**4. Final evaluation of the articles of the systematic review virtual health library**

A total of 146 articles were evaluated, of which 97 were not concordant, 28 were highly concordant and 21 were relatively concordant.

The red non-concordance refers to the fact that the article does not make any reference to the stated objectives and the subject of interest, to the research question, or to the uses.

The highly concordant articles, identified with the green color, are directly related to the subject of interest, the stated objectives, the research question, and the uses.

In the case of the relatively concordant category, identified with the yellow color, they have partially related to the subject matter and the research question, but not to the objectives or uses (see **Table 4**).

During the Virtual Health Library Systematic Review, 135 articles from the MEDLINE database, seven articles from the LILACS database, and four from the IBECS database were analyzed.

*Artificial Intelligence in Medicine and Surgery – An Exploration of Current Trends Potential…*


#### **Table 4.**

*Compliance follow-up matrix resulting from the evaluation of systematic review articles. Colombia, 2019.*

Twenty-eight highly concordant articles were identified, whose analysis consisted of organizing them by areas of artificial intelligence; to be able to visualize the approaches that are currently being applied in medicine, surgical procedures, and medical specialties. This analysis produced the following findings from the evidence found by area of artificial intelligence:
