PubMed · 10130884
The source did not provide an abstract. Follow the original record for more information.
Explore connections, maps & timelines
J Petersen. 1993. Automating training records.. https://pubmed.ncbi.nlm.nih.gov/10130884/
Cite the original work for its findings. Save a collection to share your selection of sources.
KEEP EXPLORING
Explore the source record for details and available documents.
The anticipated introduction of universal antenatal screening can be expected to increase the workload of haemoglobinopathy laboratories throughout the country. We have devised a rule-based system to process those results that does not require skilled interpretation, thereby freeing staff time for more specialized work. The system relies on a calculated test to create a total haemoglobin peak value, which incorporates the values for HbA, HbA2 and HbF, the MCV and MCH from the full blood count. Each parameter has a series of defined ranges which, when subjected to an interpretation process within the laboratory computer system, generates an automated result text for the sample. During a 6-month verification period, the automated result interpretation system in conjunction with laboratory information systems (LIS) validation reduced the number of samples requiring manual review by 60%. The system was found to be 100% sensitive and 61.8% specific. We feel that the current specificity is acceptable in order to maintain a safe system. The ability to concentrate on potentially abnormal results will allow laboratories and health care workers more time to develop appropriate and timely frameworks to deal with abnormal results.
Explore the source record for details and available documents.