SEARCH · Search PubMed

Results for “predictive modelling”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

SourceResearch connectionFrom yearThrough year

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

At least 163 records · Page 9Linked to original sources

PubMed2001

[Splice site mutations and atherosclerosis: mechanisms and prediction models].

Nucleotide variants in genes of the lipid metabolism influence the risk of premature atherosclerosis. Ten percent of all single nucleotide substitutions in these genes involve splice sites. The effects of these changes on mRNA splicing and phenotypic severity, however, are not inherently obvious from the nucleotide sequence. This review presents various genes of lipid metabolism with splicing mutations known to influence the risk of premature atherosclerosis. Mechanisms of pre-mRNA splicing are illustrated and different models for prediction of the effect of nucleotide substitutions on splice-site function are presented. The role of information theory-based models is emphasized along with its role for prediction of splice-site function and phenotypic severity of atherosclerosis.

Age Factors↗

PubMed1978

Predictive model of dynamic response of the human head/neck system to -Gx impact acceleration.

This paper describes the mathematical framework, underlying an empirical model, that predicts human head response using only the motion present at vertebra T1. Based on this framework, a model for --Gx impact acceleration was developed from data obtained on six volunteer subjects participating in the NAMRL impact acceleration experiments. Model performance was evaluated by comparing the errors in the predicted head responses with the normal variations observed between the responses of different subjects under identical impact accelerations. Independent sets of data were used for building and testing the model. The results of the evaluation indicate that the model will be useful in subsequent studies of human response to impact acceleration.

Acceleration↗

PubMed2004

Uro-gramma: the importance of updating the predictive model.

OBJECTIVES: Uro-gramma is a probabilistic predictive model of pathological staging of prostate cancer (Pca) from preoperative parameters (PSA, clinical Gs, clinical stage) published in 2000. Aim of this study is to improve Uro-gramma, updating it, to take into account the continuous evolution of the population. MATERIALS AND METHODS: From 1998 to 2000, 991 Pca patients have undergone radical prostatectomy in several Italian urological centers. Inclusion criteria were: preoperative PSA < 50 ng/ml, clinical stage < or = T3c, availability of a bioptic Gs and pathological staging. A predictive model has been estimated for each year and its behaviour on the following years tested, using Hosmer and Lemeshow tests, which compare the expected rate with the observed one. RESULTS: The mean age was 66.3 years. Pca familiarity was present in 3.2% of the patients in 1998, 2.6% in 1999 and 7.4% in 2000. PSA values < 10 have increased (from 41% to 47%) and those > 10 decreased (from 59% to 53%). The mean number of bioptic samples per patient has increased from 4.9 to 6, while clinical and pathologic Gs have remained stable. An increase in the rate of organ confined Pca has been noticed, either clinically (87.4% in 1998, 92% in 2000) or pathologically (55.2% in '98, 57% in 2000). Staging lymphadenectomy has been performed in 88% of the pts in 1998, 94.2% in 1999 and 94.5% in 2000, whereas the % of N+ patients has moved from 11.3% in 1998 to 9.8% in 2000. CONCLUSIONS: Uro-gramma update results show that the mean estimate error has fallen from 6.4% to 1.2%. The new model strengthens the previous one and confirms its validity, but it also underlines the need of a constant update to take into account the continuous evolution of the population and of the methods of diagnosis and staging.

Aged↗

PubMed2006

[Development of a predicting model of survival rates for patients with salivary adenoid cystic carcinoma].

