Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “predictability”

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.

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

At least 1,441 records · Page 80Linked to original sources
PubMed2005

Long-term prospective on-line real-time seizure prediction.

OBJECTIVE: Epilepsy, one of the most common neurological disorders, constitutes a unique opportunity to study the dynamics of spatiotemporal state transitions in real, complex, nonlinear dynamical systems. In this study, we evaluate the performance of a prospective on-line real-time seizure prediction algorithm in two patients from a common database. METHODS: We previously demonstrated that measures of chaos and angular frequency, estimated from electroencephalographic (EEG) signals recorded at critical sites in the cerebral cortex, progressively converge (i.e. become dynamically entrained) as the epileptic brain transits from the asymptomatic interictal state to the ictal state (seizure) (Iasemidis et al., 2001, 2002a, 2003a). This observation suggested the possibility of developing algorithms to predict seizures well ahead of their occurrences. One of the central points in those investigations was the application of optimization theory, specifically quadratic zero-one programming, for the selection of the critical cortical sites. This current study combines that observation with a dynamical entrainment detection method to prospectively predict epileptic seizures. The algorithm was tested in two patients with long-term (107.54h) and multi-seizure EEG data B and C (Lehnertz and Litt, 2004). RESULTS: Analysis from the 2 test patients resulted in the prediction of up to 91.3% of the impending 23 seizures, about 89+/-15min prior to seizure onset, with an average false warning rate of one every 8.27h and an allowable prediction horizon of 3h. CONCLUSIONS: The algorithm provides warning of impending seizures prospectively and in real time, that is, it constitutes an on-line and real-time seizure prediction scheme. SIGNIFICANCE: These results suggest that the proposed seizure prediction algorithm could be used in novel diagnostic and therapeutic applications in epileptic patients.

Brain Mapping↗
PubMed2006

Predicted current densities in the brain during transcranial electrical stimulation.

OBJECTIVE: We sought an electrical modeling approach to evaluate the potential application of finite element method (FEM) modeling to predict current pathways and intensities in the brain after transcranial electrical stimulation. METHODS: A single coronal MRI section through the head, including motor cortex, was modeled using FEM. White matter compartments with both anatomically realistic anisotropies in resistivity and with a homogeneous resistivity were modeled. Current densities in the brain were predicted for electrode sites on the scalp and after theoretical application of a conductive head restraint device. RESULTS: Localized current densities were predicted for the model with white matter anisotropies. Differences in predicted peak current densities were related to location of stimulation sites relative to deep sulci in the brain and scalp shunting that was predicted to increase with inter-electrode proximity. A conductive head restraint device was predicted to shunt current away from the brain when a constant current source was used. CONCLUSIONS: The complex geometry of different tissue compartments in the head and their contrasting resistivities may jointly determine the strength and location of current densities in the brain after transcranial stimulation. This might be predictable with FEM incorporating white matter anisotropies. Conductive head restraint devices during surgery may be contraindicated with constant current stimulation. SIGNIFICANCE: Individually optimized tcMEP monitoring and localized transcranial activation in the brain might be possible through FEM modeling.

Anisotropy↗
PubMed2006

Averting iatrogenic hypoglycemia through glucose prediction in clinical practice: progress towards a new procedure in diabetes.

BACKGROUND: Hypoglycemia is a risk factor common to all insulin therapy. The hypothesis is that efforts to reduce or prevent this adverse side effect may fail because providers generally lack the resources to predict not only future blood glucose levels but also future risks of hypoglycemia. This lack has been remedied. A controlled study was undertaken to test the hypothesis. METHODS: Twenty-two insulin dependent subjects suffering more than one (1) episode/week of hypoglycemia with similar insulin regimens, similar diabetes education and similar self-management training participated in this study. For all subjects, a remote monitoring resource (registry and database) was used to capture daily SMBG and afford a return path for provider interventions and decision support. Identical telemedical methods were used which differed only for the provider either by the presence (prediction group) or by the absence (control group) of an on-screen, visual display of predicted glycemia and predicted risks of hypoglycemia. The study lasted 2 months. RESULTS: Over an average of 41 days from baseline to follow up and while using the glycemic prediction resource, providers intervened more effectively in the prediction group reducing rates of hypoglycemia nine-fold (P<0.0001) and insulin therapy by just -9 U/day (P<0.01). Mean pre-meal glycemia was not compromised. Over 61 days from baseline to final follow up but without glycemic predictions in the control group, providers' interventions were less effective and resulted in no net changes in rates of hypoglycemia, daily insulin therapy, or mean pre-meal glycemia. CONCLUSIONS: Given knowledge of future glycemia and future risks of hypoglycemia, providers in clinical practice can now avert iatrogenic hypoglycemia in less than 2 months. A shared diabetes data center furnishing remote data capture and decision support is fundamental to the implementation of this as a new clinical procedure in diabetes.

