PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “Prognostic model”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 325 records · Page 18Linked to original sources

The myelodysplastic syndromes: classification and prognosis.

Myelodysplastic syndrome is a neoplastic clonal stem cell disorder characterized clinically by bone marrow failure and a tendency to progress to acute myelogenous leukemia. Dysplasia is the pathologic hallmark. The French-American-British classification served as the gold standard for more than two decades. Under the auspice of the World Health Organization, more than 100 hematopathologists in a 3-year cumulative effort issued the new World Health Organization classification, which recognizes multilineage dysplasia. Refractory anemia with excess blasts is divided into two groups. Chronic myelomonocytic leukemia is reclassified under a separate category. Refractory anemia with excess blasts in the transformation group was omitted. Finally, 5q-syndrome is a new subgroup. In addition to the pathologic classification, various prognostic predictors were formatted into scoring systems. Bone marrow blast percentage, cytopenias, and cytogenetics are the backbone for those prognostic models. The International Prognostic Scoring System is a product of pooled data from previous scoring systems and a useful tool to predict survival and acute myelogenous leukemia evolution. This paper discusses the classification and prognosis of myelodysplastic syndromes and their evolution.

Classification↗

Statistical models for predicting response to interferon-alpha and spontaneous seroconversion in children with chronic hepatitis B.

To develop prognostic models for identifying children with hepatitis B who are likely to respond to interferon-alpha (IFN-alpha) or to spontaneously seroconvert, we evaluated results of a multinational controlled trial comprising 70 children with chronic hepatitis B who received IFN-alpha and 74 children who did not receive therapy. Prognostic models were developed using SMILES (similarity of least squares), which is a data analysis network that incorporates multidimensional relationships in the clinical data of complex diseases. Commonly collected clinical data included age, gender, serum aminotransferase (aspartate aminotransferase [AST] and alanine aminotransferase [ALT]) and hepatitis B virus (HBV) DNA levels, and IFN-alpha dose. Additional data included pretreatment directional information (e.g. increases or decreases in serum aminotransferase and HBV DNA levels), liver biopsy results, race and transmission mode. Using data available prior to initiation of treatment, the SMILES models achieved prospective predictions of 89% for responders, 96% for non-responders, 100% for seroconverters and 93% for non-seroconverters. Although not predictive by themselves, the variables that had the greatest impact on predictions for IFN-alpha response were HBV DNA pretreatment direction, baseline HBV DNA, IFN-alpha dose and gender. The variables that had the greatest impact on predictions for spontaneous seroconversion were ALT pretreatment direction, baseline HBV DNA level, age and AST pretreatment direction. Therefore, these models may be useful in determining, in children with hepatitis B, the likelihood of response to IFN-alpha and spontaneous seroconversion.

Adolescent↗

Alternative splicing in ovarian cancer.

Ovarian cancer is the second leading cause of gynecologic cancer death worldwide, with only 20% of cases detected early due to its elusive nature, limiting successful treatment. Most deaths occur from the disease progressing to advanced stages. Despite advances in chemo- and immunotherapy, the 5-year survival remains below 50% due to high recurrence and chemoresistance. Therefore, leveraging new research perspectives to understand molecular signatures and identify novel therapeutic targets is crucial for improving the clinical outcomes of ovarian cancer. Alternative splicing, a fundamental mechanism of post-transcriptional gene regulation, significantly contributes to heightened genomic complexity and protein diversity. Increased awareness has emerged about the multifaceted roles of alternative splicing in ovarian cancer, including cell proliferation, metastasis, apoptosis, immune evasion, and chemoresistance. We begin with an overview of altered splicing machinery, highlighting increased expression of spliceosome components and associated splicing factors like BUD31, SF3B4, and CTNNBL1, and their relationships to ovarian cancer. Next, we summarize the impact of specific variants of CD44, ECM1, and KAI1 on tumorigenesis and drug resistance through diverse mechanisms. Recent genomic and bioinformatics advances have enhanced our understanding. By incorporating data from The Cancer Genome Atlas RNA-seq, along with clinical information, a series of prognostic models have been developed, which provided deeper insights into how the splicing influences prognosis, overall survival, the immune microenvironment, and drug sensitivity and resistance in ovarian cancer patients. Notably, novel splicing events, such as PIGV|1299|AP and FLT3LG|50,941|AP, have been identified in multiple prognostic models and are associated with poorer and improved prognosis, respectively. These novel splicing variants warrant further functional characterization to unlock the underlying molecular mechanisms. Additionally, experimental evidence has underscored the potential therapeutic utility of targeting alternative splicing events, exemplified by the observation that knockdown of splicing factor BUD31 or antisense oligonucleotide-induced BCL2L12 exon skipping promotes apoptosis of ovarian cancer cells. In clinical settings, bevacizumab, a humanized monoclonal antibody that specifically targets the VEGF-A isoform, has demonstrated beneficial effects in the treatment of patients with advanced epithelial ovarian cancer. In conclusion, this review constitutes the first comprehensive and detailed exposition of the intricate interplay between alternative splicing and ovarian cancer, underscoring the significance of alternative splicing events as pivotal determinants in cancer biology and as promising avenues for future diagnostic and therapeutic intervention.