OBJECTIVE: To establish a predicting model of survival rates and to evaluate the weighted contributions of each key prognostic factor of the patients with salivary adenoid cystic carcinoma (SACC). METHODS: One hundred and eighteen follow-up cases with SACC were analyzed for the survival study with retrospective cohort method. Ten possible clinical and pathologic factors were selected. A multivariate analysis was performed by Cox proportional hazard model and prognostic index (PI) was calculated. According to the PI, all cases were divided into three risk subgroups respectively: lower, intermediate and higher risk subgroups. Ten-year survival rate and median survival time were calculated and the predicting models of survival rates were established. RESULTS: The significant prognostic factors influencing the survival rate were age at diagnosis, clinical presentation, TNM clinical stage, treatment, surgical margins (P < 0.05). The predicting formula was PI = 0.031X(2) + 0.665X(5) + 0.420X(6)-0.576X(7) + 0.999X(10). According to the value of PI, the prognosis of the patients was significantly different among the three subgroups (P < 0.05). In the three risk subgroups, 10-year survival rates were 83.56%, 31.45% and 11.20% respectively, the median survival time was 18 years, 7 years and 4 years respectively. CONCLUSIONS: The established predicting model of survival rates can predict the prognosis of the patients with SACC.

Adolescent↗

PubMed2026

Risk prediction models for blood transfusion in patients undergoing total hip and knee arthroplasty: a systematic review and meta-analysis.

OBJECTIVE: To systematically review and evaluate published risk prediction models for perioperative blood transfusion in patients undergoing total hip or knee arthroplasty (THA/TKA). METHODS: We systematically searched PubMed, Web of Science, the Cochrane Library, and Embase from inception to May 31, 2025. Two researchers independently screened the literature, extracted data, and assessed the risk of bias and applicability using the Prediction model Risk Of Bias Assessment Tool (PROBAST). The area under the receiver operating characteristic curve (AUC) values were pooled via a meta-analysis using Stata 18.0. RESULTS: d Fourteen studies containing 36 prediction models were included. The incidence of blood transfusion among THA/TKA patients ranged from 3.2% to 30.8%. Preoperative hemoglobin (Hb) level, tranexamic acid (TXA) use, operative duration, intraoperative blood loss, and age were the most frequently incorporated predictors. Model sensitivity ranged from 58% to 94.5%, and specificity ranged from 71.3% to 94%. Meta-analysis showed that the pooled AUC value of the 13 validated models was 0.87 (95% CI: 0.85-0.90), suggesting good discriminatory performance. All models were rated as having a high risk of bias. The applicability of four studies was rated as unclear. CONCLUSION: Although the included studies demonstrated promising discriminative ability of prediction models for blood transfusion in THA/TKA, all were assessed as having a high risk of bias using the PROBAST tool. Therefore, future research should prioritize the development of models with larger sample sizes, rigorous study designs, and multicenter external validation.

Humans↗

PubMed2026

The Progress of Gout Prediction Models Based on Multi-source Data.

INTRODUCTION: Gout, a highly serious inflammatory disease that is caused by monosodium urate crystals, is becoming an increasingly significant health concern. Artificial Intelligence and multi-omics-based research have made significant gains for the early detection and prevention of gout based on diverse approaches. This review intends to summarize current advances in forecasting gout susceptibility and gout-related symptoms, evaluate the predictive efficacy of different features, and ascertain which clinical and omics characteristics are most effective in these prediction models. METHODS: We explored the PubMed database after 2010 using keywords such as "gout", "predictive model", "risk prediction", and "machine learning", and confined our search to Englishlanguage articles. The original peer-reviewed research articles that developed gout models were selected. Research that was not original or lacked internal validation was excluded. RESULTS: Clinical features, genomics, microbiomics, radiomics, and metabolomics have been utilized to construct models related to gout and have demonstrated excellent predictive performance. Multisource data prediction models usually exhibit better effectiveness. DISCUSSION: Gout-oriented models performed excellently in predictive performance but present limitations in certain clinical and omics domains. However, if they are to affect actual patient care, they must overcome some external confirmation roadblocks and the fiscal and practical implications they will face ahead of time. CONCLUSION: This review indicates that clinical and multi-omics models of gout are significant instruments for clinical decision-making. The models constructed in these studies may be crucial for the treatment of gout and its practical benefits.

Gout↗

PubMed2003

Stratification of adverse outcomes by preoperative risk factors in coronary artery bypass graft patients: an artificial neural network prediction model.