Adolescent↗
PubMedDate not supplied

Prediction of drug sensitivity and drug resistance in cancer by transcriptional and proteomic profiling.

The oncologist's challenges, particularly with advanced cancers, are (a) how to predict tumor response to a given drug or regimen; (b) how to predict which tumors of identical histology will remain indolent and which will be likely to progress; and (c) how to determine the appropriate timing of the emergence of drug-resistant cancer cells and hence switch to appropriate therapy. These issues are still unresolved; current clinical practice is hampered by the complexity and heterogeneity of anti-tumor drug resistance where multiple cellular, tumor microenvironment and host factors operate simultaneously. The rapid accumulation of genomic and proteomic databases for complex biological systems, such as cancer, together with advances in technology platforms, have paved the way to an increased molecular understanding and prediction of antitumor drug response. The complex phenotype of drug resistance can now be dissected and specific, clinically relevant markers pinpointed. Several microarray studies of genetic patterns from untreated and pre-treated cancers have provided "fingerprints" that can predict response to therapeutics. Nevertheless, such approaches require further validation in experimental models and in large clinical trials before their routine clinical use. Moreover, comparative transcriptional profiling alone is unlikely to predict drug sensitivity/resistance, a dynamic process where protein phosphorylation, protein trafficking, and protein-protein interactions with secondary effectors play key roles in the fate of cancer cells following therapeutic stress. Functional proteomics is potentially more predictive, but still faces technical challenges with regards to sampling, tumor heterogeneity, and lack of standardized methodologies. These obstacles are surmountable with current concerted research efforts and availability of powerful high-throughput genomic and proteomic instrumentations, and thus approaches to predict and overcome drug resistance could be rationalized.

Animals↗
PubMed2004

Prediction of successful vaginal delivery in women undergoing external cephalic version at term for breech presentation.

OBJECTIVE: The aim of this study is to evaluate the clinical and sonographic predictors of ultimate successful vaginal delivery in women undergoing external cephalic version. STUDY DESIGN: The study population consisted of women with external cephalic version performed at or after 36 weeks of gestation. They were randomized into group A or B, each consisted of half of the total study population. Logistic regression was performed on group A to identify the significant independent variables in predicting successful cephalic vaginal delivery, which were used to construct a prediction model. The derived regression model was then tested in group B to assess its accuracy. RESULTS: The study included 407 pregnancies. Maternal weight (kg) at the time of version, maternal height (cm), multiparity, engagement of fetal presenting part, and fetal head palpable were significant independent variables of successful version and vaginal delivery (regression coefficients are: -0.084, 0.085, 1.752, -1.271, and 0.725, respectively). A prediction model was constructed based on these independent variables. The weighted average of the overall accuracy in predicting success or failure of version and vaginal delivery was 70.9%. The regression model was then applied to study group B. Using 0.50 as the cutoff value, the sensitivity, specificity, positive, and negative predictive values were 75.4, 58.8, 73.7, and 61.0%, respectively. CONCLUSION: A regression model constructed based on clinical variables failed to provide an accurate predictive tool of successful external cephalic version and vaginal delivery. However, in women who are equivocal about external cephalic version, a high prediction of success would be encouraging.

Adult↗
PubMed2005

HKF-R 10 - screening for predicting chronicity in acute low back pain (LBP): a prospective clinical trial.

