor scope operability. Significant risk factors in the multivariate analysis included type 2 [odds ratio (OR) 6.88], type 3 (OR 7.74), type 4 (OR 4.06) papillae, PAD (OR 2.26), and poor scope operability (OR 4.03). Pattern recognition scores were significantly higher in the salvage group (1.31 vs. 3.43, P < 0.0001). Type 2-4 papillae, PAD, and poor scope operability are significant risk factors for cannulation difficulty. Pattern recognition scores based on these factors can predict cannulation difficulty and aid in selecting between conventional and salvage methods.

Maximum Diameter of Ileocecal Lymph Nodes Measured Using Abdominal Ultrasonography Allows for the Discrimination of Yersinia pseudotuberculosis Infection from Kawasaki Disease.

This study aimed to determine whether the maximum diameter of ileocecal lymph nodes measured using abdominal ultrasonography is useful for differentiating Yersinia pseudotuberculosis infection from Kawasaki disease in the acute phase. The optimal maximum diameter cutoff of the ileocecal lymph nodes was also explored to optimize differentiation between these 2 diseases. We included pediatric patients <15 years old who met the diagnostic criteria for Kawasaki disease. Stool culture testing, loop-mediated isothermal amplification of stool specimens, and serological diagnosis were performed to confirm the presence or absence of Y. pseudotuberculosis infection. Of the 122 patients included in the analysis, 17 were confirmed to have Y. pseudotuberculosis infection and 105 were not. The age (in months), white blood cell count, C-reactive protein level, prediction score (risk score) for nonresponse to intravenous immunoglobulin, and number of intravenous immunoglobulin doses did not differ significantly between the Y. pseudotuberculosis-positive and -negative groups. The maximum diameter of ileocecal lymph nodes was 6.0 (5.5-9.5) mm in the Y. pseudotuberculosis-positive group and 3.0 (2.5-3.8) mm in the Y. pseudotuberculosis-negative group (numbers presented as median, interquartile range), with a significantly larger diameter in the Y. pseudotuberculosis-positive group (P < 0.01, Mann-Whitney U test), suggesting potency of ultrasonography. In patients meeting the diagnostic criteria for Kawasaki disease, the possibility of Y. pseudotuberculosis infection is significantly higher if the maximum ileocecal lymph node diameter ≥5.1 mm. Its sensitivity and specificity being 100%, and 89.5%, respectively.

Assessing the Utility of Prediction Scores PAINT, ISARIC4C, CHIS, and COVID-GRAM at Admission and Seven Days after Symptom Onset for COVID-19 Mortality.

The COVID-19 pandemic underscores the need for accurate prognostic tools to predict patient outcomes. This study evaluates the effectiveness of four prominent COVID-19 prediction scores-PAINT, ISARIC4C, CHIS, and COVID-GRAM-at two critical time points: at admission and seven days post-symptom onset, to assess their utility in predicting mortality among hospitalized patients. Conducted at the Clinical Emergency Hospital Pius Brînzeu in Timișoara, this retrospective analysis included adult patients hospitalized with confirmed SARS-CoV-2 infection. Eligible patients had complete data for the scores at both time points. Statistical analysis involved ROC curves and logistic regression to assess the scores' predictive accuracy for mortality. The study included 215 patients, split into 139 survivors and 76 non-survivors. At admission, the PAINT, ISARIC4C, CHIS, and COVID-GRAM scores significantly differentiated between the survival outcomes (p < 0.0001). The best cutoff values at admission were 6.26 for PAINT, 7.95 for ISARIC4C, 5.58 for CHIS, and 0.63 for COVID-GRAM, corresponding to sensitivities of 85.47%, 80.56%, 88.89%, and 83.33% and specificities of 77.34%, 82.12%, 75.01%, and 78.45%, respectively. By day seven, the cutoff values increased, indicating deteriorating conditions in patients who eventually succumbed to the virus. The hazard ratios at admission for exceeding these cutoffs were significant: PAINT (HR = 3.45), ISARIC4C (HR = 2.89), CHIS (HR = 4.02), and COVID-GRAM (HR = 3.15), highlighting the scores' abilities to predict severe outcomes. One week post symptom onset, these scores' predictive values and corresponding hazard ratios increased, further validating their prognostic significance over time. The evaluated COVID-19 prediction scores robustly predict mortality at admission and become more predictive by the seventh day of symptom onset. These findings support the use of these scores in clinical settings to facilitate early identification and intervention for high-risk patients, potentially improving patient outcomes during the ongoing global health crisis.

