Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “External validity”

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 1,801 records · Page 100Linked to original sources
PubMed2003

Rapid determination of lidocaine solutions with non-column chromatographic diode array UV spectroscopy and multivariate calibration.

A new method for the rapid determination of pharmaceutical solutions is proposed. A conventional HPLC system with a Diode Array Detector (DAD) was used with no chromatographic column connected. As eluent, purified water (Milli Q) was used. The pump and autosampler of the HPLC system were mainly utilised as an automatic and convenient way of introducing the sample into the DAD. The method was tested on the local anaesthetic compound lidocaine. The UV spectrum (245-290 nm) from the samples analysed in the detector was used for multivariate calibration for the determination of lidocaine solutions. The content was determined with PLS regression. The effect on the predictive ability of three factors: flow, data-collection rate and rise time as well as two ways of exporting a representative UV spectrum from the DAD file collected was investigated by means of an experimental design comprising 11 experiments. For each experiment, 14 solutions containing a known content of lidocaine were analysed (0.02-0.2 mg ml(-1)). From these 14 samples two calibration sets and two test sets were made and as the response in the experimental design the Root Mean Square Error of Prediction (RMSEP) values from the predictions of the two test sets were used. When the factor setting giving the lowest RMSEP was found, this setting was used when analysing a new calibration set of 12 lidocaine samples (0.1-0.2 mg ml(-1)). This calibration model was validated by two external test sets, A and B, analysed on separate occasions for the evaluation of repeatability (test set A) and determination over time (test set B). For comparison, the reference method, liquid chromatography, was also used for analysis of the ten samples in test set B. This comparison of the two methods was done twice on different occasions. The results show that in respect of accuracy, precision and repeatability the new method is comparable to the reference method. The main advantages compared with liquid chromatography are the much shorter time of analysis (<30 s) as well as the automatic and simple analytical procedure and the low consumption of organic solvents.

Calibration↗
PubMed2005

Molecular modeling and 3D-QSAR studies of indolomorphinan derivatives as kappa opioid antagonists.

Molecular modeling and 3D-QSAR studies were performed on 31 indolomorphinan derivatives to evaluate their antagonistic behaviors on kappa opioid receptor and provide information for further modification of this kind of compounds. Best predictions were obtained with CoMFA standard model (q2 = 0.693, N = 4, r2 = 0.900) and CoMSIA combined model (q2 = 0.617, N = 4, r2 = 0.904). Both models were further validated by an external test set of eight compounds with satisfactory predictions: r2 = 0.607 for CoMFA and r2 = 0.701 for CoMSIA. In addition, the 3D structure of human kappa opioid receptor was constructed based on the crystal structure of bovine rhodopsin, and the CoMSIA contour plots were then mapped into the structural model of kappa opioid receptor-GNTI complex to identify key residues, which might account for kappa antagonist potency and selectivity. The roles of nonconserved Glu297 and conserved Lys227 of human kappa opioid receptor were then discussed.

Amino Acid Sequence↗
PubMed2006

A risk score for mortality after allogeneic hematopoietic cell transplantation.

BACKGROUND: Despite recent advances, mortality rates after allogeneic hematopoietic cell transplantation remain high and cannot be accurately predicted. OBJECTIVE: To develop a reliable and valid predictor of all-cause mortality during the first 2 years after allogeneic hematopoietic cell transplantation. DESIGN: Retrospective cohort. SETTING: Tertiary hematopoietic cell transplantation center. PATIENTS: Patients (n = 2802) who received a first hematopoietic cell transplant between 1990 and 2002 were assigned to a development group or a validation group. MEASUREMENTS: Potential predictor variables were assessed with univariate and multivariable Cox proportional hazards methods to generate a prediction model. The c-statistic was calculated for 5 validation cohorts to assess model performance across early and late time periods and among patients with different diagnoses. RESULTS: The authors constructed a 50-point Pretransplantation Assessment of Mortality (PAM) score that incorporated 8 pretransplantation clinical variables: patient age, donor type, disease risk, conditioning regimen, FEV1, carbon monoxide diffusion capacity, serum creatinine level, and serum alanine aminotransferase concentration. The risk for death within 2 years for patients with PAM scores in the highest category was significantly higher than for those with scores in the lowest category. C-statistic values ranged from 0.69 to 0.76 for all validation cohorts. LIMITATIONS: The predictor model was not validated in an external cohort and is only useful for predicting the risk for death within the first 2 years after hematopoietic cell transplantation. CONCLUSIONS: Integrating pretransplantation clinical variables into a single score reliably predicts survival within 2 years after allogeneic hematopoietic cell transplantation. Accurate estimates of the risk for death may be useful in clinical trials and in epidemiologic studies. Such information can also be used to help physicians counsel patients regarding the expected outcomes of this potentially curative procedure.