We constructed and internally validated an artificial neural network (ANN) model for prediction of in-hospital major adverse outcomes (defined as death, cardiac arrest, coma, renal failure, cerebrovascular accident, reinfarction, or prolonged mechanical ventilation) in patients who received "on-pump" coronary artery bypass grafting (CABG) surgery. We retrospectively analyzed a 5-year CABG surgery database with a final study population of 563 patients. Predictive variables were limited to information available before the procedure, and outcome variables were represented only by events that occurred postoperatively. The ANN's ability to discriminate outcomes was assessed using receiver-operating characteristic (ROC) analysis and the results were compared with a multivariate logistic regression (LR) model and the QMMI risk score (RS) model. A major adverse outcome occurred in 12.3% of all patients and 18 predictive variables were identified by the ANN model. Pairwise comparison showed that the ANN model significantly outperformed the RS model (AUC = 0.886 vs.0.752, p = 0.043). However, the other two pairs, ANN vs. LR models (AUC = 0.886 vs. 0.807, p = 0.076) and LR vs. RS models (AUC = 0.807 vs. 0.752, p = 0.453) performed similarly well. ANNs tend to outperform regression models and might be a useful screening tool to stratify CABG candidates preoperatively into high-risk and low-risk groups.

Cardiopulmonary Bypass↗

PubMed2000

How can the incidence of negative specimens resulting from large loop excision of the cervical transformation zone (LLETZ) be reduced? An analysis of negative LLETZ specimens and development of a predictive model.

OBJECTIVE: To analyse biopsies of large loop excision of the transformation zone of the cervix; to identify factors associated with negative histology; and to develop predictive models in order to reduce the number of negative loop excisions. DESIGN: Retrospective analysis of patient notes and audit database. SETTING: Colposcopy clinic of a large district general hospital in North Staffordshire. POPULATION: Four hundred and fifty-two women who underwent a large loop excision of the transformation zone (LLETZ) procedure for suspected cervical intraepithelial neoplasia. METHODS: Women who underwent a LLETZ procedure were placed in two different groups, one positive for cervical intra epithelial neoplasia and the other negative for cervical intra epithelial neoplasia. Information was obtained on a number of clinical and colposcopic variables. Analysis was undertaken to determine if there were any differences between the two groups. These factors were then identified and three predictive models generated. Receiver-operator characteristic curves were used to assess and test these models. MAIN OUTCOMES MEASURES: To identify factors associated with negative histology on a LLETZ specimen. To predict how to reduce the number of negative LLETZ specimens. RESULTS: Four hundred and fifty-two women underwent a LLETZ procedure, 88 were negative (19%) and 364 were positive (81%). In women who were treated at their first visit, 56/316 (18%) had negative histology. There were significant associations between negative histology in the LLETZ and negative or low grade cytological atypia, negative colposcopic findings and years of age > 50 in both bivariate analysis and stepwise logistic regression. In the predictive models, the sensitivity ranged between 72% and 80%, the specificity 59%-72%, and the area under the receiver-operator characteristic was 0.75-0.77. If we had used the predictor models and managed women with negative or low grade cervical atypia and negative colposcopy findings conservatively, we would have reduced the negative biopsy rate from 19% to 14%, but five cases of high grade disease and 25 cases of low grade disease would have been missed. If we had also included women aged > 50 years in this model, the negative biopsy rate would have dropped from 19% to 15%, with only one case of high grade disease and 11 cases of low grade disease missed. All these women would require continued cytological and colposcopic surveillance. Importantly, no cases of invasion would have been missed. CONCLUSION: Using a predictive model can reduce the number of negative LLETZ specimens, but at the expense of continued cytological and colposcopic surveillance and cannot be recommended in normal practice. This raises the question whether current standards for negative histology in LLETZ specimens are set unrealistically high.

Adolescent↗

PubMed1998

Biodegradation kinetics of naphthalene in nonaqueous phase liquid-water mixed batch systems: comparison of model predictions and experimental results.

