Noted that there are various QSAR modeling based screening techniques while essentially they are learning based, thus only the typical SVR method was selected for comparison.
Crude and multivariable logistic regression models were applied with 'clinic based testing' and 'school based screening' as outcome variables.
The pharmacophore model was created with the same queries that were used for shape based screening.
In the multivariable model using school based screening as independent outcome, current urogenital symptoms in girls and in both genders combined were the only significant traditional risk factor.
The pharmacophore model based virtual screening unveils two lead compounds from the TIMBAL database of small molecules modulating the targeted protein-protein interactions.
This has important implications for the evaluation of the specificity and modelling of cost effectiveness of HPV based screening programmes.
Crude and multivariable logistic regression models were applied with two dependent variables: 1) 'clinic based testing'; yes/no, and 2) 'school based screening'; yes/no.
The predicted probabilities for screening from the latent class model, based on posterior medians, were dichotomized into screening and non-screening using a cut-off of 50% probability.
A semi-analytical model based on a thin screen assumption for fluid flow calculation and on correction factors to Stokes law for drag force is used to compute the particle trajectory.
The microsimulation model used in the present study fitted the data better when a somewhat higher threshold diameter, corresponding to a lower screening test sensitivity, and diameter of clinical detection is used compared with a model based on data from randomised screening trials (extrapolated to the current period) and from the US.
Initially, the model based on data from randomised screening trials (extrapolated to the current period) and US data simulated a more favourable tumour size distribution than observed in the trial for both groups.