Humans↗

Comprehensive investigation identifies CPSF3 as a novel prognostic and oncogenic biomarker in bladder cancer.

BACKGROUND: Bladder cancer (BC) remains a prevalent malignancy worldwide, with rising incidence rates each year. Despite progress in therapeutic strategies, many patients suffer recurrence or progression, emphasizing the urgent need for novel prognostic biomarkers and therapeutic targets. This research evaluated the prognostic relevance and functional role of Cleavage and Polyadenylation Specificity Factor 3 (CPSF3) in BC. METHODS: We analyzed CPSF3 expression using The Cancer Genome Atlas data and immunohistochemistry on a cohort of 203 BC patients. A nomogram incorporating CPSF3 expression was developed based on CPSF3 expression for prediction of overall survival and disease-free survival. Immune infiltration analyses and transcriptome sequencing were performed to explore underlying biological mechanisms. In vitro and in vivo experiments were utilized to examine the results of CPSF3 silencing on bladder cancer cell growth, colony-forming ability and cell cycle transitions. RESULTS: Elevated CPSF3 expression was significantly linked to unfavorable overall survival and disease-free survival both in TCGA datasets and our cohort. The CPSF3-based nomogram outperformed conventional prognostic models. CPSF3 expression was associated with tumor-infiltrating immune cells and immune checkpoint markers. Enrichment analysis revealed CPSF3 enrichment in cell cycle-related pathways. Suppression of CPSF3 expression led to marked reductions in cell proliferation, colony formation, tumor growth in animal models and inhibited G1 to S phase progression. CONCLUSION: CPSF3 is a promising prognostic biomarker for BC and may play a crucial role in BC progression. Incorporating CPSF3 into clinical prognostic models may enhance prediction of patient outcomes. CPSF3 may represent a promising therapeutic target for BC management.

Bladder cancer↗

Overview of prognostic factors in non-Hodgkin's lymphoma.

The non-Hodgkin's lymphomas are a biologically heterogeneous group of diseases with varying clinical presentations and outcomes. A number of studies have identified variables that carried independent prognostic significance. Although several staging systems had evolved that incorporated these prognostic variables, they were still unable to predict outcome. Ideally, the object of a staging system is to predict the likelihood of treatment response, time to progression or disease-free survival, and overall survival, and to provide a way to compare the outcome of similar groups of patients among various clinical trials. The need for such a system led to the creation of prognostic models such as the M.D. Anderson Tumor Score and, more recently, the International Prognostic Index. These prognostic models may identify those patients at highest risk for treatment failure, thereby identifying those patients who may require different therapeutic approaches.

Biomarkers, Tumor↗

Pretransplant prediction of prognosis after liver transplantation in primary sclerosing cholangitis using a Cox regression model.

