17 important role in the interpretation of results , for instance in flagging outlier values as well as setting up alerts and reminders . Furthermore , the CDSS can be used to extend the value of laboratory test results to avoid future invasive procedures , as illustrated in a study among patients with hepatitis B and C conducted by Keltch et al . Although a liver biopsy is considered to be the gold standard for the diagnosis of the disease , the incorporation of an artificial intelligence system into the CDSS combining the results from serum markers , imaging , and genetic tests provided greater accuracy than biopsies , thus avoiding the need for this invasive procedure . 15 CDSS that incorporated magnetic resonance imaging have also shown success in the classification of brain tumours . 16
A literature review by Peiffer-Smadja et al 17 aimed to inform clinicians about the use of machine learning ( ML ) for diagnosis , classification , outcome prediction and antimicrobial management in infectious diseases . Of the 60 unique ML-CDSS found , 62 % focused on bacterial infections , 17 % on viral infections , 15 % on tuberculosis and 7 % on any kind of infection . These addressed the diagnosis of infection , the prediction , early detection or stratification of sepsis , the prediction of treatment response or antibiotic resistance , the choice of antibiotic regimen and the choice of a combination antiretroviral therapy . The authors concluded that future ML-CDSS in infectious diseases should be developed in diverse health settings and be embedded in a structured process of integration into clinical settings .
Effective communication between the laboratory and the clinic is essential Although medical laboratories are essential for the diagnosis , treatment and management of patients , the testing workflow does not always proceed adequately owing to factors such as capacity , logistics , infrastructure , and availability of appropriate technology . A study conducted in a hospital in Norway identified common communication barriers between the microbiology laboratory and the clinical units at different levels , namely errors and omissions in request forms , deficient integration of laboratory and clinical information technology systems , and reporting of results . Also of concern was the perceived insufficient knowledge of procedures from both sides and specifically a lack of understanding of the potential and limitations of the tests by the requesting physician . 18
Ineffective communication results in a climate of mistrust when a clinician requests a large number of tests , for example , unaware of the workload of the laboratory staff ; the ordering physician may also fail to convey the medical value of a given assay to the laboratory . An efficient clinical laboratory maximises low cost , high quality , and timely delivery of results , and it is important that hospital administrators >