Adult↗
PubMed2026

Natural language processing-based model to predict radiation pneumonitis in patients with locally advanced non-small cell lung cancer undergoing chemoradiotherapy: a retrospective cohort study.

BACKGROUND: Radiation pneumonitis (RP) remains a significant treatment-related toxicity in patients with unresectable, locally advanced non-small cell lung cancer (NSCLC) undergoing chemoradiotherapy (CRT). Most existing predictive models rely on static baseline demographic or dosimetry variables and lack real-time clinical applicability. We developed a novel predictive framework that integrates longitudinal symptom data extracted from clinical notes using natural language processing (NLP) with clinical and dosimetry features to improve early RP prediction. METHODS: We retrospectively identified 227 patients with locally advanced NSCLC treated with definitive CRT at a high-volume cancer center in the United States. We included all patients older than 18 years who were diagnosed between Jan 1, 2006, and Dec 31, 2022 with histologically or cytologically confirmed unresectable Stage 2 or 3 NSCLC and treated with conformal radiotherapy to a minimum dose of &#x2265;45 Gy with or without chemotherapy. Of these, 31 RP events were identified through manual adjudication using radiologic criteria and chart review. NLP was used to extract the temporal relationship of 16 pre-specified symptoms with treatment from over 100,000 clinical notes spanning pre- and during-treatment intervals. We trained and validated machine learning models on combinations of baseline clinical data, radiation dosimetry, and NLP-derived symptom features. Model performance was evaluated using a nested cross-validation framework, with an outer cross-validation loop reserved for performance assessment and an inner cross-validation loop used for model training and integration, and summarized using area under the receiver operating characteristic curve (AUC) and partial AUC (pAUC) at high specificity thresholds. Clinical utility was evaluated using decision curve analysis (DCA). FINDINGS: The best-performing model incorporated longitudinal NLP features and achieved a median AUC of 0.759 (90% confidence interval 0.753-0.766), significantly outperforming baseline models using only dosimetry (AUC 0.613) or clinical variables (AUC 0.635). NLP-based features such as cough trajectory, shortness of breath, and wheezing were among the most important predictors. Inclusion of NLP-derived symptom data improved early identification of high-risk patients, particularly in the clinically relevant high-specificity range (pAUC 0.021 vs. 0.010 for dosimetry alone). DCA showed that the calibrated MLP model provided greater net benefit than default strategies of treating all or no patients across clinically relevant threshold possibilities. INTERPRETATION: In this early work, NLP-based extraction of longitudinal symptoms from routine clinical documentation meaningfully enhances RP prediction in patients undergoing CRT for NSCLC. This approach leverages existing electronic health record infrastructure to deliver real-time, scalable, and interpretable risk estimates, offering a pathway toward potential early intervention and personalized toxicity management. The model and DCA requires external and prospective validation before clinical deployment; as such, future work should focus on this validation and integration into clinical decision support systems. FUNDING: AstraZeneca.

Chemoradiotherapy↗
PubMed2005

Quantitative study of the structure-retention index relationship in the imine family.

The Kováts retention index is one of the most popular descriptors of the performance of organic compounds in gas chromatography (GC). The mathematical modeling of this index is an interesting and open problem in analytical chemistry. In this paper, two models for the prediction of the Kováts retention index are presented. Topologic, topographic and quantum-chemical descriptors were used as structural descriptors. Multiple linear regression (MLR) analysis provides the first model using the forward stepwise procedure for the variable selection. For the second one, an ensemble of artificial neural network (ANN) was constructed using the pruning algorithm. Both methods were validated by an external set of compounds, by the Golbraikh and Tropsha method and by the leave-one-out (LOO) and the leave many out (LMO) procedures. The R2, RMScv and Q2, values for the training sets were 0.884, 0.589 and 0.830 for NN and 0.974, 0.417 and 0.970 for MLR models, respectively. The robustness of both models was demonstrated. Both portrait the chromatographic performance of the sample but in this case, the results of MLR equation are better than the NN ones. The MLR model is recommended because of its simplicity.

