**14. Discussion**

There are many technological applications being made focused on the area of medicine, but not all of them are structured under artificial intelligence and expert systems. This led to evaluate current healthcare technologies in terms of artificial intelligence and expert systems.

#### **Figure 5.**

*Development of artificial intelligence application by medical specialty.*

**Figure 6.** *Decision making by artificial intelligence specialty.*

Important findings were obtained with respect to the interventions that have been applied in the clinical part.

Among the most relevant findings, it was recognized that the areas of artificial intelligence where different medical applications were developed were automatic reasoning, computer vision, expert systems, and deep learning with neural networks.

The articles identified as highly concordant were in English, whose applications were implemented in developed countries. This demonstrates the great void of projects and studies on technologies in developing countries, where there is little economic investment in technology and research.

In addition, the epidemiological studies that apply to each of the areas of artificial intelligence were identified, represented by observational, descriptive, analytical, predictive, and experimental studies.

When comparing this article with another research conducted on artificial intelligence in communicable diseases, we can observe that the total number of articles found according to the degree of agreement was 70, compared to 146 in this study; 16 studies were highly concordant compared to 28 in this research. The trend by areas of artificial intelligence focused mainly on automatic reasoning, computer vision, expert systems, and neural networks; in this study, the same result was obtained according to the approach.

In both studies, it could be observed that the area of neural networks is the least developed, the most published and very concordant studies were in the English language in developed countries and in the Medline database, which demonstrates the economic, educational, and research investment that these countries have made in comparison with developing countries.

In addition, it reflects the transformation and technological advances that these countries have made in different areas. The areas of health where there was greater development were public health and epidemiology in infectious diseases, unlike this study, which was in the area of internal medicine.

On the other hand, it made it possible to identify which clinical issues have been evaluated and do not require further research, as well as which texts or topics do not require further research. It also allowed professionals to keep up to date on current trending topics.

We were able to assess the consistency of studies and explore the main sources of variability in studies with apparently beneficial results. Different predictive models of morbidity were identified.

The quality assessment of the body of evidence obtained was performed by two reviewers or peers independently, and the differences in the results of the evaluation of each aspect were discussed until a consensus was reached. The differences between the peers between each of the filters were minimal.
