<![CDATA[Electronic Medical Records (EMRs) are digital applications of machine learning models that function to receive and store clinical data related to medical information for the purposes of basic clinical diagnostic tests. The integrative review aims to provide a synthesis of new findings from several articles on EMRs for the early detection of basic clinical diagnoses with a variety of existing populations. Using four databases, we reviewed 11 articles. All authors involved review abstracts and full text according to predetermined criteria. The selected articles are then integrated into the publication quality assessment matrix, further included in machine learning algorithms for diagnostic determination of the disease. Reviewed articles are excluded in the form of artificial intelligence. The PRISMA flowchart identified 1962 articles and the final selection found 11 articles. Circulating system networks dominate machine learning models (66.6%). The study netted an average population of 490.5 and the artificial intelligence system managed to detect 9 body systems from different body systems. A total of 11 articles were selected, more than half of which were Caucasian (80.90%) and white (72.95%), but only 1 article was represented by Caucasian ethnicity, while white race was almost in every article. African-American and Black racial groups were in the middle position at 29.95% and 17.50%. The racial representation with the least percentage below 10% was Hispanic and Asian (6.10% and 2.17%). This machine learning has proven to be very accurate for detecting disease diagnoses in hospital, other health clinics. Therefore, the further development of this application for the purpose of establishing clinical diagnosis precisely and accurately.]]>