Establishment and validation of a predictive model for spontaneous preterm birth in singleton pregnant women

IntroductionIn the current study, we screened for highly sensitive and specific predictors of premature birth, with the aim to establish an sPTB prediction model that is suitable for women in China and easy to operate and popularize, as well as to establish a sPTB prediction scoring system for early, intuitive, and effective assessment of premature birth risk.MethodsA total of 685 pregnant women with a single pregnancy during the second trimester (16–26 weeks) were divided into premature and non-premature delivery groups based on their delivery outcomes. Clinical and ultrasound information were collected for both groups, and risk factors that could lead to sPTB in pregnant women were screened and analyzed using a cut-off value. A nomogram was developed to establish a prediction model and scoring system for sPTB. In addition, 119 pregnant women who met the inclusion criteria for the modeling cohort were included in the external validation of the model. The accuracy and consistency of the model were evaluated using the area under the receiver operating characteristic (ROC) and C-calibration curves.ResultsMultivariate logistic regression analysis showed a significant correlation (P < 0.05) between the number of miscarriages in pregnant women, history of miscarriages in the first week of pregnancy, history of preterm birth, CL of pregnant women, open and continuous cervical opening, and the occurrence of sPTB in pregnant women. We drew a nomogram column chart based on the six risk factors mentioned above, obtained a predictive model for sPTB, and established a scoring system to divide premature birth into three risk groups: low, medium, and high. After validating the model, the Hosmer Lemeshow test indicated a good fit (p = 0.997). The modeling queue C calibration curve was close to diagonal (C index = 0.856), confirming that the queue C calibration curve was also close to diagonal (C index = 0.854). The AUCs of the modeling and validation queues were 0.850 and 0.881, respectively.ConclusionOur predictive model is consistent with China’s national conditions, as well as being intuitive and easy to operate, with wide applicability, thus representing a helpful tool to assist with early detection of sPTB in clinical practice, as well as for clinical management in assessing low, medium, and high risks of sPTB.

Prediction Model of In-Hospital Death for Acute Exacerbation of Chronic Obstructive Pulmonary Disease Patients Admitted to Intensive Care Unit: The PD-ICU Score

Introduction: Patients with acute exacerbation of chronic obstructive pulmonary disease (AECOPD) admitted to intensive care unit (ICU) are exposed to poor clinical outcomes, and no specific prognostic models are available among this population. We aimed to develop and validate a risk score for prognosis prediction for these patients. Methods: This was a multicenter observation study. AECOPD patients admitted to ICU were included for model derivation from a prospective, multicenter cohort study. Logistic regression analysis was applied to identify independent predictors for in-hospital death and establish the prognostic risk score. The risk score was further validated and compared with DECAF, BAP-65, CURB-65, and APACHE II score in another multicenter cohort. Results: Five variables were identified as independent predictors for in-hospital death in APCOPD patients admitted to ICU, and a corresponding risk score (PD-ICU score) was established, which was composed of procalcitonin >0.5 μg/L, diastolic blood pressure <60 mm Hg, need for invasive mechanical ventilation, disturbance of consciousness, and blood urea nitrogen >7.2 mmol/L. Patients were classified into three risk categories according to the PD-ICU score. The in-hospital mortality of low-risk, intermediate-risk, and high-risk patients was 0.3%, 7.3%, and 27.9%, respectively. PD-ICU score displayed excellent discrimination ability with an area under the receiver-operating characteristic curve (AUC) of 0.815 in the derivation cohort and 0.754 in the validation cohort which outperformed other prognostic models. Conclusion: We derived and validated a simple and clinician-friendly prediction model (PD-ICU score) for in-hospital mortality among AECOPD patients admitted to ICU. With good performance and clinical practicability, this model may facilitate early risk stratification and optimal decision-making among these patients.

