PubMed Health⌕ Search

Biomedical subjects

K A Spackman

Publications and source records attributed to K A Spackman.

8 recordsLinked to original sources
PubMed1992

Combining logistic regression and neural networks to create predictive models.

Neural networks are being used widely in medicine and other areas to create predictive models from data. The statistical method that most closely parallels neural networks is logistic regression. This paper outlines some ways in which neural networks and logistic regression are similar, shows how a small modification of logistic regression can be used in the training of neural network models, and illustrates the use of this modification for variable selection and predictive model building with neural networks.

Algorithms↗
PubMed1991

Maximum likelihood training of connectionist models: comparison with least squares back-propagation and logistic regression.

This paper presents maximum likelihood back-propagation (ML-BP), an approach to training neural networks. The widely reported original approach uses least squares back-propagation (LS-BP), minimizing the sum of squared errors (SSE). Unfortunately, least squares estimation does not give a maximum likelihood (ML) estimate of the weights in the network. Logistic regression, on the other hand, gives ML estimates for single layer linear models only. This report describes how to obtain ML estimates of the weights in a multi-layer model, and compares LS-BP to ML-BP using several examples. It shows that in many neural networks, least squares estimation gives inferior results and should be abandoned in favor of maximum likelihood estimation. Questions remain about the potential uses of multi-level connectionist models in such areas as diagnostic systems and risk-stratification in outcomes research.

Diagnosis, Computer-Assisted↗
PubMed1991

Information workstations in clinical pathology.

Multitasking operating systems and expanding networks now permit smooth access to remote computers, peripherals, data, and information resources. Graphic user interfaces and productivity-enhancing software packages reduce the need for training and memorization of commands. New models of desktop computers based on "data-centered" software architecture can enhance workstation usefulness even more. Pathologists need to consider how these tools might improve access to and management of information and knowledge.

Computer Systems↗
PubMed1990

A knowledge-based system for transfusion advice.

A knowledge-based system has been designed for evaluating the appropriateness of transfusion of non-red blood cell blood components. The goal of the system is to assist the blood bank physician in quality assurance efforts by automatically identifying cases of inappropriate transfusion before the blood is issued. Evaluation of a working prototype system shows that it is indeed capable of serving this function. The system identifies and summarizes cases, but it leaves consultation, education, and decision making to the blood bank physician. Small "expert systems" such as this may find use in quality assurance activities throughout the laboratory.

Blood Transfusion↗
PubMed1987

Knowledge-based systems in laboratory medicine and pathology. A review and survey of the field.

Knowledge-based systems are computer systems designed to handle knowledge-intensive tasks, usually involving reasoning and inference. They are increasingly being applied to problems in laboratory medicine and pathology. In this article we provide a brief introduction to the basic concepts of knowledge-based systems, review some of their published applications, and report on an informal survey of specialists in laboratory medicine and pathology. The survey, sent to 102 individuals, indicated that 24% were involved in developing knowledge-based systems, with most systems at an early stage of development. Recent advances in knowledge-based systems research as well as survey responses suggest that this technology will have increasing value in laboratory medicine and pathology.

Artificial Intelligence↗
PubMed1985

A program for machine learning of counting criteria: empirical induction of logic-based classification rules.

A program has been developed which derives classification rules from empirical observations and expresses these rules in a knowledge representation format called 'counting criteria'. Decision rules derived in this format are often more comprehensible than rules derived by existing machine learning programs such as AQ11. Use of the program is illustrated by the inference of discrimination criteria for certain types of bacteria based upon their biochemical characteristics. The program may be useful for the conceptual analysis of data and for the automatic generation of prototype knowledge bases for expert systems.

Artificial Intelligence↗