SEARCH · PubMed Health

Results for “predictability”

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 73 records · Page 4Linked to original sources

PubMed1996

Assessment of protein fold predictions from sequence information: the predicted alpha/beta doubly wound fold of the von Willebrand factor type A domain is similar to its crystal structure.

The fold of the von Willebrand Factor type A domain (vWF-A) was predicted to be similar to an alpha/beta doubly wound fold in the GTP-binding domain of ras-p21, despite the lack of sequence or functional similarity. This was subsequently confirmed by the vWF-A crystal structure from complement receptor type 3. The prediction is now reviewed. The vWF-A secondary structure was predicted with 62 to 75% accuracy and 12 of the 13 secondary structure elements were identified correctly. Accessibility predictions were 69 to 71% accurate. The fold recognition analysis was confirmed, but was much improved by averaging the results from 70 complete vWF-A sequences. The related folds of ras-P21 and flavodoxin scored highly. In addition, both the mapping of the predicted vWF-A secondary structure elements with those in 12 known alpha/beta folds and two Asp residues at the C-terminal ends of two adjacent beta-strands matched well with ras-p21 and flavodoxin. The predicted Mg(2+)-binding site, two disulphide bridges and the secondary structure topology were largely accurate. The exception is the reversal of a beta-hairpin at one end of the central beta-sheet. We conclude that non-homologous folds with dissimilar functions can be predicted from sequence data with reasonable accuracy, and that the accuracy in this case was principally limited at the periphery of the fold.

Amino Acid Sequence↗

PubMed2004

Controlled test for predictive power of Lyapunov exponents: their inability to predict epileptic seizures.

Lyapunov exponents are a set of fundamental dynamical invariants characterizing a system's sensitive dependence on initial conditions. For more than a decade, it has been claimed that the exponents computed from electroencephalogram (EEG) or electrocorticogram (ECoG) signals can be used for prediction of epileptic seizures minutes or even tens of minutes in advance. The purpose of this paper is to examine the predictive power of Lyapunov exponents. Three approaches are employed. (1) We present qualitative arguments suggesting that the Lyapunov exponents generally are not useful for seizure prediction. (2) We construct a two-dimensional, nonstationary chaotic map with a parameter slowly varying in a range containing a crisis, and test whether this critical event can be predicted by monitoring the evolution of finite-time Lyapunov exponents. This can thus be regarded as a "control test" for the claimed predictive power of the exponents for seizure. We find that two major obstacles arise in this application: statistical fluctuations of the Lyapunov exponents due to finite time computation and noise from the time series. We show that increasing the amount of data in a moving window will not improve the exponents' detective power for characteristic system changes, and that the presence of small noise can ruin completely the predictive power of the exponents. (3) We report negative results obtained from ECoG signals recorded from patients with epilepsy. All these indicate firmly that, the use of Lyapunov exponents for seizure prediction is practically impossible as the brain dynamical system generating the ECoG signals is more complicated than low-dimensional chaotic systems, and is noisy.

Cerebral Cortex↗

PubMed2002

Fold prediction of helical proteins using torsion angle dynamics and predicted restraints.

We describe a procedure for predicting the tertiary folds of alpha-helical proteins from their primary sequences. The central component of the procedure is a method for predicting interhelical contacts that is based on a helix-packing model. Instead of predicting the individual contacts, our method attempts to identify the entire patch of contacts that involve residues regularly spaced in the sequences. We use this component to glue together two powerful existing methods: a secondary structure prediction program, whose output serves as the input to the contact prediction algorithm, and the tortion angle dynamics program, which uses the predicted tertiary contacts and secondary structural states to assemble three-dimensional structures. In the final step, the procedure uses the initial set of simulated structures to refine the predicted contacts for a new round of structure calculation. When tested against 24 small to medium-sized proteins representing a wide range of helical folds, the completely automated procedure is able to generate native-like models within a limited number of trials consistently.

Algorithms↗

PubMed2004

Predicted 3-D structures for mouse I7 and rat I7 olfactory receptors and comparison of predicted odor recognition profiles with experiment.