Liver transplantation remains the only treatment for patients with end-stage primary sclerosing cholangitis (PSC); however, selection criteria for the procedure and its timing remains uncertain. The aim of this study was to identify pretransplant variables associated with survival after transplantation and to devise a Cox regression model for prediction of post-transplant survival. We studied 118 patients transplanted for PSC at the Queen Elizabeth Hospital, Birmingham, UK, being followed for up to 91/4 years after the procedure. The association between pretransplant data and the post-transplant survival up to 1 year was studied using the logrank test (univariate analyses) and Cox multiple regression analysis. Univariate analyses showed the following variables to be associated with a decreased post-transplant survival: high serum creatinine, high serum bilirubin, biliary tree malignancy, previous upper abdominal surgery, hepatic encephalopathy, ascites, and Crohn's disease, whereas ulcerative colitis was associated with increased post-transplant survival (all P </=.05). The final multiple Cox regression model included the following significant variables: inflammatory bowel disease, ascites, previous upper abdominal surgery, serum creatinine, and biliary tree malignancy (all P <.03). Biliary tree malignancy could be omitted from the Cox model with only slight loss of information. The results were validated using the data of 30 independent PSC patients from another center. These results can improve selection of patients with PSC for liver transplantation. The developed prognostic model for transplantation can be used in parallel with previously published prognostic models for nontransplantation. The obtained prognostic estimates will provide additional information that is useful for optimal timing of liver transplantation in the individual patient.

Adult↗

Integration of single-cell transcriptomics and genomic mutation analysis identifies an immunotherapy-resistant tumor subcluster and validates ARNTL2 as a malignant driver in lung adenocarcinoma.

BACKGROUND: Immunotherapy resistance in lung adenocarcinoma (LUAD) remains a critical clinical challenge, and the mechanisms underlying resistance-associated intratumoral heterogeneity are poorly characterized. METHODS: We performed single-cell RNA sequencing of LUAD patients receiving neoadjuvant immunotherapy (responders vs. non-responders), integrating inferCNV, GSVA, and differential expression analyses. Cluster-specific genes were validated across seven independent cohorts (TCGA-LUAD, GSE13213, GSE26939, GSE29016, GSE30219, GSE31210, GSE42127). A multi-algorithm machine learning framework was used to construct a prognostic model, and the immune microenvironment was characterized using TCIA scoring, seven infiltration algorithms, and ESTIMATE. ARNTL2 function was assessed by CCK-8 and Transwell assays in A549 and H1299 cells. RESULTS: Non-responders showed significant enrichment of epithelial cells, depletion of cytotoxic T/NK cells, and elevated copy number variation burden versus responders (p < 0.0001). A resistance-enriched malignant subcluster (Cluster 2) exhibited hyperproliferative and metabolic reprogramming signatures with upregulated KRT17, S100A2, and CST6, which showed tumor-specific overexpression, adverse prognostic value, and genomic amplification across cohorts. CoxBoost combined with survivalSVM achieved optimal predictive performance (C-index = 0.686), yielding robust risk stratification (HR: 2.54-10.51, all p < 0.05). Low-risk patients showed greater immune infiltration and higher TCIA immunophenoscores. ARNTL2 was an independent prognostic factor (HR: 2.07-4.64) strongly correlated with risk score (r = 0.69), and its knockdown suppressed proliferation and invasion in both LUAD cell lines (all p < 0.05). CONCLUSION: This study identifies a resistance-associated malignant subcluster in LUAD, constructs a validated CoxBoost + survivalSVM prognostic model with robust immune stratification, and establishes ARNTL2 as a core oncogenic driver and therapeutic target.

ARNTL2↗

Prognostic factors in non-seminomatous germ cell tumours of the testis. Experience at the Institut Gustave-Roussy.

The original prognostic model in non-seminomatous germ cell tumours of the testis at the Institut Gustave-Roussy was updated in a larger patient population. As in the former model human chorionic gonadotrophin (hCG) and alpha-feto-protein (AFP) serum concentrations attained prognostic significance in the multivariate analysis. We studied two important issues in this patient population: the impact of expressing hCG values in mass units and of the introduction of etoposide-containing chemotherapy. Etoposide has a major impact on the determination of the prognostic group limits which are considerably increased. It is, thus, suggested to reevaluate prognostic models in patient populations treated by modern chemotherapy regimens.