STUDY DESIGN: Prospective cohort study. OBJECTIVES: To develop a short instrument to reliably predict chronicity in low back pain (LBP). SUMMARY OF BACKGROUND DATA: Health care expenditures on the treatment of low back pain continue to increase. It is therefore important to prevent the development of chronicity. In Germany, there is at present no early risk assessment tool to predict the risk of developing chronic LBP for patients presenting with acute LBP. Undertaken in an orthopedic practice setting, this study examined known risk factors for chronicity. It resulted in the development of a short questionnaire that successfully predicted the course of chronicity with an accuracy of 78%. METHODS: A cohort of 192 orthopaedic outpatients was assessed for clinical, behavioral, emotional, and cognitive parameters bsed on a self-report test battery of 167 established items predictive for chronicity in LBP. Chronicity was defined as back pain persisting for longer than six months. Logistic regression analysis was performed to evaluate the predictive value of all items significantly associated with the dependent variable. RESULTS: The study found the following items to have the strongest predictive value in the development of chronicity: "How strong was your back pain during the last week when it was most tolerable?" and the question "How much residual pain would you be willing to tolerate while still considering the therapy successful?" These were followed by the variables for "Duration of existing LBP" (more than eight days), the patient's educational level (low levels are related to higher risks of chronicity) and pain being experienced elsewhere in the body. Other significant factors were five items assessing depression (Zung) and the palliative effect of therapeutic massage (where a positive correlation was found). Female patients have a higher risk for chronicity, as do patients with a high total score on the scales assessing "catastrophizing thoughts" and thoughts of "helplessness". CONCLUSION: Using the items listed above, the study was able to predict a patient's risk of developing chronic LBP with a probability of 78%. These items were assembled in a brief questionnaire and were paired with a corresponding evaluative tool. This enables practitioners to assess an individual patient's risk for chronicity by means of a simple calculator in just a few minutes. A validation study for the questionnaire is currently being prepared. MINI ABSTRACT: The objective of this study was the development of a brief questionnaire to assess the risk for chronicity for LBP.

Acute Disease↗
PubMed2004

Preoperative neural network using combined magnetic resonance imaging variables, prostate specific antigen, and Gleason score to predict prostate cancer recurrence after radical prostatectomy.

OBJECTIVE: An artificial neural network analysis (ANNA) was developed to predict the biochemical recurrence more effectively than regression models based on the combined use of pelvic coil magnetic resonance imaging (pMRI), prostate specific antigen (PSA) and biopsy Gleason score in patients with clinically organ-confined prostate cancer after radical prostatectomy (RP). METHODS: Two-hundred-and-ten patients undergoing retropubic RP with pelvic lymphadenectomy were evaluated. Predictive study variables included clinical TNM classification, preoperative serum PSA, biopsy Gleason score, transrectal ultrasound (TRUS) findings, and pMRI findings. The predicted result was a biochemical failure (PSA >or=0.1 ng/ml). Using a five-way cross-validation method, the predicted ability of ANNA for a validation set of 200 randomly selected patients was compared with those of Cox regression analysis and "Kattan nomogram" by area under the receiver operating characteristic curve (AUC) analysis. RESULTS: Seventy-three patients (35%) failed at median follow-up of 61 (mean: 60, range: 2-94) months. Using similar input variables, the AUC of ANNA (0.765, 95% Confidence Interval [CI]: 0.704-0.825) was comparable (p > 0.05) to those for Cox regression (0.738, 95%CI: 0.691-0.819) and Kattan nomogram (0.728, 95%CI: 0.644-0.819). Contrarily, adding the pMRI findings, the ANNA is significantly (p < 0.05) superior to any other predictive model (0.897, 95%CI: 0.841-0.977). The Gleason score represented the most influential predictor (relative weight: 2.4) of PSA recurrence, followed by pMRI (2.2), and PSA (2.0). CONCLUSION: ANNA is superior to regression models to predict accurately biochemical recurrence. The relative importance of pMRI and the utility of ANNA to predict the PSA failure in patients referred for RP must be confirmed in further trials.

Biomarkers, Tumor↗
PubMed2004

Clinical prediction rule for atrial fibrillation after coronary artery bypass grafting.