The first step in the perception of an odor is the activation of one or more olfactory receptors (ORs) following binding of the odorant molecule to the OR. In order to initiate the process of determining how the molecular level receptor-odorant interactions are related to odor perception, we used the MembStruk computational method to predict the three-dimensional (3-D) structure of the I7 OR for both mouse and rat. We then used the HierDock ligand docking computational method to predict the binding site and binding energy for the library of 56 odorants to these receptors for which experiment response data are now available. We find that the predicted 3-D structures of the mouse and rat I7 OR lead to predictions of odorant binding that are in good agreement with the experimental results, thus validating the accuracy of both the 3-D structure and the predicted binding site. In particular we predict that heptanal and octanal both bind strongly to both mouse and rat I7 ORs, which conflicts with the older literature but agrees with recent experiments. To provide the basis of additional validations of our 3-D structures, we also report the odorant binding site for a new odorant (8-hydroxy-octanal) with a novel functionality designed to bind strongly to mouse I7. Such validated computational methods should be very useful in predicting the structure and function of many other ORs.

Amino Acid Sequence↗

PubMed1994

Presenting pulse pressure predicts thrombolytic therapy-related intracranial hemorrhage. Thrombolytic Predictive Instrument (TPI) Project results.

BACKGROUND: In selecting patients with acute myocardial infarction for thrombolytic therapy, it is important to identify patients who are at high risk for intracranial hemorrhage, for whom thrombolytic therapy is ill advised. We hypothesized that presenting pulse blood pressure, representing the "hammer" effect on cerebral vessels and the effects of age on arterial compliance, might predict thrombolysis-related intracranial hemorrhage better than systolic, diastolic, or mean arterial blood pressures. METHODS AND RESULTS: Of 3483 Thrombolytic Predictive Instrument (TPI) Project subjects receiving thrombolytic therapy for acute infarction, we identified and obtained detailed clinical data on the 19 with treatment-related intracranial hemorrhages confirmed by computed tomography and on 175 matched controls. Systolic, diastolic, mean arterial, and pulse blood pressures were each significantly related to the occurrence of intracranial hemorrhage, with pulse pressure most highly related. The mean pulse pressure in patients who developed intracranial hemorrhage was 63 mm Hg, 34% higher than the 47 mm Hg mean value for those not developing hemorrhage (P = .0001). Excess pulse pressure, defined as the extent to which a patient's pulse pressure exceeded 40 mm Hg for systolic blood pressures of at least 120 mm Hg, was even more strongly related: its mean value of 23 mm Hg for patients was 130% higher than its mean value of 10 mm Hg for controls (P < .0001). With logistic regression models to estimate the relative risks (odds ratios) for intracranial hemorrhage conferred by each form of blood pressure, the relative risk for hemorrhage was greatest for excess pulse pressure: for each 10-point pulse pressure excess, the relative risk for intracranial hemorrhage was increased by 1.85 (P = .0002; 95% confidence interval [CI], 1.34 to 2.55) by itself and 1.76 (P = .001; 95% CI, 1.26 to 2.46) when adjusted for age. In this sample, excess pulse pressure by itself predicted hemorrhage as well as systolic pressure and age together. When excess pulse pressure was combined with age to make a logistic regression model predicting intracranial hemorrhage, age contributed less to the prediction than when combined with the other blood pressure forms, even though this model predicted better than any other combination of age and pressure (receiver-operating characteristic curve area, 0.82 versus 0.77 with systolic pressure and age, 0.75 with mean arterial pressure, 0.71 with diastolic pressure, and 0.81 with both systolic and diastolic pressures). CONCLUSIONS: We found that excess pulse blood pressure predicted thrombolysis-related intracranial hemorrhage better than other forms of pretreatment blood pressure, perhaps better describing the pathophysiology of intracranial hemorrhage, including the effect of age. These findings will need confirmation in larger studies with comparable clinical detail.

Aged↗

PubMed1993

Prediction of lung function in Hispanics using local ethnic-specific and external non-ethnic-specific prediction equations.

We compared locally derived prediction equations for the FVC and FEV1 in Hispanics with external prediction equations derived from non-Hispanic whites. Algebraic subtraction of the external non-Hispanic equations from the New Mexico Hispanic equations showed that the differences would vary with age and height. The study population included 442 patients with Hispanic surnames between the ages of 25 and 80 yr evaluated at the University of New Mexico Hospital pulmonary function laboratory. We calculated percent predicted values for FVC and FEV1 using spirometric prediction equations from a New Mexico Hispanic population and from non-Hispanic white populations in Salt Lake City, Tucson, and six other U.S. cities. We employed either 80% of predicted or the lower fifth percentile as the criterion for separating normal from abnormal values. We found that the concordance of the classifications by the external and internal equations varied among the external equations and with the criterion used for abnormality. The classification of the FVC and FEV1 as normal or abnormal was influenced by the regression equation in about 5 to 10% of the subjects. Subjects with discordant classifications were not clearly predicted by age or height, although they tended to be at the extremes of the age and height distributions. This study shows that classification of lung function by locally derived ethnic-group-specific prediction equations may lead to a differing clinical categorization for some persons in comparison with external non-ethnic-group-specific equations.

