ABSTRACT
In 1988 the Veterans' Benefits and Services Act attempted to solve the problem of the lack of adequate VA healthcare facilities in rural areas by establishing a demonstration program using mobile clinics. Six clinics operated in areas that were at least 100 miles from a VA healthcare facility during the time period between October 1, 1992 and May 28, 1994. This article evaluated the effect of the mobile clinics' structural limitations on clinical care, the increased number of sites on VA usage, and cost. Limited space for storage of medical records and the unavailability of laboratory, electrocardiographic, or radiographic facilities significantly affected clinical practice. However, even with these space limitations, veterans' use of healthcare in the areas served by the mobile clinics increased significantly in comparison to reference areas. The direct costs per visit averaged more than three times what the VA would have reimbursed the private sector.
Subject(s)
Mobile Health Units/organization & administration , Rural Health Services/supply & distribution , United States Department of Veterans Affairs , Demography , Health Care Costs , Health Services Accessibility , Humans , Mobile Health Units/economics , Physicians/supply & distribution , Pilot Projects , Program Evaluation , Rural Health Services/economics , Rural Health Services/statistics & numerical data , United States , WorkloadABSTRACT
A number of techniques exist with which neural network architectures such as multilayer perceptrons and radial basis function networks can be trained. These include backpropagation, k-means clustering and evolutionary algorithms. The latter method is particularly useful as it is able to avoid local optima in the search space and can optimise parameters for which no gradient information exists. Unfortunately, only moderately sized networks can be trained by this method, owing to the fact that evolutionary optimisation is very computationally intensive. In this paper a novel algorithm (CERN) is therefore proposed which uses a special form of combinatorial search to optimise groups of neural nodes. Oriented, ellipsoidal basis nodes optimised with CERN achieved significantly better accuracy with fewer nodes than spherical basis nodes optimised by k-means clustering. Multilayer perceptrons optimised by CERN were found to be as accurate as those trained by advanced gradient descent techniques. CERN was also found to be significantly more efficient than a conventional evolutionary algorithm that does not use a combinatorial search.
