EFFICACY OF ARTIFICIAL INTELLIGENCE TO DIAGNOSE ANTERIOR SEGMENT DISEASES: A SCOPING REVIEW

OBJECTIVE: The purpose of the study is to find the efficacy of artificial intelligence in diagnosing anterior segment diseases.
METHODOLOGY:
Existing literature on efficacy of artificial intelligence or machine learning in diagnosing anterior segment ocular diseases from the year 2015 to 2020 were identified and summarized from 4 databases: Scopus, Pubmed, Google Scholar and Cochrane. All relevant articles were recorded.
RESULTS:
A total of one hundred fifty five studies were identified in the literature search. Out of which twenty two studies were analyzed. A majority of the twenty two studies demonstrated the efficiency of artificial intelligence to diagnose anterior segment diseases. Even though the methodology slightly varied from each research paper, all most all of the studies included in this review show a high sensitivity and specificity.
CONCLUSION
In this era of pandemic which ravages the whole world, a new interest grows towards diagnosis with minimal contact; to not only protect ourselves but also our patients. All published studies included in our review shows that sensitivity and specificity rate of machine learning is high and can be used for diagnosis of different anterior segment diseases.
KEYWORDS: Anterior segment of eye, Artificial Intelligence, Machine learning, Neural Network, Sensitivity, Specificity.