A model is formulated to describe dissolution of naphthalene from an insoluble nonaqueous phase liquid (NAPL) and its subsequent biodegradation in the aqueous phase in completely mixed batch reactors. The physicochemical processes of equilibrium partitioning and mass transfer of naphthalene between the NAPL and aqueous phases were incorporated into the model. Biodegradation kinetics were described by Monod's microbial growth kinetic model, modified to account for the inhibitory effects of 1,2-naphthoquinone formed during naphthalene degradation under certain conditions. System parameters and biokinetic coefficients pertinent to the NAPL-water systems were determined either by direct measurement or from nonlinear regression of the naphthalene mineralization profiles obtained from batch reactor tests with two-component NAPLs comprised of naphthalene and heptamethylnonane. The NAPLs contained substantial mass of naphthalene, and naphthalene biodegradation kinetics were evaluated over the time required for near complete depletion of naphthalene from the NAPL. Model predictions of naphthalene mineralization time profiles compared favorably to the general trends observed in the data obtained from laboratory experiments with the two-component NAPL, as well as with two coal tars obtained from the subsurface at contaminated sites and composed of many different PAHs (polycyclic aromatic hydrocarbon compounds). The effects of varying the NAPL mass and the naphthalene mole fractions in the NAPL are discussed. It was observed that the time to achieve a given percent removal of naphthalene does not change significantly with the initial mass of naphthalene in a fixed volume of the NAPL. Significant changes in the mineralization profiles are observed when the volume (and mass) of NAPL in the system is changed.

Biodegradation, Environmental↗

PubMed1993

A predictive model for aggressive non-Hodgkin's lymphoma.

BACKGROUND: Although many patients with intermediate-grade or high-grade (aggressive) non-Hodgkin's lymphoma are cured by combination chemotherapy, the remainder are not cured and ultimately die of their disease. The Ann Arbor classification, used to determine the stage of this disease, does not consistently distinguish between patients with different long-term prognoses. This project was undertaken to develop a model for predicting outcome in patients with aggressive non-Hodgkin's lymphoma on the basis of the patients' clinical characteristics before treatment. METHODS: Adults with aggressive non-Hodgkin's lymphoma from 16 institutions and cooperative groups in the United States, Europe, and Canada who were treated between 1982 and 1987 with combination-chemotherapy regimens containing doxorubicin were evaluated for clinical features predictive of overall survival and relapse-free survival. Features that remained independently significant in step-down regression analyses of survival were incorporated into models that identified groups of patients of all ages and groups of patients no more than 60 years old with different risks of death. RESULTS: In 2031 patients of all ages, our model, based on age, tumor stage, serum lactate dehydrogenase concentration, performance status, and number of extranodal disease sites, identified four risk groups with predicted five-year survival rates of 73 percent, 51 percent, 43 percent, and 26 percent. In 1274 patients 60 or younger, an age-adjusted model based on tumor stage, lactate dehydrogenase level, and performance status identified four risk groups with predicted five-year survival rates of 83 percent, 69 percent, 46 percent, and 32 percent. In both models, the increased risk of death was due to both a lower rate of complete responses and a higher rate of relapse from complete response. These two indexes, called the international index and the age-adjusted international index, were significantly more accurate than the Ann Arbor classification in predicting long-term survival. CONCLUSIONS: The international index and the age-adjusted international index should be used in the design of future therapeutic trials in patients with aggressive non-Hodgkin's lymphoma and in the selection of appropriate therapeutic approaches for individual patients.

Age Factors↗

PubMed2004

The use of laboratory-determined ion exchange parameters in the predictive modelling of field-scale major cation migration in groundwater over a 40-year period.