Adult↗

PubMed2001

Can we predict radiation-induced changes in pulmonary function based on the sum of predicted regional dysfunction?

PURPOSE: To determine whether changes in whole-lung pulmonary function test (PFT) values are related to the sum of predicted radiation therapy (RT)-induced changes in regional lung perfusion. PATIENTS AND METHODS: Between 1991 and 1998, 96 patients (61% with lung cancer) who were receiving incidental partial lung irradiation were studied prospectively. The patients were assessed with pre- and post-RT PFTs (forced expiratory volume in one second [FEV1] and diffusion capacity for carbon monoxide [DLCO]) for at least a 6-month follow-up period, and patients were excluded if it was determined that intrathoracic recurrence had an impact on lung function. The maximal declines in PFT values were noted. A dose-response model based on RT-induced reduction in regional perfusion (function) was used to predict regional dysfunction. The predicted decline in pulmonary function was calculated as the weighted sum of the predicted regional injuries: equation [see text] where Vd is the volume of lung irradiated to dose d, and Rd is the reduction in regional perfusion anticipated at dose d. RESULTS: The relationship between the predicted and measured reduction in PFT values was significant for uncorrected DLCO (P = .005) and borderline significant for DLCO (P = .06) and FEV1 (P = .08). However, the correlation coefficients were small (range,.18 to.30). In patients with lung cancer, the correlation coefficients improved as the number of follow-up evaluations increased (range,.43 to.60), especially when patients with hypoperfusion in the lung adjacent to a central mediastinal/hilar thoracic mass were excluded (range,.59 to.91). CONCLUSION: The sum of predicted RT-induced changes in regional perfusion is related to RT-induced changes in pulmonary function. In many patients, however, the percentage of variation explained is small, which renders accurate predictions difficult.

Adult↗

PubMed2001

Prediction of hepatic metabolic clearance: comparison and assessment of prediction models.

OBJECTIVE: To perform a comparative quantitative evaluation of the prediction accuracy for human hepatic metabolic clearance of 5 different mathematical models: allometric scaling (multiple species and rat only), physiologically based direct scaling, empirical in vitro-in vivo correlation, and supervised artificial neural networks. METHODS: The mathematical prediction models were implemented with a publicly available dataset of 22 extensively metabolised compounds and compared for their prediction accuracy using 3 quality indicators: prediction error sum of squares (PRESS), r2 and the fold-error. RESULTS: Approaches such as physiologically based direct scaling, empirical in vitro-in vivo correlation and artificial neural networks, which are based on in vitro data only, yielded an average fold-error ranging from 1.64 to 2.03 and r2 values greater than 0.77, as opposed to r2 values smaller than 0.44 when using allometric scaling combining in vivo and in vitro preclinical data. The percentage of successful predictions (less than 2-fold error) ranged from 55% (rat allometric scaling) to between 64 and 68% with the other approaches. CONCLUSIONS: On the basis of a diverse set of 22 metabolised drug molecules, these studies showed that the most cost-effective and accurate approaches, such as physiologically based direct scaling and empirical in vitro-in vivo correlation, are based on in vitro data alone. Inclusion of in vivo preclinical data did not significantly improve prediction accuracy; the prediction accuracy of the allometric approaches was at the lower end of all methods compared.

Animals↗

PubMed2006

Prediction of visceral fat area in Japanese adults: proposal of prediction method applicable in a field setting.