OBJECTIVES: This study was designed to devise and validate a practical prediction rule for atrial fibrillation/atrial flutter (AF) after coronary artery bypass grafting (CABG) using easily available clinical and standard electrocardiographic (ECG) criteria. BACKGROUND: Reported prediction rules for postoperative AF have suffered from inconsistent results and controversy surrounding the added predictive value of a prolonged P-wave duration. METHODS: In 1,851 consecutive patients undergoing CABG with cardiopulmonary bypass, preoperative clinical characteristics and standard 12-lead ECG data were examined. Patients were continuously monitored for the occurrence of sustained postoperative AF while hospitalized. Multiple logistic regression was used to determine significant predictors of AF and to develop a prediction rule that was evaluated through jackknifing. RESULTS: Atrial fibrillation occurred in 508 of 1,553 patients (33%). Multivariate analysis showed that greater age (odds ratio [OR] 1.1 per year [95% confidence intervals (CI) 1.0 to 1.1], p < 0.0001), prior history of AF (OR 3.7 [95% CI 2.3 to 6.0], p < 0.0001), P-wave duration >110 ms (OR 1.3 [95% CI 1.1 to 1.7], p = 0.02), and postoperative low cardiac output (OR 3.0 [95% CI 1.7 to 5.2], p = 0.0001) were independently associated with AF risk. Using the prediction rule we defined three risk categories for AF: <60 points, 61 of 446 (14%); 60 to 79 points, 330 of 908 (36%); and >or=80 points, 117 of 199 (59%). The area under the receiver-operator characteristic curve for the model was 0.69. CONCLUSIONS: These data show that post-CABG AF can be predicted with moderate accuracy using easily available patient characteristics and may prove useful in prognostic and risk stratification of patients after CABG. The presence of intraatrial conduction delay on ECG contributed least to the prediction model.

Adrenergic beta-Antagonists↗
PubMed2006

Elevated pulmonary artery pressure by Doppler echocardiography predicts hospitalization for heart failure and mortality in ambulatory stable coronary artery disease: the Heart and Soul Study.

OBJECTIVES: We compared the predictive ability of tricuspid regurgitation (TR) and end-diastolic pulmonary regurgitation (EDPR) gradients in outpatients with coronary artery disease. BACKGROUND: The TR and EDPR gradients, in conjunction with right atrial pressure, provide Doppler estimates of pulmonary artery systolic and diastolic pressures. We hypothesized that increases in TR or EDPR gradients in stable coronary artery disease would predict heart failure (HF) hospitalization or cardiovascular (CV) death. METHODS: We measured TR and EDPR gradients in 717 adults with completed outcome adjudications who were recruited for the Heart and Soul Study. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated for HF hospitalization, CV death, all-cause death, and the combined end point. Multivariate adjustments were made for age, gender, race, history of CV or pulmonary disease, functional class, and left ventricular ejection fraction. RESULTS: There were 63 HF hospitalizations, 19 CV deaths, and 86 all-cause deaths at the 3-year follow-up. There were 466 measurable EDPR gradients and 573 measurable TR gradients. Age-adjusted ORs for EDPR >5 mm Hg predicted HF hospitalization (2.7, 95% CI 1.3 to 5.5, p = 0.006), all-cause death (2.5, 95% CI 1.4 to 4.4, p = 0.002), and HF hospitalization or CV death (2.7, 95% CI 1.4 to 5.2, p = 0.004). Age-adjusted OR for TR >30 mm Hg predicted HF hospitalization (3.4, 95% CI 1.9 to 6.2, p < 0.0001) and HF hospitalization or CV death (3.0, 95% CI 1.7 to 5.3, p = 0.0001). Multivariate adjusted OR per 5-mm Hg incremental increases in EDPR predicted HF hospitalization or CV death (1.9, 95% CI 1.01 to 3.6, p = 0.046) and all-cause death (1.7, 95% CI 1.05 to 2.8, p = 0.03). Multivariate adjusted OR per 10-mm Hg incremental increases in TR predicted HF hospitalization or CV death (1.6, 95% CI 1.1 to 2.4, p = 0.008). CONCLUSIONS: Increases in EDPR or TR gradients predict HF hospitalization or CV death among ambulatory adults with coronary artery disease.

Aged↗
PubMed2004

Molecular-based prediction of early recurrence in hepatocellular carcinoma.