Antineoplastic Combined Chemotherapy Protocols↗

Integrative analysis of single-cell sequencing identifies CD8+ TIM3+ CD101+ T cell-associated genes as prognostic biomarkers in breast cancer.

BACKGROUND: Breast cancer is a prevalent and deadly malignancy that significantly impacts women's quality of life and imposes financial burdens. Despite therapeutic advancements, tumour heterogeneity and frequent relapses remain major challenges. Accordingly, this study aimed to characterize immune features associated with CD8+ TIM3+ CD101+ T cells and develop a prognostic signature for breast cancer. METHODS: This study integrated single-cell and bulk transcriptomic datasets to characterize CD8+ TIM3+ CD101+ T cell (CCT)-related immune features and construct a prognostic signature in breast cancer. Single-cell RNA-seq data were sourced from the Gene Expression Omnibus (GEO) repository, and bulk transcriptomic data were from The Cancer Genome Atlas (TCGA) and GEO databases. Analytical methods included pseudo-time trajectory reconstruction (Monocle2), intercellular signalling analysis (CellChat), functional enrichment (ClusterProfiler), and immune profiling (ssGSEA). Prognostic modeling was conducted using least absolute shrinkage and selection operator (LASSO) Cox regression, with validation via Kaplan-Meier and time-dependent receiver operating characteristic (ROC) analyses. RESULTS: Single-cell analysis identified 17 clusters spanning seven cell types, including T cells, myeloid cells, and epithelial cells. T-cell sub-clustering revealed four subtypes. Pseudotime analysis suggested a potential state-transition relationship between CD8+ CD101- TIM3+ and CD8+ CD101+ TIM3+ T-cell states. A total of 121 differentially expressed genes were enriched in vital biological processes. An 11-gene prognostic model showed strong predictive power across cohorts. Single-cell T-cell reclustering identified a CD8+ CD101+ TIM3+ T-cell subpopulation, which was primarily characterized by the expression of markers such as CD101 and HAVCR2/TIM3. CONCLUSIONS: This study maps cellular heterogeneity and molecular networks in breast cancer, offering insights for targeted therapy and improved prognosis.

Breast invasive carcinoma↗

Prospective study of indolent non-follicular non-Hodgkin's lymphoma: validation of Gruppo Italiano per lo studio dei linfomi (GISL) prognostic criteria for watch and wait policy.

Only recently both the Revised European American Lymphoma (REAL) and World Health Organization (WHO) classifications clearly identified indolent non-follicular non-Hodgkin's lymphoma (NHL) as a distinct group of precise histological entities. Therefore, prognostic models, specifically designed for this NHL subset, are still lacking. In this study, we prospectically evaluated the prognostic criteria proposed by the Gruppo Italiano per lo studio dei linfomi (GISL) to identify patients with an indolent non-progressive clinical course, eligible for a watch and wait policy within this histological subset defined according to stringent criteria of histomorphology and immunophenotype. Fifty-three patients affected with small lymphocytic, marginal zone, lymphoplasmacytic lymphoma and lacking at presentation the following: B symptoms, bulky disease, anemia, thrombocytopenia, diffuse pattern of bone marrow infiltration and short tumor doubling time, were registered in a prospective therapeutic GISL trial and addressed to a watch and wait program. After 41.3 months of median follow-up, the median progression free survival (PFS) was not reached and 73% of cases did not progress. When additional variables were considered, in order to improve the prognostic model, it was evident that LDH level and the number of extranodal sites were of statistical significance in the multivariate analysis. Based on this finding, a prognostic score was devised which was able to further identify a small group of patients more likely to undergo early progression, and thus suitable for immediate treatment. In conclusion, the GISL definition of indolent disease is a reliable tool to design the appropriate therapeutic strategy in this histological setting.

Adult↗

The timing of liver transplantation in primary biliary cirrhosis.

