Artificial neural network: predicted vs observed survival in patients with colonic cancer.

PURPOSE: An Internet-web-based artificial neural network has been developed for practicing clinical oncologists and medical researchers as part of an ongoing program designed for the implementation of advanced neural networks for prognostic estimates and eventually for management/treatment decisions in individual patients with colonic cancer. METHODS: An interdisciplinary team of academic oncologists and physicists has configured and implemented a Partial Logistic Artificial Neural Network and trained it to predict cancer-related survival in patients with confirmed colorectal cancer by using a database (1,558 patients) made available for the study by the Information & Statistics Division of National Health Service Scotland. The reliability of the trained network was evaluated against Kaplan-Meier observed survival plots of a random sample of 300 patients not used in the training but forming part of the same data set. RESULTS: The predicted survival curves obtained as the output from the artificial neural network showed close agreement with observed actual survival rates of a cohort of 300 patients with four grades of risk of dying from the cancer within five years of diagnosis. CONCLUSIONS: The web-based Partial Logistic Artificial Neural Network system accurately predicts survival after staging and treatment of colonic cancer. It can be made web-accessible where it is powerful enough to serve hundreds of users simultaneously.

Intraocular light scatter as modeled through a stratified medium.

Intraocular light scatter is modeled through a stratified medium. It is shown that the eye lens possesses properties of an interference filter. In conditions that may mimic opacification, as found with cataract formation, such a filter may have stop zones with reflectance up to 1 in the band of visible light.

Electrostatic interaction between stereocilia: I. Its role in supporting the structure of the hair bundle.

This paper provides theoretical estimates for the forces of electrostatic interaction between adjacent stereocilia in auditory and vestibular hair cells. Estimates are given for parameters within the measured physiological range using constraints appropriate for the known geometry of the hair bundle. Stereocilia are assumed to possess an extended, negatively charged surface coat, the glycocalyx. Different charge distribution profiles within the glycocalyx are analysed. It is shown that charged glycocalices on the apical surface of the hair cells can support spatial separation between adjacent stereocilia in the hair bundles through electrostatic repulsion between stereocilia. The charge density profile within the glycocalyx is a crucial parameter. In fact, attraction instead of repulsion between adjacent stereocilia will be observed if the charge of the glycocalyx is concentrated near the membrane of the stereocilia, thereby making this type of charge distribution unlikely. The forces of electrostatic interaction between stereocilia may influence the mechanical properties of the hair bundle and, being strongly non-linear, contribute to the non-linear phenomena that have been recorded from the periphery of the auditory and vestibular systems.

Electrostatic interaction between stereocilia: II. Influence on the mechanical properties of the hair bundle.

This paper is based on our model [Dolgobrodov et al., 2000. Hear. Res., submitted for publication] in which we examine the significance of the polyanionic surface layers of stereocilia for electrostatic interaction between them. We analyse how electrostatic forces modify the mechanical properties of the sensory hair bundle. Different charge distribution profiles within the glycocalyx are considered. When modelling a typical experiment on bundle stiffness measurements, applying an external force to the tallest row of stereocilia shows that the asymptotic stiffness of the hair bundle for negative displacements is always larger than the asymptotic stiffness for positive displacements. This increase in stiffness is monotonic for even charge distribution and shows local minima when the negative charge is concentrated in a thinner layer within the cell coat. The minima can also originate from the co-operative effect of electrostatic repulsion and inter-ciliary links with non-linear mechanical properties. Existing experimental observations are compared with the predictions of the model. We conclude that the forces of electrostatic interaction between stereocilia may influence the mechanical properties of the hair bundle and, being strongly non-linear, contribute to the non-linear phenomena, which have been recorded from the auditory periphery.

[Numerical modeling of convection diffusion of endolymph in porous structures of the vestibular apparatus]. / Chislennoe modelirovanie konvektivnoi diffuzii éndolimfy v poristykh strukturakh vestibuliarnogo apparata.

The mathematical algorithm of the diffusive model of vestibular reception is discussed. The conditions to reduce arising 3-dim boundary value problem to 1-dim one are found. And it is shown, that separate diffusive canaliculi as a mechanical transducers demonstrate the apparently clear properties of the frequency selectivity relating to their transduction of the endolymph space pressure into the endolymph concentration changes in their outputs. Some numerical experimental results are presented and discussed.