OBJECTIVE: This study aimed to develop a prediction equation for the visceral fat area at the umbilical level (VFA(L4-5)) in Japanese adults, using internal fat mass (IFM) estimated from a few anthropometric variables. METHODS: Subjects were 112 adults aged from 25 to 82 years (body mass index (BMI)=24.2+/-3.1 kg/m(2), ranged from 15.7 to 31.2 kg/m(2)). Another 60 adults aged from 21 to 71 years were recruited for the crossvalidation group (BMI=24.5+/-4.0 kg/m(2), ranged from 17.1 to 34.6 kg/m(2)). We examined (1) the prediction of IFM based on a small number of skinfold thicknesses; (2) the prediction of VFA(L4-5) using IFM and (3) the application of bioelectrical impedance analysis (BIA) measurement. VFA(L4-5) was measured by computed tomography (reference value). Total fat mass was measured by dual-energy X-ray absorptiometry (DXA) and single-frequency BIA with 8-point tactile electrodes. RESULTS: Three skinfolds at the abdomen, side chest and suprailiac were selected to estimate IFM. From IFM estimated using these three skinfolds, waist-to-hip ratio (WHR), sex and age, about 75% of the variance of VFA(L4-5) could be explained (Eq(VFA)1: R=863, R(2)=0.745, standard error of estimate (s.e.e.)=20.483 cm(2)). When substituting IFM based on BIA measurement (IFM(BIA)) into Eq(VFA)1, there were no significant mean differences from the reference in both equations, and high correlations were found (r=0.860, s.e.e.=20.902 cm(2)), although a significant mean difference in total fat mass was found between DXA and BIA measurements. The prediction equation using IFM(BIA) (Eq(VFA)2) could have prediction accuracy comparable with that of Eq(VFA)1 (Eq(VFA)2: R=879, R(2)=0.773, s.e.e.=20.324 cm(2)). Furthermore, when applying these equations to the crossvalidation group, there were cross-validity in both equations. CONCLUSION: This study proposed a prediction equation for VFA(L4-5) from WHR and IFM based on three skinfolds, and the validity of BIA measurement in Japanese adults. We can propose the procedure for a field setting.

Absorptiometry, Photon↗

PubMed2003

Application of the CFU-GM assay to predict acute drug-induced neutropenia: an international blind trial to validate a prediction model for the maximum tolerated dose (MTD) of myelosuppressive xenobiotics.

In a previous study of prevalidation, a standard operating procedure (SOP) for two independent in vitro tests (human and mouse) had been developed, to evaluate the potential hematotoxicity of xenobiotics from their direct and the adverse effects on granulocyte-macrophages (CFU-GM). A predictive model to calculate the human maximum tolerated dose (MTD) was set up, by adjusting a mouse-derived MTD for the differential interspecies sensitivity. In this paper, we describe an international blind trial designed to apply this model to the clinical neutropenia, by testing 20 drugs, including 14 antineoplastics (Cytosar-U, 5-Fluorouracil, Myleran, Thioguanine, Fludarabine, Bleomycin, Methotrexate, Gemcitabine, Carmustine, Etoposide, Teniposide, Cytoxan, Taxol, Adriamycin); two antivirals (Retrovir, Zovirax,); three drugs for other therapeutic indications (Cyclosporin, Thorazine, Indocin); and one pesticide (Lindane). The results confirmed that the SOP developed generates reproducible IC90 values with both human and murine GM-CFU. For 10 drugs (Adriamycin, Bleomycin, Etoposide, Fludarabine, 5-Fluorouracil, Myleran, Taxol, Teniposide, Thioguanine, and Thorazine), IC90 values were found within the range of the actual drug doses tested (defined as the actual IC90). For the other 10 drugs (Carmustine, Cyclosporin, Cytosar-U, Cytoxan, Gemcitabine, Indocin, Lindane, Methotrexate, Retrovir, and Zovirax) extrapolation on the regression curve out of the range of the actual doses tested was required to derive IC90 values (extrapolated IC90). The model correctly predicted the human MTD for 10 drugs out of 10 that had "actual IC90 values" and 7 drugs out of 10 for those having only an extrapolated IC90. Two of the incorrect predictions (Gemcitabine and Zovirax) were within 6-fold of the correct MTD, instead of the 4-fold range required by the model, whereas the prediction with Cytosar-U was approximately 10-fold in error. A possible explanation for the failure in the prediction of these three drugs, which are pyrimidine analogs, is discussed. We concluded that our model correctly predicted the human MTD for 20 drugs out of 23, since the other three drugs (Topotecan, PZA, and Flavopiridol) were tested in the prevalidation study. The high percentage of predicitivity (87%), as well as the reproducibility of the SOP testing, confirm that the model can be considered scientifically validated in this study, suggesting promising applications to other areas of research in developing validated hematotoxicological in vitro methods.