Development and validation of machine-learning models of diet management for hyperphenylalaninemia: a multicenter retrospective study

BackgroundAssessing dietary phenylalanine (Phe) tolerance is crucial for managing hyperphenylalaninemia (HPA) in children. However, traditionally, adjusting the diet requires significant time from clinicians and parents. This study aims to investigate the development of a machine-learning model that predicts a range of dietary Phe intake tolerance for children with HPA over 10 years following diagnosis.MethodsIn this multicenter retrospective observational study, we collected the genotypes of phenylalanine hydroxylase (PAH), metabolic profiles at screening and diagnosis, and blood Phe concentrations corresponding to dietary Phe intake from over 10 years of follow-up data for 204 children with HPA. To incorporate genetic information, allelic phenotype value (APV) was input for 2965 missense variants in the PAH gene using a predicted APV (pAPV) model. This model was trained on known pheno-genotype relationships from the BioPKU database, utilizing 31 features. Subsequently, a multiclass classification model was constructed and trained on a dataset featuring metabolic data, genetic data, and follow-up data from 3177 events. The final model was fine-tuned using tenfold validation and validated against three independent datasets.ResultsThe pAPV model achieved a good predictive performance with root mean squared error (RMSE) of 1.53 and 2.38 on the training and test datasets, respectively. The variants that cause amino acid changes in the region of 200–300 of PAH tend to exhibit lower pAPV. The final model achieved a sensitivity range of 0.77 to 0.91 and a specificity range of 0.8 to 1 across all validation datasets. Additional assessment metrics including positive predictive value (0.68–1), negative predictive values (0.8–0.98), F1 score (0.71–0.92), and balanced accuracy (0.8–0.92) demonstrated the robust performance of our model.ConclusionsOur model integrates metabolic and genetic information to accurately predict age-specific Phe tolerance, aiding in the precision management of patients with HPA. This study provides a potential framework that could be applied to other inborn errors of metabolism.

Analysis of an Infectious Disease Vaccination Prediction System Based on the MF-Conv LSTM Model

Infectious diseases can seriously threaten people's life safety and have a serious impact on social stability. Therefore, it should improve society’s stability under infectious diseases and ensure the safety of people's lives. A personnel flow feature extraction model based on Multi-Feature Convolutional Long Short-Term Memory (MF-Conv LSTM) is designed based on the characteristics of human daily activity behavior. This can optimize the accuracy of transmission simulation prediction for infectious disease vaccination. When using multi-feature ensemble analysis to extract human daily activity features as input for infectious disease simulation and prediction models, the learner's prediction score for the recurrent infectious diseases reached 0.8705. When using multi-feature ensemble analysis, the predicted scores did not exceed 0.85. The designed infectious disease vaccine transmission prediction model can accurately simulate the infectious behavior of infectious diseases. This provides direction for developing strategies to disrupt the infectious diseases’ spread. This reduces the infectious diseases’ harm to people's personal safety and improves social stability during the spread of large-scale infectious diseases.

Dissecting transcriptome signals of anti-PD-1 response in lung adenocarcinoma

Immune checkpoint blockades are actively adopted in diverse cancer types including metastatic melanoma and lung cancer. Despite of durable response in 20–30% of patients, we still lack molecular markers that could predict the patient responses reliably before treatment. Here we present a composite model for predicting anti-PD-1 response based on tumor mutation burden (TMB) and transcriptome sequencing data of 85 lung adenocarcinoma (LUAD) patients who received anti-PD-(L)1 treatment. We found that TMB was a good predictor (AUC = 0.81) for PD-L1 negative patients (n = 20). For PD-L1 positive patients (n = 65), we built an ensemble model of 100 XGBoost learning machines where gene expression, gene set activities and cell type composition were used as input features. The transcriptome-based models showed excellent accuracy (AUC > 0.9) and highlighted the contribution of T cell activities. Importantly, nonresponder patients with high prediction score turned out to have high CTLA4 expression, which suggested that neoadjuvant CTLA4 combination therapy might be effective for these patients. Our data and analysis results provide valuable insights into developing biomarkers and strategies for treating LUAD patients using immune checkpoint inhibitors.