Imines↗
PubMed2003

Exploring and enhancing engagement to the psychosocial intervention 'Preparing for Parenthood'.

BACKGROUND: Poor compliance by participants consenting to be randomised to receive both physical and mental health promotion interventions represents a potentially serious threat to external and internal validity of those interventions. Quantitative and qualitative investigation of possible predictors of engagement forms an appropriate basis for efforts to enhance it. METHODS: Eight 'Preparing for Parenthood' intervention courses of a randomized controlled trial (RCT) underpinned a quantitative study. One 'Preparing for Parenthood' (PFP) intervention course, run upon completion of the RCT, formed a qualitative study. All nine courses followed identical procedures to enable clear comparisons. The three factors quantitatively explored with respect to engagement in health promoting behaviours were: locus of control (LOC), psychosocial support, and life events. The qualitative study utilised grounded theory analysis, the participants reflecting upon their experiences of the intervention and/or their reasons for not attending the course; nine interviews were completed. RESULTS: Participants in the quantitative and qualitative studies were divided into three sub-groups: compliant, non-compliant, and refusers. None of the three health promoting variables predicted compliance to a statistically significant degree. However, a variable from the trial analysis was found to reach significance; those women who had had less contact with the National Health Service in the 12 months prior to the baseline assessment were more likely to refuse the invitation to PFP. The qualitative study produced nine main themes that had influenced participant engagement at both the initial recruitment stage and during the course itself. CONCLUSIONS: In combination these findings may contribute to the future design of both effective and acceptable interventions to prevent postnatal depression. One such modified intervention is described and its impact on engagement outlined.

Depression, Postpartum↗
PubMed2000

Patient satisfaction with screening flexible sigmoidoscopy.

BACKGROUND: Screening flexible sigmoidoscopy is an underused cancer prevention procedure. Physicians often cite patient discomfort as a reason for not requesting sigmoidoscopy, but patient experiences and attitudes toward sigmoidoscopy have not been well studied. OBJECTIVE: To measure patient satisfaction and the determinants of satisfaction with screening sigmoidoscopy. METHODS: An instrument to assess satisfaction with screening sigmoidoscopy was developed. Responses were evaluated with a factor analysis, tested for reproducibility and internal consistency, and validated against an external standard. RESULTS: A total of 1221 patients (666 men and 555 women; mean age, 61.8 years) were surveyed after sigmoidoscopy. Examinations were performed by a nurse practitioner (n = 668), internist (n = 344), or gastrointestinal specialist (n= 184). More than 93% of the participants strongly agreed or agreed they would be willing to undergo another examination, and 74.9% would strongly recommend the procedure to their friends. Regarding pain and discomfort, 76.2% strongly agreed or agreed that the examination did not cause a lot of pain, 78.1% stated that it did not cause a lot of discomfort, and 68.5% thought that it was more comfortable than they expected. Fifteen percent to 25% of the patients indicated they had a lot of pain, great discomfort, or more discomfort than expected. Women were more likely to have significant pain or discomfort than men (adjusted odds ratio, 2.9; 95% confidence interval, 1.9-4.3; P<.001). CONCLUSIONS: Approximately 70% of individuals who undergo screening sigmoidoscopy are satisfied and find the procedure more comfortable than expected, whereas only 15% to 25% find the procedure unpleasant. Physicians should not project discomfort onto patients as a reason for not requesting screening sigmoidoscopy.

Factor Analysis, Statistical↗
PubMed2001

Rigorous new approach to constructing a gold standard for validating new diagnostic criteria, as exemplified by the eosinophilia-myalgia syndrome.