Acute Disease↗

PubMed2001

Predictive genetic testing in children.

Predictive genetic testing should only be performed on children if it is in their best interests. "Interests" include psychosocial elements. Predictive testing is performed on children when there are interventions to prevent disease or to detect and treat it early and it is necessary to begin these interventions in childhood. It is also performed for diseases known to commence in childhood. Predictive testing in children for adult-onset conditions for which there is no medical intervention is highly controversial. Competent children and adolescents can consent to predictive genetic testing. Predictive testing can result in harm, such as discrimination (eg, in insurance entitlement or employment) and stigmatisation. Predictive testing can have important non-medical benefits in terms of self-knowledge and life planning.

Australia↗

PubMed2006

Chemosensitivity prediction in esophageal squamous cell carcinoma: novel marker genes and efficacy-prediction formulae using their expression data.

Esophageal cancer is a highly lethal disease and the optimal therapy remains unclear. Since adjuvant chemotherapy gives a better chance of survival, we attempted to develop a chemosensitivity prediction model to improve individual responses to therapy. Comprehensive gene expression analyses (cDNA and oligonucleotide microarrays) and MTT assay of 8 drugs in 20 KYSE squamous cell carcinoma cell lines were performed to distinguish candidate marker genes whose expression levels reproducibly correlated with cellular drug sensitivities. After confirmation with real-time RT-PCR, we performed multiple regression analyses to develop drug-sensitivity prediction formulae using the quantified expression data of selected marker genes. Using the same sets of genes, we also constructed prediction models for individual clinical responses to 5-FU-based chemotherapy using 18 cases. We selected 5 better marker genes, known as drug sensitivity determinants, identified 9 novel predictive genes for 4 of 8 anticancer drugs [5-FU, CDDP, DOX, and CPT-11 (SN-38)], and developed highly predictive formulae of in vitro sensitivities to the 4 drugs and clinical responses to 5-FU-based adjuvant chemotherapies in terms of overall and disease-free survivals. Our selected genes are likely to be effective drug-sensitivity markers and formulae using the 9 novel genes would provide advantages in prediction.

Antineoplastic Agents↗

PubMed1985

Prediction of functional capacity and use of exercise testing for predicting risk after acute myocardial infarction.

This study evaluated whether an ischemic exercise test response or functional capacity could be predicted from data available during hospitalization in patients discharged after acute myocardial infarction (AMI). The value of exercise test variables for predicting death and new AMI within 1 year was also examined. Among 1,469 patients, 466 (32%) underwent treadmill exercise testing around the time of discharge. An ischemic exercise test response (ST-segment depression or angina) could not be predicted. Good functional capacity (more than 4 METs) could be predicted from age and ST-segment changes at rest. Among the 60% of the patients who were predicted to have functional capacity of more than 4 METs, only 15% had poor functional capacity at the time of testing. Multivariate analysis for predicting death and new infarction selected only functional capacity (continuous variable in METs), which classified 72% of the patients into a low-risk group with less than a 2% rate of death and new AMI in the first year. The high-risk group (29% of the patients) had an 18% rate of death or new AMI. It is concluded that functional capacity is the most important exercise test variable and that patients likely to have good functional capacity can be identified on the basis of age and ST-segment changes at rest. Further, the level of functional capacity on exercise testing can identify groups of patients with very low and relatively high risk of death or new AMI within 1 year.

Adrenergic beta-Antagonists↗

PubMed1993

Prediction of pain in patients with chronic low back pain: effects of inaccurate prediction and pain-related anxiety.

This study investigated predictions of pain intensity, reports of pain and anxiety, frequency of pain-related anxiety symptoms, and range of motion, in 43 patients exposed to pain during a physical examination. All patients had primary complaints of low back pain. The pain stimuli used for this study included back and/or leg pain produced by repeatedly raising the extended leg of the patient to the point of pain tolerance. Generally, findings demonstrated that (a) predictions of pain were a function of discrepancies between previous predictions and experiences of pain, (b) patients reporting greater pain-related anxiety showed a tendency to overpredict new pain events, but corrected their predictions readily, (c) patients reporting less pain-related anxiety displayed a persistent tendency to underpredict pain, and (d) higher predictions of pain, independent of pain reports, related to less range of motion during a procedure that involved painful movement. Discussion focuses on differences between these results and those of previous studies and the implications of inaccurate prediction for continued pain and disability.

