A COMPUTER-ASSISTED VALIDATION SYSTEM FOR LABORATORY DATA IN HEMATOLOGY - VALAB-HAEMATO

Validation of laboratory reports is the ultimate step before transmiss ion of results to the clinician. The biologist checks the intrinsic co nsistency of the data as well as their possible medical value that is liable to lead to other investigations. Such a policy, when performed on all the data, is time-consuming, boring and uncertain. This,step ma y be simplified by the use of a computerized expert system. The comput er assisted validation system presented here concerns routine haematol ogy data (Valab-haemato(R)). Like its predecessor devoted to clinical chemistry (Valab-Biochem(R)) it is based on the performance of a power ful inference engine which generates a decision-making tree for each r eport according to the data. This adaptability gives the system a capa city very close to human reasoning. In its haematology version the sys tem deals with many variables including sex, age, origin of the patien t (hospital ward), and the haematological data (blood cell count, diff erential, reticulocyte count, various information drawn from microscop e examination of the blood smear as well as any report concerning the blood sample, erythrocyte sedimentation rate). Previous data are also taken into account, as well as the normal ranges, the values beyond wh ich no result can be automatically validated and the delta-check. Some information definitely prevents validation of the result, others can be validated if they have been previously approved. Whereas the method of reasoning is fixed, all items are changeable in order to adapt the system to the type of activity of the laboratory. The quality of the system has been evaluated according to an epidemiological technique us ed for evaluation of the quality of a test: thus a series of 357 repor ts given by the computerized system were compared first with the separ ate findings of the three haematologists who had participated in the e stablishment of the database and the construction of the logic of the system, then with their consensual findings. Results were as follows: sensibility = 0.837 (target: 1), specificity = 0.838 (target: 1) and m ajor error = 0.163 (target: 0). Since it has been installed, the compu terized system runs 24 hours per day, the biologist Inking into consid eration only those reports not validated by the system. Evaluation of the performance of the system over a three month period showed that it had validated 77.7% of the reports processed in the laboratory, that is 86.5% of the reports taken into consideration. After 18 months util ization, all the clinicians of the hopital having been duly informed n o complaint that might be related to the validation process has been r ecorded.