BACKGROUND: Constructing diagnostic criteria, a common problem in clinical medicine, is particularly difficult for diseases that lack a pathognomonic "gold standard." To develop an improved strategy for constructing such criteria, we used the eosinophilia-myalgia syndrome as an example. The goal, for research classifications, was to construct validated clinically sensible criteria and to develop improved methods that can be used for other disorders. METHODS: Using a "pattern-based" approach with data from several separate sources, a committee of investigators first prepared and informally tested criteria for the diagnosis of eosinophilia-myalgia syndrome. A gold standard challenge set of reports of cases and noncases was independently generated and separately validated by an external panel of clinical experts. The criteria were then tested using the gold standard set, and interobserver variability and diagnostic accuracy were determined. RESULTS: Interobserver variability showed the following mean proportionate agreements: 98.7% for the presence of specific criteria elements, 99% to 100% for diagnosis, and 97% to 98% for diagnostic pattern. kappa Values were correspondingly high. Diagnostic accuracy showed sensitivity at 88%, specificity at 97%, and overall accuracy at 92%. CONCLUSIONS: The proposed criteria are accurate and reproducible, and can be used in future clinical investigations of the eosinophilia-myalgia syndrome. The new strategy and methods developed for this challenge can be valuable for solving analogous problems in constructing criteria for other clinical disorders.

Diagnostic Techniques and Procedures↗
PubMed1991

Technetium 99m mercaptoacetyltriglycine gamma camera clearance calculations: methodological problems.

Major sources of errors in the gamma-camera methods for the calculation of renal clearance are the accuracy of background correction for obtaining the true renal time-activity curve and the validity of the externally recorded pre-cordial activity as an estimate of the plasmatic time-activity curve. With technetium 99m mercaptoacetyltriglycine (99mTc-MAG3), because of its high protein plasma binding, one could expect minimal extravascular diffusion and hence a more accurate externally detected plasmatic curve. The high extraction rate should reduce the influence of the background, but, on the other hand, the effect of hepatobiliary excretion on the calculation of renal clearance might be significant. Our results suggest that the hepatobiliary excretion of 99mTc-MAG3 does not influence the gamma-camera renal clearance determination, even in patients with low renal function. However, the pre-cordial curve does not reflect accurately the plasmatic disappearance curve; its calibration with a single plasma sample taken at the 20th min is responsible for significant errors, probably because of an unfavourable ratio between the intravascular and extravascular activities at the 20th min.

Adult↗
PubMed1997

[Comparison of psychodiagnostic methods for dementia and deterioration].

Several cognitive impairment/dementia diagnostic methods were examined concerning their results: (a) in the differentiation of accidental and pathological ability changes; (b) in the demonstration/exclusion of cognitive impairment; and (c) the degree of dementia. Sixty-five patients suffering from brain damage were examined with five accepted methods of diagnosing dementia (Syndrom-kurztest, KAI-MWT-Methode, Demenz-Test, Mini-Mental-Status-Test, Wurzer-Methode) and a comprehensive performance test battery as an external criterion for valid determination of the degree of dementia and definite differentiation of accidental and pathological performance changes. The five methods of diagnosing dementia differ appreciably in determining the degree of severity (r = 0.44). They are effective in registering accidental performance changes in comparison with the outer criterion, but pathological changes are inaccurately registered (hit rate 88%/53%); the results concerning the degree of cognitive impairment are identical: between 25 and 43% (r = 0.43). High rates (58%) of false-negative diagnoses are especially apparent in the range of slight and intermediate cognitive impairment. The methods examined are only useful for the demonstration and not for the exclusion of severe cognitive impairment (dementia) and in no circumstances for the registration of slight/intermediate cognitive impairment. Valid diagnosis of cognitive impairment/dementia necessitates the use of test batteries that differentiate functions including the premorbid performance level.

Adult↗
PubMed2006

Investigation of extraction procedures and HPLC-DAD/MS for the determination of the brominated flame retardant tetrabromobisphenol A bis(2,3-dibromopropylether) in environmental samples.

A method for the determination of the novel brominated flame retardant tetrabromobisphenol A bis(2,3-dibromopropylether), 1,1'-(isopropylidene)bis[3,5-dibromo-4-(2,3-dibromo-propoxy)-benzene] (TBBPA-dbpe), was developed. Technical TBBPA-dbpe was purified and the results of a thorough physical characterisation are reported. The application of APCI-MS is discussed and the fragmentation patterns are described. Quantification of TBBPA-dbpe was done by HPLC-DAD using external calibration. The validation of the method was accomplished using sediment and sewage sludge samples spiked with defined amounts of authentic TBBPA-dbpe. The average recovery rates of TBBPA-dbpe from spiked samples ranged from 35 to 91% (sediment) and from 57 to 98% (sewage sludge) depending on the respective extraction method. Pressurised fluid extraction (PFE) and fluidised bed extraction were superior to classical Soxhlet and sonication procedures and yielded recovery rates between 90 and 98% with relative standard deviations of 2%. The limits of detection (DTC), identification (ID) and determination (DTM) using HPLC-DAD were 10, 21 and 30 ng g(-1) in sediment and 22, 44 and 72 ng g(-1) in sewage sludge, respectively.

