PubMed Health⌕ Search

Biomedical subjects

M J Roger

Publications and source records attributed to M J Roger.

4 recordsLinked to original sources

[Computer-assisted diagnostic system: methods used (author's transl)].

Having chosen a computer-assisted diagnostic system (ADM) based on a logic method of the documentation type, a real example of a questionnaire is presented. The procedures actually utilized as well as those now being studied are presented and discussed in the light of the difficulties encountered. After giving various detailed examples, the fundamental procedures in developing the computer-assisted diagnostic system are described. These have been developed keeping in mind that the system must preserve utility, feasibility, transparency, simplicity, liberty of choice and scope but must also handle all difficulties and causes of error that are inherent in the process of arriving at a medical diagnosis.

Computers↗

[By-Products of the computer-assisted diagnostic system (ADM) (author's transl)].

The data-bank of signs and symptoms and classes of diseases developed for the computer-assisted medical diagnostic system (ADM) allows us to develop a series of interesting by-products. Some by-products are utilitarian: the inverted file and lexicon subset of signs, symptoms and diseases organized by medical discipline. Other by-products in the area of operational research serve to increase the quality and productivity of medical practice. Among the most interesting we note one by-product that helps teach physical diagnosis, one for therapeutics, one for the partial automation of observations, and one for assistance in decision-making in various specialized medical functions. The number and importance of these by-products demonstrates the powerful potential of the computer-assisted diagnostic system (ADM).

Diagnosis, Computer-Assisted↗

[Creation, development and maintenance of the data-base of a computer-assisted diagnostic system (ADM) (author's transl)].

The creation of the data-base for our ADM system begun in 1972, tests on a high productivity type of procedure which we consider should be explained. In organizing and creating our data-bank our routine procedures are: first, the capture of signs, symptoms and pathological data after the text editing of the description of diseases as prepared by medical students under the direction of the Medical Computing Service; second, the control and coding of the data by the personnel of the Medical Computing Service: third, the machine storage of the data and the automated logical controls by the Medical Computing Service; and fourth, the semantic control of the data as well as the availability of each new version. Other related procedures based on the individual theses and various studies are described as examples. From 1972 to 1980 an impressive data-bank was created holding descriptions of 2200 diseases with 22000 signs and symptoms and 3500 bibliographies. It is now foreseen that between 1980 and 1983, the prototype will be placed in service and adapted. The creation and maintenance of such a data-bank with all the problems that it entails, and developed under such economical conditions justifies our presentation.

Data Display↗