ABSTRACT
The aim of this study was to develop a computer expert system that could reproduce a pathologist's diagnosis of iron deficiency from the data obtained from blood tests. 275 cases were collected for construction and testing of the expert system. The expert system used a combination of fuzzy set logic and cut-off points from 14 parameters to arrive at one of 5 diagnostic categories graded from "iron deficient" to "no evidence of iron deficiency". The agreement between pathologist and expert system was 0.91 (Spearman rank correlation coefficient) in the learning population; this dropped to 0.79 in the test population. Absolute agreement on diagnostic category was reached in 71% of cases. In no case was there disagreement by more than 3 grades.
Subject(s)
Anemia, Hypochromic/diagnosis , Diagnosis, Computer-Assisted , Expert Systems , Evaluation Studies as Topic , Female , Humans , Male , Middle AgedABSTRACT
It frequently happens that following a patient consultation similar data are recorded in a number of ways in a number of places. This paper describes a method which has been used in one busy clinic to capture this data only once, effecting a saving in consultant's time, avoiding delay in data capture and ensuring consistency of databases. The implementation, using a mixture of custom programs and commercial database software, is discussed. Examples of the sort of data input and user interfaces and some future directions for extension are given. Limitations of commercial packages encountered are mentioned.
Subject(s)
Data Collection/methods , Medical Records Systems, Computerized , Patient Discharge , New Zealand , Referral and Consultation , Software , User-Computer InterfaceABSTRACT
A method is described to capture data on the physical examination of children. A personal computer with a keyboard, keypad and screen is used along with a menu driven software package. Little knowledge of computing is necessary and the system can be used to produce hard copy when a suitable printer is added. Accumulated data may be transferred to a statistical package for future analysis. The method used to capture data from children with respiratory disease, particularly acute asthma, is described here, but it is stressed that the method can be adapted easily to data collection from the results of the physical examination of any system.
Subject(s)
Information Storage and Retrieval , Microcomputers , Pediatrics/methods , Physical Examination , Asthma , Child , HumansABSTRACT
On the basis of a modified version of the LOWTRAN 6 model, an absolute comparison is made of typical and background-limited IR sensors operating in the 3-5-microm and 8-12-microm wavebands in a tropical maritime environment. Allowance is made for slant paths and a variety of targets and backgrounds including hot targets and backgrounds spectrally different from the targets. It is found that with current detector technology the 8-12-microm waveband is superior for all except very hot targets at long ranges. The validity of various approximations is also investigated, and in particular it is found that the blackbody composite background approximation should not be used for slant paths.
