Search

In this paper, we have introduced a general modeling approach for dynamic nonlinear systems that utilizes a variant of the simulated annealing algorithm for training the Laguerre–Volterra network (LVN) to overcome the local minima and convergence problems and employs a pruning technique to achieve sparse LVN representations with \ell 1 regularization. We tested this new approach with computer simulated systems and extended it to autoregressive sparse LVN (ASLVN) model structures that are suitable for input–output modeling of nonlinear systems that exhibit transitions in dynamic states, such as the Hodgkin–Huxley (H-H) equations of neuronal firing. Application of the proposed ASLVN to the H-H equations yields a more parsimonious input–output model with improved predictive capability that is amenable to more insightful physiological/biological interpretation. [ABSTRACT FROM PUBLISHER]

Human atherosclerotic arterial samples were ablated via a fiber with a XeCl excimer laser. The resulting tissue fluorescence was recorded for each ablating pulse. The pulse-to-pulse evolution of the fluorescence intensity at 430 nm was obtained and compared to the histological findings. The characteristic transition observed in such curves exhibited good correlation with the transition from the atheromatous layer to the normal media, as determined by the histological examination. The establishment of such a relation led to the development of a simple computer algorithm able to detect plaque to normal media transitions. The limitations of this approach are discussed.