BACKGROUND/AIMS: Hepatocellular carcinoma (HCC) has a very poor prognosis, due to the high incidence of tumor recurrence. As the current morphological indicators are often insufficient for therapeutic decisions, we sought to identify additional biologic indicators for early recurrence. METHODS: We analyzed gene expression using a PCR-based array of 3,072 genes in 100 HCC patients. Informative genes predicting early intrahepatic recurrence were selected by random permutation testing, and a weighted voting prediction method was constructed. Following estimation of prediction accuracy, a multivariate Cox analysis was performed. RESULTS: By permutation testing, we selected 92 genes demonstrated distinct expression patterns differing significantly between recurrence cases and recurrence-free cases. Our prediction method, using the 20 top-ranked genes, correctly predicted the early intrahepatic recurrence for 29 of 40 cases within the validation group, and the odds ratio was 6.8 (95%CI 1.7-27.5, P = 0.010). The 2-year recurrence rates in the patients with the good signature and those with the poor signature were 29.4 and 73.9%, respectively. Multivariate Cox analysis revealed that molecular-signature was an independent indicator for recurrence (hazard ratio 3.82, 95%CI 1.44-10.10, P = 0.007). CONCLUSIONS: Our molecular-based prediction method using 20 genes is clinically useful to predict early recurrence of HCC.

Aged↗
PubMed2006

Prediction of CD4(+) T cell epitopes restricted to HLA-DP4 molecules.

We have set up a method to predict peptide binding to HLA-DP4 molecules. These HLA II molecules are the most frequent worldwide and hence are an interesting target for epitope-based vaccines. The prediction is based on quantitative matrices built with binding data for peptides substituted at anchoring positions for HLA-DP4. A set of 98 peptides of various origins was used to compare the prediction with binding activity. At different prediction thresholds, the positive predictive value and the sensitivity of the prediction ranged from 50% to 80%, demonstrating its efficiency. This prediction method can be applied to the entire genomes of pathogens and large peptide sequences derived from tumor antigens.

Amino Acid Sequence↗
PubMed2006

Breath-hold single-photon emission tomography and computed tomography for predicting residual pulmonary function in patients with lung cancer.

OBJECTIVE: We sought to evaluate the utility of integrated breath-hold single-photon emission tomography and computed tomography imaging compared with that of simple calculation with the lung segment-counting technique for predicting residual pulmonary function in patients undergoing surgical intervention for lung cancer. METHODS: A prospective series of 22 patients undergoing anatomic lung resection for cancer were enrolled in this study. Postoperative residual forced expiratory volume in 1 second was predicted by measuring the radioactivity counts of the affected lobes or segments to be resected within the entire lungs by placement of regions of interest on single-photon emission tomography and computed tomography images. Residual forced expiratory volume in 1 second was also estimated by using the segment-counting technique. RESULTS: Both predicted values agreed well with postoperative forced expiratory volume in 1 second. Although the residual forced expiratory volume in 1 second predicted by means of single-photon emission tomography and computed tomography correlated well with that predicted by using segment counting, the values were significantly underestimated by the segment-counting technique in 4 outliers with severe emphysema. There were 2 patients with borderline pulmonary functional reserve whose residual forced expiratory volume in 1 second values were predicted more accurately by means of single-photon emission tomography and computed tomography than by using segment counting. CONCLUSION: Integrated breath-hold single-photon emission tomography and computed tomography images allow the accurate prediction of postoperative pulmonary function but without statistical superiority over the simple segment-counting technique. Further study of the usefulness of single-photon emission tomography and computed tomography in patients with severe emphysema and borderline lung function should prove valuable because the segment-counting technique underestimates pulmonary functional reserve in these patients.

Aged↗
PubMed2006

Deficits in predictive smooth pursuit after mild traumatic brain injury.

Given that even mild traumatic brain injury (TBI) may produce extensive diffuse axonal injury (DAI), we hypothesized that mild TBI patients would show deficits in predictive smooth pursuit eye movements (SPEM), associated with impaired cognitive functions, as these processes are dependent on common white matter connectivity between multiple cerebral and cerebellar regions. The ability to predict target trajectories during SPEM was investigated in 21 mild TBI patients using a periodic sinusoidal paradigm. Compared to 26 control subjects, TBI patients demonstrated decreased target prediction. TBI patients also showed increased eye position error and variability of eye position, which correlated with decreased target prediction. In all subjects, average target prediction, eye position error and eye position variability correlated with scores related to attention and executive function on the California Verbal Learning Test (CVLT-II). However, there were no differences between TBI and control groups in average eye gain or intra-individual eye gain variability, or in performance on the Wechsler Abbreviated Scale of Intelligence (WASI), suggesting that the observed deficits did not result from general oculomotor impairment or reduced IQ. The correlation between SPEM performance and CVLT-II scores suggests that predictive SPEM may be a sensitive assay of cognitive functioning, including attention and executive function. This is the first report to our knowledge that TBI patients show impaired predictive SPEM and eye position variability, and that these impairments correlate with cognitive deficits.