Primary biliary cirrhosis, the most common chronic cholestatic liver disease in adults, usually progresses to cirrhosis and its complications. Ursodeoxycholic acid therapy delays disease progression, but most patients will ultimately succumb. Liver transplantation is now accepted as the standard treatment for end-stage PBC. Development of major complications of portal hypertension and liver failure, poor quality of life and short survival without transplantation are the major indications for this surgical intervention in patients with primary biliary cirrhosis. Resource use is another key variable to be considered in the timing of liver transplantation. Prognostic models have been developed to predict survival and resource utilization with and without liver transplantation. Prognostic models aid the clinician in the selection and timing of liver transplantation in the patient with primary biliary cirrhosis.

Adult↗

The relationship between an objective response to chemotherapy and survival in advanced colorectal cancer.

This analysis was conducted to evaluate the independent relationship between survival and response to chemotherapy in advanced colorectal cancer. In order to correct for the guarantee time effect, patients dying before the response evaluation were excluded from the analyses. A previously constructed prognostic model containing 11 variables was applied to 324 patients. When the response categories were analysed together with the prognostic variables, it was found that a response was associated with a definite survival advantage (P < 0.001), whereas the influence of all the other variables decreased. The corrected survival advantage (relative progressive disease) was 11 months after a complete response, 6 months after a partial response and 4 months after stable disease. The survival advantage was of a similar magnitude when the analyses were repeated in an independent population comprising 198 patients in whom the prognostic model was extended to include also a set of laboratory values. The results show that a response to chemotherapy is associated with a longer survival also after correction for the guarantee time effect and the distribution of prognostic variables.

Adult↗

Risk factors predicting complications in blood culture-proven typhoid fever in adults.

To create a prognostic model for complications of blood culture-proven typhoid fever in adults (> or = 15 y old), a retrospective cohort was assembled though review of the medical records of the hospitalized patients treated for typhoid fever over a 3-y period ending January 1995. Of the 59 patients included, 21 (35.6%) developed various complications of typhoid fever. No patient included died. Four baseline variables (abdominal pain, systolic blood pressure < 100 mmHg, hypoalbuminaemia < 32 g/l and laboratory evidence of disseminated intravascular coagulation) were independently associated with complications and were used to create a prognostic model. The prediction accuracy of the model was determined using the concordance index (c-index). The results (c-index, 0.915 [95%, CI, 89.0-93.0]) showed that the model predicted complications significantly better than chance. The model stratified patients into 3 prognostic stages: low risk for complications (0%; stage I), intermediate risk (42.9%; stage II) and high risk (92.3%; stage III) (p = 0.001). If validated in other settings, it will help clinicians in predicting complications in patients with blood culture-proven typhoid fever on admission.

Abdominal Pain↗

A per- and polyfluoroalkyl substances-based gene signature links prognosis to immune landscapes in thyroid cancer.

BACKGROUND: Thyroid cancer (THCA) is the most common endocrine malignancy with a rising global incidence and significant heterogeneity. Although per- and polyfluoroalkyl substances (PFAS) exposure is linked to thyroid dysfunction, the prognostic value of per- and polyfluoroalkyl substances-related genes (PFASRGs) and their role in the tumor immune microenvironment (TME) remain poorly understood. This study aims to systematically screen key PFASRGs and evaluate their prognostic value as biomarkers for THCA. METHODS: Utilizing The Cancer Genome Atlas (TCGA)-THCA transcriptomic data and PFASRGs, we constructed a prognostic model through differential expression analysis, univariate and multivariate Cox regression analyses, and the least absolute shrinkage and selection operator (LASSO). The model's robustness was validated using receiver operating characteristic (ROC) curves, Kaplan-Meier analysis, and clinical nomograms. Furthermore, the TME, immunotherapy response, and drug sensitivities were systematically evaluated. Distinct molecular landscapes were characterized by stratifying the cohort via unsupervised consensus clustering analysis. RESULTS: The eight-gene prognostic model demonstrated robust performance, with area under the curve (AUC) values exceeding 0.85 across all validation cohorts. High-risk patients exhibited significantly shorter overall survival and an "inflamed" TME characterized by high immune scores and checkpoint expression. In contrast, the therapeutic efficacy of anti-programmed death-ligand 1 (PD-L1) agents was more pronounced in the low-risk category, as evidenced by a superior objective response. Furthermore, distinct molecular subtypes and risk-specific sensitivities to targeted agents, such as sorafenib and sunitinib, were identified, highlighting the model's clinical utility for personalized treatment. CONCLUSIONS: We established a novel THCA prognostic framework based on eight PFASRGs. This model exhibits superior performance in risk stratification, effectively distinguishing cohorts with divergent clinical trajectories, unique immune microenvironment features, and varied therapeutic responses. Our findings provide a powerful predictive tool for refining prognostic evaluation and facilitating the implementation of personalized management strategies for THCA patients.