A new scoring system to predict survival in elderly advanced stage Hodgkin lymphoma patients

Predictive prognostic scoring (PS) systems are not primarily applicable to elderly patients with classical Hodgkin lymphoma (cHL). The objective of this study was to develop a PS system for these patients. The derivation cohort (DC) was utilized for model development, consisting of 97 variables. The resulting algorithm was named as Hodgkin’s Lymphoma Early Death in the Elderly within 12 months (HEDEL12). Internal and external validation cohorts (IVC and EVC) were employed for validation. A total of 286 patients were evaluated retrospectively. In DC 38 of 178 patients died within the first 12 months and overall survival (OS) at 12-month was 78.6%. Independent predictors of HEDEL12 were female sex, low albumin levels (<3.5 g/dL), and ECOG scores 2-4. According to HEDEL12 scores 0–1, OS at 12- months were 89.8% and 91.0% for IVC and EVC, respectively. The HEDEL12 scoring is useful in predicting the survival of advanced-stage cHL patients.

Automated Medical Records Review for Mild Cognitive Impairment and Dementia.

Unstructured and structured data in electronic health records (EHR) are a rich source of information for research and quality improvement studies. However, extracting accurate information from EHR is labor-intensive. Here we introduce an automated EHR phenotyping model to identify patients with Alzheimer's Disease, related dementias (ADRD), or mild cognitive impairment (MCI). We assembled medical notes and associated International Classification of Diseases (ICD) codes and medication prescriptions from 3,626 outpatient adults from two hospitals seen between February 2015 and June 2022. Ground truth annotations regarding the presence vs. absence of a diagnosis of MCI or ADRD were determined through manual chart review. Indicators extracted from notes included the presence of keywords and phrases in unstructured clinical notes, prescriptions of medications associated with MCI/ADRD, and ICD codes associated with MCI/ADRD. We trained a regularized logistic regression model to predict the ground truth annotations. Model performance was evaluated using area under the receiver operating curve (AUROC), area under the precision-recall curve (AUPRC), accuracy, specificity, precision/positive predictive value, recall/sensitivity, and F1 score (harmonic mean of precision and recall). Thirty percent of patients in the cohort carried diagnoses of MCI/ADRD based on manual review. When evaluated on a held-out test set, the best model using clinical notes, ICDs, and medications, achieved an AUROC of 0.98, an AUPRC of 0.98, an accuracy of 0.93, a sensitivity (recall) of 0.91, a specificity of 0.96, a precision of 0.96, and an F1 score of 0.93 The estimated overall accuracy for patients randomly selected from EHRs was 99.88%. Automated EHR phenotyping accurately identifies patients with MCI/ADRD based on clinical notes, ICD codes, and medication records. This approach holds potential for large-scale MCI/ADRD research utilizing EHR databases.

Impact of an emergency department nurse training intervention on the adequacy of thromboprophylaxis for venous thromboembolism: the PROTESU III study.

Objectives. To assess the impact of training for emergency department (ED) nurses on adequate thromboprophylaxis for patients admitted to hospital from the ED for medical conditions. Methods. Multicenter quasiexperimental pre-post study of an ED nurse training intervention in 8 hospitals. Patients were recruited from January 2022 through May 2023 in 3 phases: before nurse training, in the first month after training, and in the sixth month after training. Included were patients attended in the ED for medical conditions. Adequate thromboprophylaxis was defined as 1) use of prophylactic drugs in patients at high risk for venousthromboembolism according to the Padua Prediction Score (PPS), and 2) nonuse in patients at low risk. We compared the percentage of adequate prophylaxis in the first phase to the percentages in the second and third phases. Results. A total of 928 patients were included (326 in phase 1, 295 in phase 2, and 307 in phase 3). PPS scores indicated that 238 (73%) of the patients were at high risk in phase 1 vs 189 (64.1%, P = .016) in phase 2 and 207 (67.4%, P = .125) in phase 3. A total of 187 patients (57.4%, 95% CI, 51.8%-62.8%) were adequatelythromboprophylaxed in phase 1 vs 178 (60.%, 95% CI, 54.5%-66%) in phase 2 (absolute difference in proportions, 3.0% (95% CI, -4.8% to 10.6%; P = .462)]. In phase 3, 166 patients (54.1%, 95% CI, 48.3%-59.7%) received adequate prophylaxis (difference, -3.3% (95% CI, -11.0% to 4.4%; P = .405). Conclusions. A training intervention for ED nurses, implemented as an isolated strategy, had no impact on the adequacy of thromboprophylaxis in patients admitted from the ED for medical conditions.