An attempt has been made to estimate quantitatively cation concentration changes as estuary water invades a Triassic Sandstone aquifer in northwest England. Cation exchange capacities and selectivity coefficients for Na(+), K(+), Ca(2+), and Mg(2+) were measured in the laboratory using standard techniques. Selectivity coefficients were also determined using a method involving optimized back-calculation from flushing experiments, thus permitting better representation of field conditions; in all cases, the Gaines-Thomas/constant cation exchange capacity (CEC) model was found to be a reasonable, though not perfect, first description. The exchange parameters interpreted from the laboratory experiments were used in a one-dimensional reactive transport mixing cell model, and predictions compared with field pumping well data (Cl and hardness spanning a period of around 40 years, and full major ion analyses in approximately 1980). The concentration patterns predicted using Gaines-Thomas exchange with calcite equilibrium were similar to the observed patterns, but the concentrations of the divalent ions were significantly overestimated, as were 1980 sulphate concentrations, and 1980 alkalinity concentrations were underestimated. Including representation of sulphate reduction in the estuarine alluvium failed to replicate 1980 HCO(3) and pH values. However, by including partial CO(2) degassing following sulphate reduction, a process for which there is 34S and 18O evidence from a previous study, a good match for SO(4), HCO(3), and pH was attained. Using this modified estuary water and averaged values from the laboratory ion exchange parameter determinations, good predictions for the field cation data were obtained. It is concluded that the Gaines-Thomas/constant exchange capacity model with averaged parameter values can be used successfully in ion exchange predictions in this aquifer at a regional scale and over extended time scales, despite the numerous assumptions inherent in the approach; this has also been found to be the case in the few other published studies of regional ion exchanging flow.

Forecasting↗

PubMed1997

Comparative molecular field analysis-based predictive model of structure-function relationships of polyamine transport inhibitors in L1210 cells.

Maintenance of intracellular polyamine concentrations necessary for cell growth and proliferation is regulated in part by an energy-dependent polyamine uptake system. To obtain information on the characteristics of the polyamine uptake system in L1210 leukemia cells, we have applied computational chemistry techniques to the study of relationships between structure and function of 57 polyamine analogues. Ki values of polyamine analogues, derived from competitive inhibition of [3H]spermidine transport into L1210 cells, were chosen as the measure of biological activity. Using comparative molecular field analysis (CoMFA), a model was constructed to relate molecular structure with biological activity. The model was based on 4 monocationic, 8 dicationic, 14 tricationic, and 20 tetracationic polyamine analogues with a range of Ki values for the inhibition of [3H]spermidine uptake of 0.97-521 microM. The CoMFA model successfully predicted the inhibitory potency of 11 polyamines that had not previously been tested for polyamine uptake inhibitory activity. The 11 values predicted were within 33 +/- 62% of the actual Ki values. The test group included aziridinyl diamines, acetylated spermidines, two new oxazolidinonyl spermidines, monoaziridinyl spermidines, and a diaziridinyl spermine. Several of the compounds from this test group have been shown to have anticancer activity in mice. Consistent with the CoMFA model, certain basic functional groups, such as aziridines that have pKa values in the range of 6-7, seem to interact with the polyamine transporter in a cationic form. The results suggest that the CoMFA model is useful in drug design strategies as a predictive tool for the discovery of new anticancer agents that utilize a polyamine transporter for cellular uptake.

Animals↗

PubMed2000

A predictive model for life-threatening neutropenia and febrile neutropenia after the first course of CHOP chemotherapy in patients with aggressive non-Hodgkin's lymphoma.