Per- and polyfluoroalkyl substances-related genes ↗

scRNA-seq and bulk RNA-seq reveal the characteristics of macrophage copper metabolism and establish a risk signature in hepatocellular carcinoma.

BACKGROUND: Hepatocellular carcinoma (HCC) is a prevalent malignancy with an urgent need for improved prognostic stratification and treatment-response prediction. This study aimed to explore a macrophage copper metabolism-associated prognostic model and to investigate the relationship between this risk model and the tumor immune microenvironment. METHODS: The FindClusters function was used to analyze cell clusters, and CellChat and CellPhoneDB/LIANA were employed for cell-cell communication analysis. Copper metabolism-related genes were sourced from the MSigDB database. A prognostic risk model was established using least absolute shrinkage and selection operator (LASSO) analysis and multivariate Cox regression analysis, and a nomogram was constructed by integrating the prognostic model with clinicopathological factors. Additional analyses were performed to map the seven model genes in single-cell data, assess model uncertainty and robustness, evaluate macrophage/copper/cuproptosis-related transcriptional programs, and examine the correlations between risk score, immune infiltration and predicted drug sensitivity. RESULTS: Using single-cell RNA sequencing (scRNA-seq) data, we identified four macrophage subpopulations. Macrophages with high SPP1 expression showed close interaction with T cell populations and were associated with copper ion metabolism. By incorporating 141 copper metabolism-related genes and using The Cancer Genome Atlas Liver Hepatocellular Carcinoma (TCGA-LIHC) cohort, we constructed a seven-gene risk prediction model. Additional single-cell mapping showed that the model genes were detectable in the HCC single-cell dataset and showed a macrophage-associated expression pattern. The model showed moderate prognostic discrimination in TCGA-LIHC, whereas its external performance was heterogeneous and remained evaluable across external cohorts, with performance varying among datasets. Immune and mechanism-related analyses suggested that the risk signature was associated with macrophage-related infiltration, copper metabolism and cuproptosis-related transcriptional programs. Drug sensitivity analysis nominated Daporinad as a computationally predicted candidate compound, supporting Daporinad as a pharmacogenomic candidate for follow-up investigation. CONCLUSIONS: By integrating scRNA-seq and bulk RNA sequencing (RNA-seq) data, we constructed a macrophage copper metabolism-associated prognostic signature for HCC. The risk score was associated with survival, immune microenvironment features and predicted drug response, providing a transcriptomic framework for risk stratification and therapeutic hypothesis generation.

Hepatocellular carcinoma (HCC)↗

Community-acquired bacterial meningitis: risk stratification for adverse clinical outcome and effect of antibiotic timing.