Adolescent↗
PubMed2004

Protein C levels can be forecasted by global haemostatic tests in critically ill patients and predict long-term survival.

INTRODUCTION: We have shown the usefulness of global haemostatic tests International Normalized Ratio (INR) and Activated Partial Thromboplastin Time (APTT) for predicting survival in critically ill patients. Ability to analyse inhibitors protein C and antithrombin is limited to a small number of laboratories and often only during office hours. We therefore studied the usefulness of global haemostatic tests to predict levels of protein C and antithrombin and investigated value of these latter tests in predicting outcome. PATIENTS/METHODS: Blood samples were collected within 6 h of admission to intensive care unit (ICU) and tested regarding platelet count, INR, and APTT. If platelet count was <100x10(9) L(-1), INR >1.36 and/or APTT >45 s, a second sampling was done within 6 h after the first one for analysis of protein C and antithrombin. Ninety-two patients were included; length of stay at ICU and hospital, survival when leaving ICU and hospital and up to 5 years were recorded. RESULTS: Using univariate analysis of variance, INR and APTT separately predicted levels of protein C and to some extent antithrombin. Neither platelet count nor any combinations of global haemostatic tests were predictive. Utilising Cox regression, decreased protein C, but not antithrombin, predicted lower survival rate. CONCLUSIONS: Global haemostatic tests INR and APTT can predict levels of protein C and, though less so, antithrombin. A low protein C level indicated a sinister prognosis in the ICU setting, at the hospital, and after up to 5 years.

Aged↗
PubMed2005

Effect of patient location on the performance of clinical models to predict pulmonary embolism.

INTRODUCTION: Current clinical likelihood models for predicting pulmonary embolism (PE) are used to categorize outpatients into low, intermediate and high clinical pre-test likelihood of PE. Since these clinical prediction rules were developed using outpatients it is not known if they can be applied universally to both inpatients and outpatients with suspected PE. Thus, the purpose of this study was to determine the effect of patient location on the performance of clinical models to predict PE. MATERIALS AND METHODS: Two clinical models (Wells and Wicki) were applied to data from the multi-centered PIOPED study. The Wells score was applied to 1359 patients and the Wicki score was applied to 998 patients. 361 patients (27%) from the PIOPED study did not have arterial gas measurement and were excluded from the Wicki score patient group. Patients were stratified by their location at the time of entry into the PIOPED study as follows: outpatient/emergency, surgical ward, medicine/coronary care unit or intensive care unit. The diagnostic performance of the two clinical models was applied to the various patient locations and the performance was evaluated using the area under a fitted receiver operating characteristic curve (AUC). RESULTS: The prevalence of PE in the three clinical probability categories were similar for the two scoring methods. Both clinical models yielded the lowest diagnostic performance in patients referred from surgical wards. The AUC for both clinical prediction rules decreased significantly when applied to inpatients in comparison to outpatients. CONCLUSIONS: Current clinical prediction rules for determining the pre-test likelihood of PE yielded different diagnostic performances depending upon patient location. The performance of the clinical prediction rules decreased significantly when applied to inpatients. In particular, the rules performed least well when applied to patients referred from surgical wards suggesting these rules should not be used in this patient group. As expected the clinical prediction rules performed best in outpatients with the optimum diagnostic performance in patients referred from emergency and outpatient wards.

Adult↗
PubMed2004

Prediction of CTL epitopes using QM, SVM and ANN techniques.