The purpose of this study was to develop a model for predicting the occurrence of life-threatening neutropenia (LN, ANC < or = 0.5 x 10(9)/l) and febrile neutropenia (FN, an ANC < 0.5x10(9)/l in association with a body temperature of > or = 38.3 degrees C) after the first cycle of CHOP therapy in patients newly diagnosed with aggressive NHL. One hundred and forty-five patients, aged > or = 15 years, with newly diagnosed diffuse mixed, diffuse large-cell or large-cell immunoblastic lymphoma (IWF categories, F, G, H), who had been treated with CHOP at King Chulalongkorn Memorial Hospital between June 1994 and December 1998, were entered into the study. The criteria for eligibility included complete work-up for baseline evaluation, treatment with standard CHOP chemotherapy, at least one complete blood count performed during days 8-14 post-treatment or if at any time the patients experienced a BT of > or = 38.3 degrees C and were not treated with any colony-stimulating factors (CSFs). The median age of the patients was 47 years (range, 17-78). Forty-eight percent of the patients were in stage III/IV, 36% had ECOG performance status (PS) II-IV, 30% had > or = 2 extranodal diseases, 59% had serum LDH > 1 x normal and 23% had bone marrow involvement. The frequencies of patients in the low-, low-intermediate, high-intermediate and high risk groups according to the international index were 29%, 28%, 17% and 26%, respectively. Thirty-nine percent of the patients had LN at nadir and 33% developed FN after the first course of CHOP. By using stepwise logistic regression analysis, the pretreatment variables independently predictive of the LN at nadir and the FN were serum albumin concentration of < or = 3.5 g/dl, serum LDH > 1 x normal and whether there was bone marrow involvement of lymphoma at presentation. The model, based on the incorporation of these three factors, identified three risk groups of patients with a predicted probability of developing LN at nadir of 81.5% (95% CI, 68.5-90.7) (high risk), 23.9% (95% CI, 12.6-38.8) (intermediate risk) and 4.4% (95% CI, 0.5-15.1) (low risk). The predicted rate of FN in the three groups were 72.2% (95% CI, 58.4-83.5), 17.4% (95% CI, 7.8-31.4) and 2.2% (95% CI, 0.05-11.8), respectively. In conclusion, our model could be used as a means to identify patients with newly diagnosed aggressive NHL, treated with CHOP, who are at high risk (> or = 50% probability) of developing post-first course LN and FN, in whom CSF and/or antibiotic prophylaxis might be indicated.

Adolescent↗

PubMed2006

The effect of including C-reactive protein in cardiovascular risk prediction models for women.

BACKGROUND: While high-sensitivity C-reactive protein (hsCRP) is an independent predictor of cardiovascular risk, global risk prediction models incorporating hsCRP have not been developed for clinical use. OBJECTIVE: To develop and compare global cardiovascular risk prediction models with and without hsCRP. DESIGN: Observational cohort study. SETTING: U.S. female health professionals. PARTICIPANTS: Initially healthy nondiabetic women age 45 years and older participating in the Women's Health Study and followed an average of 10 years. MEASUREMENTS: Incident cardiovascular events (myocardial infarction, stroke, coronary revascularization, and cardiovascular death). RESULTS: High-sensitivity CRP made a relative contribution to global risk at least as large as that provided by total, high-density lipoprotein (HDL), and low-density lipoprotein (LDL) cholesterol individually, but less than that provided by age, smoking, and blood pressure. All global measures of fit improved when hsCRP was included, with likelihood-based measures demonstrating strong preference for models that include hsCRP. With use of 10-year risk categories of 0% to less than 5%, 5% to less than 10%, 10% to less than 20%, and 20% or greater, risk prediction was more accurate in models that included hsCRP, particularly for risk between 5% and 20%. Among women initially classified with risks of 5% to less than 10% and 10% to less than 20% according to the Adult Treatment Panel III covariables, 21% and 19%, respectively, were reclassified into more accurate risk categories. Although addition of hsCRP had minimal effect on the c-statistic (a measure of model discrimination) once age, smoking, and blood pressure were accounted for, the effect was nonetheless greater than that of total, LDL, or HDL cholesterol, suggesting that the c-statistic may be insensitive in evaluating risk prediction models. LIMITATIONS: Data were available only for women. CONCLUSIONS: A global risk prediction model that includes hsCRP improves cardiovascular risk classification in women, particularly among those with a 10-year risk of 5% to 20%. In models that include age, blood pressure, and smoking status, hsCRP improves prediction at least as much as do lipid measures.

Age Factors↗

PubMed1999

Validation of a predictive model for asthma admission in children: how accurate is it for predicting admissions?

We studied 364 index presentations to the Emergency Department of a children's hospital with a diagnosis of asthma. The admission rate for this group of children was about 31%. We developed a parsimonious multiple logistic regression model to predict asthma hospital admission based on asthma severity indicators. We then evaluated the model's predictive ability using two methods of cross-validation, using the same sample that was used for the predictive model, and using data from a split sample. The logistic regression model had a predictive accuracy of 90% (95% confidence interval 85-95%). The sensitivity and specificity were 86% and 88%, respectively. Cross-validation models confirmed that the predictive ability of the model was stable. In studies with limited sample sizes, it is possible to validate a model without setting aside a split sample for cross-validation.