Journal Article↗
PubMed2003

[Psychometric properties of the pain drawing and the Ransford technique in patients with chronic low back pain].

BACKGROUND: The aim of our study was the analysis of the psychometric properties of the Ransford technique which is a qualitative scoring method of the pain drawing. It has been suggested that this method could be used as a brief screening technique for psychological involvement in pain complaints. METHODS: 109 patients with chronic back pain filled in the pain drawing which was rated according to the technique described by Ransford and colleagues in 1976. As external criterium of validity we chose a differentiated expert rating of psychological interference. RESULTS: The Ransford technique showed moderate to good coefficients of reliability (re-test, inter and intra-rater). On the other hand, the Ransford technique demonstrated very low indices of validity in order to give a prognosis about psychological interference. Sensitivity and specificity of the Ransford technique demonstrated inadequate values, which furthermore were independent from cut-off points. CONCLUSION: The pain drawing and the Ransford technique are not sufficiently sensitive and therefore allow no screening for psychological interference in complaints of back pain.

Adult↗
PubMed2005

Prediction of toxicity using a novel RBF neural network training methodology.

A neural network methodology based on the radial basis function (RBF) architecture is introduced in order to establish quantitative structure-toxicity relationship models for the prediction of toxicity. The dataset used consists of 221 phenols and their corresponding toxicity values to Tetrahymena pyriformis. Physicochemical parameters and molecular descriptors are used to provide input information to the models. The performance and predictive abilities of the RBF models are compared to standard multiple linear regression (MLR) models. The leave-one-out cross validation procedure and validation through an external test set produce statistically significant R2 and RMS values for the RBF models, which prove considerably more accurate than the MLR models. [Figure: see text].

Animals↗
PubMed2007

A novel QSPR model for predicting θ (lower critical solution temperature) in polymer solutions using molecular descriptors.

In this study, we present a new model that has been developed for the prediction of θ (lower critical solution temperature) using a database of 169 data points that include 12 polymers and 67 solvents. For the characterization of polymer and solvent molecules, a number of molecular descriptors (topological, physicochemical,steric and electronic) were examined. The best subset of descriptors was selected using the elimination selection-stepwise regression method. Multiple linear regression (MLR) served as the statistical tool to explore the potential correlation among the molecular descriptors and the experimental data. The prediction accuracy of the MLR model was tested using the leave-one-out cross validation procedure, validation through an external test set and the Y-randomization evaluation technique. The domain of applicability was finally determined to identify the reliable predictions.

Linear Models↗
PubMed1981

Limitations of scalogram analysis as a method for investigating drug use behavior.

Scalogram analysis, a procedure developed for constructing attitude scales, has been recently applied to support certain conclusions about the typical development of drug use patterns. An examination of results from these applications revealed at least three potential limitations: (1) presumed drug usage patterns may not generalize to many drug users, especially adult users; (2) presumed patterns may not be supported if several commonly used psychoactive substances are included; and (3) measures of unidimensionality may not be sufficient for confirming actual unidimensionality of highly skewed drug use patterns. Therefore, the present study was designed to investigate further these potential limitations and to determine their bearing upon previous theoretical conclusions. Results confirmed the existence of all three limitations. In other words, typical applications of scalogram analysis based upon adolescent illicit drug use data have questionable external and internal validity. Therefore, previous conclusions based upon such applications should be re-evaluated and future research should employ alternative analytical procedures.

Adolescent↗
PubMed2006

A validated clinical and biochemical score for the diagnosis of acute heart failure: the ProBNP Investigation of Dyspnea in the Emergency Department (PRIDE) Acute Heart Failure Score.