Adult↗

PubMed1994

Comparative predictive value of attributional style, negative affect, and positive affect in predicting self-reported physical health and psychological health.

The main aim of the study was to determine the comparative predictive value of affect and attributions in predicting emotional and physical health. Secondary aims were to determine the comparative value of attributional, affect variables, in predicting health. Two hundred and forty subjects completed scales for the assessment of attributions, and negative and positive affect. Subjects also answered self-report questions on emotional health, physical health, number of visits to doctors for medical advice, illness, and days absent from work. The analysis of relative importance of attributional and affect variables in predicting health revealed that negative affect caused by thoughts was the best predictor of psychological health and physical health; there was some indication that positive affect caused by thoughts was the next best predictor of health. Analysis of the predictive value of attributional variables only, revealed that: (a) attributional style for bad events was a better predictor of health than attributional style for good events; and (b) of all the attributions made for good events and bad events, global attributions for bad events were the best predictor of health. Analysis of the predictive value of affect variables only showed that: (a) negative affect was a better predictor of health than positive affect; and (b) of negative affect and positive affect caused by thoughts and day-to-day experiences, negative affect caused by thoughts was the best predictor of health.

Adolescent↗

PubMed2006

Prediction of asthma in young adults using childhood characteristics: Development of a prediction rule.

OBJECTIVE: To develop an easily applicable prediction rule for asthma in young adulthood using childhood characteristics. METHODS: A total of 1,055 out of 1,328 members of a Dutch birth cohort were followed from 2 to 21 years of age. Univariate and multivariate logistic regression analyses were used to evaluate the predictive value of childhood characteristics on asthma at 21 years of age. A prognostic function was developed, and the area under the receiving operating characteristic (ROC) curve was used to estimate the predictive ability of the prognostic models. RESULTS: Of the 693 responding subjects, 86 (12%) were diagnosed with asthma. Independent prognostic factors at ages 2 and 4 years were female gender (odds ratios (OR) 1.9 and 2.1; 95% confidence intervals (CI) 1.2-3.2 and 1.3-2.5), smoking mother (OR 1.6 and 1.6; CI 1.0-2.7 and 1.0-2.6), lower respiratory tract illness (OR 1.9 and 2.4; CI 1.0-3.6 and 1.4-4.0), and atopic parents (OR 2.1 and 1.9; CI 1.3-3.4 and 1.2-3.1). The predictive power of both models was poor; area under ROC curve was 0.66 and 0.68, respectively. CONCLUSION: Asthma in young adulthood could not be predicted satisfactorily based on childhood characteristics. Nevertheless, we propose that this method is further tested as a tool to predict development of asthma.

Adolescent↗

PubMed2006

Predictability and predictiveness in health care spending.

This paper re-examines the relation between the predictability of health care spending and incentives due to adverse selection. Within an explicit model of health plan decisions about service levels, we show that predictability (how well spending on certain services can be anticipated), predictiveness (how well the predicted levels of certain services contemporaneously co-vary with total health care spending), and demand responsiveness all matter for adverse selection incentives. The product of terms involving these three measures of predictability, predictiveness, and demand responsiveness define an empirical index of the direction and magnitude of selection incentives. We quantify the relative magnitude of adverse selection incentives bearing on various types of health care services in Medicare. Our results are consistent with other research on service-level selection. The index of incentives can readily be applied to data from other payers.

Forecasting↗

PubMed2003

The seizure prediction characteristic: a general framework to assess and compare seizure prediction methods.

The unpredictability of seizures is a central problem for all patients suffering from uncontrolled epilepsy. Recently, numerous methods have been suggested that claim to predict from the EEG the onset of epileptic seizures. In parallel, new therapeutic devices are in development that could control upcoming seizures provided that their onset is known in advance. A reliable clinical application controlling seizures, consisting of a seizure prediction method and an intervention system, would improve patient quality of life. The question therefore arises as to whether the performance of the seizure prediction methods is already sufficient for clinical applications. The answer requires assessment criteria to judge and compare these methods, but recognized criteria still do not exist. Based on clinical, behavioral, and statistical considerations, we suggest the "seizure prediction characteristic" to evaluate seizure prediction methods. Results of this approach are exemplified by its application to the "dynamical similarity index" seizure prediction method using 582 hours of intracranial EEG data, including 88 seizures.

Electroencephalography↗