Acute Disease↗

PubMed2003

Accident prediction models for urban roads.

This paper describes some of the main findings from two separate studies on accident prediction models for urban junctions and urban road links described in [Uheldsmodel for bygader-Del1: Modeller for 3-og 4-benede kryds. Notat 22, The Danish Road Directorate, 1995; Uheldsmodel for bygader- Del2: Modeller for straekninger. Notat 59, The Danish Road Directorate, 1998] (Greibe and Hemdorff, 1995, 1988). The main objective for the studies was to establish simple, practicable accident models that can predict the expected number of accidents at urban junctions and road links as accurately as possible. The models can be used to identify factors affecting road safety and in relation to 'black spot' identification and network safety analysis undertaken by local road authorities. The accident prediction models are based on data from 1036 junctions and 142 km road links in urban areas. Generalised linear modelling techniques were used to relate accident frequencies to explanatory variables. The estimated accident prediction models for road links were capable of describing more than 60% of the systematic variation ('percentage-explained' value) while the models for junctions had lower values. This indicates that modelling accidents for road links is less complicated than for junctions, probably due to a more uniform accident pattern and a simpler traffic flow exposure or due to lack of adequate explanatory variables for junctions. Explanatory variables describing road design and road geometry proved to be significant for road link models but less important in junction models. The most powerful variable for all models was motor vehicle traffic flow.

Accidents, Traffic↗

PubMed2002

Recent innovations in intensive care unit risk-prediction models.

During the past 20 years, ICU risk-prediction models have undergone significant development, validation, and refinement. Among the general ICU severity of illness scoring systems, the Acute Physiology and Chronic Health Evaluation (APACHE), Mortality Prediction Model (MPM), and the Simplified Acute Physiology Score (SAPS) have become the most accepted and used. To risk-adjust patients with longer, more severe illnesses like sepsis and acute respiratory distress syndrome, several models of organ dysfunction or failure have become available, including the Multiple Organ Dysfunction Score (MODS), the Sequential Organ Failure Assessment (SOFA), and the Logistic Organ Dysfunction Score (LODS). Recent innovations in risk adjustment include automatic physiology and diagnostic variable retrieval and the use of artificial intelligence. These innovations have the potential of extending the uses of case-mix and severity-of-illness adjustment in the areas of clinical research, patient care, and administration. The challenges facing intensivists in the next few years are to further develop these models so that they can be used throughout the IUC stay to assess quality of care and to extend them to more specific patient groups such as the elderly and patients with chronic ICU courses.

Artificial Intelligence↗

PubMed2003

A climate-based model predicts the spatial distribution of the Lyme disease vector Ixodes scapularis in the United States.

An understanding of the spatial distribution of the black-legged tick, Ixodes scapularis, is a fundamental component in assessing human risk for Lyme disease in much of the United States. Although a county-level vector distribution map exists for the United States, its accuracy is limited by arbitrary categories of its reported presence. It is unknown whether reported positive areas can support established populations and whether negative areas are suitable for established populations. The steadily increasing range of I. scapularis in the United States suggests that all suitable habitats are not currently occupied. Therefore, we developed a spatially predictive logistic model for I. scapularis in the 48 conterminous states to improve the previous vector distribution map. We used ground-observed environmental data to predict the probability of established I. scapularis populations. The autologistic analysis showed that maximum, minimum, and mean temperatures as well as vapor pressure significantly contribute to population maintenance with an accuracy of 95% (p < 0.0001). A cutoff probability for habitat suitability was assessed by sensitivity analysis and was used to reclassify the previous distribution map. The spatially modeled relationship between I. scapularis presence and large-scale environmental data provides a robust suitability model that reveals essential environmental determinants of habitat suitability, predicts emerging areas of Lyme disease risk, and generates the future pattern of I. scapularis across the United States.

Animals↗