BACKGROUND: No method integrating amino-terminal pro-brain natriuretic peptide (NT-proBNP) testing with clinical assessment for the evaluation of patients with suspected acute heart failure (HF) has been described. METHODS: Amino-terminal pro-brain natriuretic peptide results and clinical factors from 599 patients with dyspnea were analyzed. The beta coefficients of the 8 independent predictors of HF were used to assign a weighted integeric score for predictor. The sum of these integers provided a diagnostic HF "score" for each patient. Receiver operating characteristic curve analysis determined the optimal cut point for the diagnosis of acute HF. The performance of the score was evaluated in the development cohort and subsequently in a patient population from a separate clinical trial of patients with dyspnea conducted in Christchurch, New Zealand. RESULTS: Eight factors comprised the score: elevated NT-proBNP (4 points), interstitial edema on chest x-ray (2 points), orthopnea (2 points), absence of fever (2 points), loop diuretic use, age > 75 years, rales, and absence of cough (all 1 point). Median scores in patients with acute HF were higher than those without acute HF (9 vs 3 points, P < .001). At a cut point of > or = 6 points, the score had a sensitivity of 96% and a specificity of 84% for the diagnosis of acute HF (P < .001). The score improved diagnostic accuracy over NT-proBNP testing alone and retained discriminative capacity in patients in whom clinical uncertainty was present. Lastly, the accuracy of the score was validated in the external data set of patients with suspected acute HF. CONCLUSION: We report a simple and accurate scoring system combining NT-proBNP testing and clinical assessment for the diagnosis or exclusion of acute HF in patients with dyspnea.

Acute Disease↗
PubMed2005

Prognosis after aortic valve replacement with the Carpentier-Edwards pericardial valve: use of microsimulation.

BACKGROUND: The second-generation Carpentier-Edwards pericardial valve (Edwards Lifesciences LLC, Irvine, CA) is widely used for aortic valve replacement. However, knowledge on the long-term outcomes of patients after valve implantation is incomplete. We used microsimulation to calculate the long-term outcome of any given patient after aortic valve replacement with the Carpentier-Edwards pericardial valve. METHODS: A meta-analysis of 8 reports on aortic valve replacement with the Carpentier-Edwards pericardial valve (2,685 patients; 12,250 patient years) was used to estimate the hazards of valve-related events other than structural valvular deterioration. Structural valvular deterioration was described by age-dependent Weibull curves calculated from 18-year follow-up, premarket approval, Carpentier-Edwards pericardial primary data. These estimates provided the input data for the parameters of the microsimulation model, which was then used to calculate the outcomes of patients of different ages after valve implantation. The model estimates of survival were validated using two external data sets. RESULTS: The Weibull analysis estimated a median time to reoperation for structural valvular deterioration ranging from 18.1 years for a 55-year-old male to 23.2 years for a 75-year-old male. For a 65-year-old male, microsimulation calculated a life expectancy and event-free life expectancy of 10.8 and 9.1 years, respectively. The lifetime risk of at least one valve-related event was 38% and that of reoperation due to structural valvular deterioration 17%, respectively, for this patient. The model estimates of survival showed good agreement with external data. CONCLUSIONS: Microsimulation provides detailed insight into the long-term prognosis of patients after aortic valve replacement. The Carpentier-Edwards pericardial valve performs satisfactorily and offers a low lifetime risk of reoperation due to structural valvular deterioration, especially for elderly patients requiring aortic valve replacement.

Adult↗
PubMed2004

Molecular classification of scrapie strains in mice using gene expression profiling.

Transmissible spongiform encephalopathy strains demonstrate specific prion characteristics, each with specific incubation times, and strain-specific patterns of deposition of the misfolded isoform of prion, PrPSc, in the brains of infected individuals. Different biochemical properties, including glycosylation profiles and the degree of proteinase resistance, have been shown to be strain-specific. However, no relationship between these properties and the phenotypic differences in the subsequent diseases has as yet been determined. Here we explore the utility of gene expression profiles to identify differences in the host response to different strains of prion agent. We identify 114 genes that exhibit significantly different levels of expression in mice infected with three strains of scrapie. These genes represent a pool of genes involved in a strain-specific response to prion disease. We have identified the most discriminatory genes from this list utilizing a wrapper-based feature selection algorithm with external cross-validation.

Algorithms↗

Refine your search to explore more results.