BACKGROUND: Community-acquired bacterial meningitis causes substantial morbidity and mortality in adults. OBJECTIVE: To create and test a prognostic model for persons with community-acquired bacterial meningitis and to determine whether antibiotic timing influences clinical outcome. DESIGN: Retrospective cohort study; patients were divided into derivation and validation samples. SETTING: Four hospitals in Connecticut. PATIENTS: 269 persons who, between 1970 and 1995, had community-acquired bacterial meningitis microbiologically proven by a lumbar puncture done within 24 hours of presentation in the emergency department. MEASUREMENTS: Baseline clinical and laboratory features and times of arrival in the emergency department, performance of lumbar puncture, and administration of antibiotics. The target end point was the development of an adverse clinical outcome (death or neurologic deficit at discharge). RESULTS: For the total group, the hospital mortality rate was 27%. Fifty-six of 269 patients (21 %) developed a neurologic deficit, and in 9% the neurologic deficit persisted at discharge. Three baseline clinical features (hypotension, altered mental status, and seizures) were independently associated with adverse clinical outcome and were used to create a prognostic model from the derivation sample. The prediction accuracy of the model was determined by using the concordance index (c-index). For both the derivation sample (c-index, 0.73 [95% CI, 0.65 to 0.81]) and the validation sample (c-index, 0.81 [CI, 0.71 to 0.92]), the model predicted adverse clinical outcome significantly better than chance. For the total group, the model stratified patients into three prognostic stages: low risk for adverse clinical outcome (9%; stage I), intermediate risk (33%; stage II), and high risk (56%; stage III) (P=0.001). Adverse clinical outcome was more common for patients in whom the prognostic stage advanced from low risk (P=0.008) or intermediate risk (P=0.003) at arrival in the emergency department to high risk before administration of antibiotics. CONCLUSIONS: In persons with community-acquired bacterial meningitis, three baseline clinical features of disease severity predicted adverse clinical outcome and stratified patients into three stages of prognostic severity. Delay in therapy after arrival in the emergency department was associated with adverse clinical outcome when the patient's condition advanced to the highest stage of prognostic severity before the initial antibiotic dose was given.

Adolescent↗

Using logistic regression analysis in preliminary differential diagnosis of adnexal masses.

The aim of our study was to generate predictive model, which would allow to estimate the influence of analyzed parameters on predictive accuracy of differential diagnosis of adnexal masses and to evaluate prospectively diagnostic efficacy of the statistic model in the new set of patients. A total of 686 women diagnosed and surgically treated in the Gynecological and Obstetrical Teaching Hospital of University of Medical Sciences in Poznan, Poland, were recruited into the study. Preoperative diagnostics included gynecological examination, ultrasonographic evaluation, tumor Doppler blood flow analysis, and serum levels of CA125 and TPS. In order to find the best combination of features and to calculate the individual probability of the malignancy, stepwise logistic regression analysis with quasi-Newton estimation was applied. The essential part of the best prognostic model, described by foregoing variables, is as follows: [z = -6.005 + 0.058 x age + 1.174 x septa + 1.317 x tumor localization + 1.185 x ascites + 2.28 x solid element + 2.429 x vessels localization -2.386 x PI + 0.084 x MEDV]. The highest sensitivity and specificity for the obtained model were 87.84% and 93.74%, respectively. Prognostic model, constructed with the use of logistic regression analysis, is characterized by higher sensitivity and specificity than individually applied diagnostic tests. Prospective evaluation of this model application in a larger group of patients with adnexal masses will enable precise assessment of its objective clinical usefulness.

Adnexal Diseases↗

Prediction of survival in patients with esophageal carcinoma using artificial neural networks.

BACKGROUND: Accurate estimation of outcome in patients with malignant disease is an important component of the clinical decision-making process. To create a comprehensive prognostic model for esophageal carcinoma, artificial neural networks (ANNs) were applied to the analysis of a range of patient-related and tumor-related variables. METHODS: Clinical and pathologic data were collected from 418 patients with esophageal carcinoma who underwent resection with curative intent. A data base that included 199 variables was constructed. Using ANN-based sensitivity analysis, the optimal combination of variables was determined to allow creation of a survival prediction model. The accuracy (area under the receiver operator characteristic curve [AUR]) of this ANN model subsequently was compared with the accuracy of the conventional statistical technique: linear discriminant analysis (LDA). RESULTS: The optimal ANN models for predicting outcomes at 1 year and 5 years consisted of 65 variables (AUR = 0.883) and 60 variables (AUR = 0.884), respectively. These filtered, optimal data sets were significantly more accurate (P < 0.0001) than the original data set of 199 variables. The majority of ANN models demonstrated improved accuracy compared with corresponding LDA models for 1-year and 5-year survival predictions. Furthermore, ANN models based on the optimal data set were superior predictors of survival compared with a model based solely on TNM staging criteria (P < 0.0001). CONCLUSIONS: ANNs can be used to construct a highly accurate prognostic model for patients with esophageal carcinoma. Sensitivity analysis based on ANNs is a powerful tool for seeking optimal data sets.

Diagnosis, Computer-Assisted↗