Cytotoxic T lymphocyte (CTL) epitopes are potential candidates for subunit vaccine design for various diseases. Most of the existing T cell epitope prediction methods are indirect methods that predict MHC class I binders instead of CTL epitopes. In this study, a systematic attempt has been made to develop a direct method for predicting CTL epitopes from an antigenic sequence. This method is based on quantitative matrix (QM) and machine learning techniques such as Support Vector Machine (SVM) and Artificial Neural Network (ANN). This method has been trained and tested on non-redundant dataset of T cell epitopes and non-epitopes that includes 1137 experimentally proven MHC class I restricted T cell epitopes. The accuracy of QM-, ANN- and SVM-based methods was 70.0, 72.2 and 75.2%, respectively. The performance of these methods has been evaluated through Leave One Out Cross-Validation (LOOCV) at a cutoff score where sensitivity and specificity was nearly equal. Finally, both machine-learning methods were used for consensus and combined prediction of CTL epitopes. The performances of these methods were evaluated on blind dataset where machine learning-based methods perform better than QM-based method. We also demonstrated through subgroup analysis that our methods can discriminate between T-cell epitopes and MHC binders (non-epitopes). In brief this method allows prediction of CTL epitopes using QM, SVM, ANN approaches. The method also facilitates prediction of MHC restriction in predicted T cell epitopes.

Algorithms↗
PubMed1997

Validation of a clinical prediction rule for left ventricular ejection fraction after myocardial infarction in patients > or = 65 years old.

We sought to validate a previously described clinical prediction rule for classifying left ventricular ejection fraction (LVEF) after acute myocardial infarction (AMI). As part of the Connecticut cohort of the Cooperative Cardiovascular Project (CCP) pilot study, we identified 3,093 Medicare patients who had been admitted to hospitals throughout Connecticut with an AMI in 1992 and 1993. Retrospective chart review and detailed electrocardiogram interpretation were performed. Of the 1,891 patients with an interpretable EF, 1,378 (73%) had > or = 1 of the rule's exclusion criteria. Of the remaining 513 patients, the clinical prediction rule had a positive predictive value of 89% (i.e., 456 of 513 patients had an EF > or = 40%). In a multivariate model, presentation > 6 hours after the onset of chest pain, a history of bypass surgery, and diabetes mellitus were associated with patients in whom the rule did not correctly predict an EF > or = 40%. Excluding patients with these characteristics from the rule increased the positive predictive value from 89% to 93% and excluded an additional 239 patients. The EF could not be predicted among the patients who did not meet the rule's criteria. In conclusion, a previously published clinical prediction rule for the classification of the EF in patients after an AMI correctly classified 8 of every 9 eligible elderly patients as having an EF > or = 40%. Thus, while not performing as well as it did in the original study, our findings support the use of this rule in providing clinicians with an objective method for estimating an EF > or = 40% in a specific subset of elderly patients.

Aged↗
PubMed2004

Positron emission tomography scanning poorly predicts response to preoperative chemotherapy in non-small cell lung cancer.

BACKGROUND: The ability to accurately predict pathologic response to preoperative chemotherapy may have a significant impact on the treatment strategy for non-small cell lung cancer (NSCLC). The purpose of this study was to examine the accuracy of positron emission tomography (PET) scanning in predicting the pathologic response to preoperative chemotherapy in the primary tumor and draining lymph nodes. METHODS: A total of 25 patients were enrolled in two separate phase II trials investigating induction chemotherapy for NSCLC. All patients underwent pre-treatment and post-treatment PET scans followed by surgical resection. A significant PET scan response was defined as a reduction in the standard uptake value by 50% or more. We defined a major pathologic response as either no disease or microscopic disease only in the primary tumor. The percentage change in standard uptake value was then calculated and correlated with pathologic response in the primary tumor. In addition, the presence or absence of nodal metastases as determined by the postchemotherapy PET scan was compared with final pathologic nodal stage. RESULTS: The positive and negative predictive values for PET detection of major pathologic response in the primary tumor were 43% and 100%, respectively. Positron emission tomography did not accurately predict nodal status in 52% of patients. The positive and negative predictive values of PET to detect node-positive disease were 73% and 64%, respectively. For N2 disease the positive predictive value of PET scans was less than 20%. CONCLUSIONS: Positron emission tomography scanning does not reliably predict pathologic response to preoperative chemotherapy in NSCLC in either the primary tumor or the draining lymph nodes.

Carcinoma, Non-Small-Cell Lung↗