Investigating the level of consciousness with artificial intelligence, a narrative review

Background and aims: The occurrence of a disorder in the level of consciousness may be relatedto one of the following four causes. Reduction of brain oxygenation (hypoxia and reduction ofblood supply), neurological damage (stroke and stroke), metabolic disorders (heart failure, liverand kidney failure and diabetic ketoacidosis) and toxicological (high dose of medicine, alcoholor drug poisoning). Many instruments have been developed to assess the level of alertness, whichmostly not only measure different aspects of alertness, but also provide more reliable informationthat helps identify potential problems. Artificial Intelligence is a general term that implies the useof a computer to model intelligent behavior with minimal human intervention. Artificial NeuralNetworks (ANNs) are essential tools in machine learning that have attracted increasing attention.Structure, and learning of ANNs to better address a wide range of challenges in brain research.The processing capabilities of artificial intelligence are not far from the processes in the brain. Artificialintelligence uses the process of deep learning to perform computational problems quicklyand uses a network of layered algorithms that are interconnected to solve more and more complexproblems. Considering the extraordinary capabilities of artificial intelligence, the purpose of thisreview is to know the different capacities of artificial intelligence that can help to automaticallydetect the level of consciousness.Method: the main aim of this article is to collect data on the various theories of consciousnesspublished between ۲۰۰۰-۲۰۲۳ and to synthesize them to provide a general overview of this topic.We performed a literature review using PubMed and Google Scholar for relevant articles.Keywords include artificial intelligence, disorders of consciousness, and level of consciousness,and thus, we initially found ۳۱ studies as relevant studies in PubMed and Google scholar. Next,the title and abstract of more relevant studies were screened and the full text of ۱۱ studies wasreviewed.Results: Finally, ۷ studies met the inclusion criteria. It seems that consciousness research that hassought to use artificial intelligence has grown exponentially in recent years. Most of the studiesused the features extracted from the EEG signal to build a predictive model of consciousness.Currently, most researches in the field of consciousness medicine use artificial intelligence algorithmssuch as machine learning (ML) and perceptron (one of the artificial intelligence algorithms)to predict the pattern of differential brain activity in the level of consciousness.Conclusion: Artificial intelligence is a practical tool to care for the patient and to increase thediagnostic capabilities of the level of consciousness and to perform high-sensitivity therapeuticand surgical methods. The deep learning method has a high ability to model and diagnose thepatient’s symptoms and can identify the symptoms more accurately and provide them to the experts.However, in the field of health, more data is needed for the application of perceptron andDL algorithms. Also, using these tests before using them in medical centers requires more studies.

کلیدواژه ها:

Artificial intelligence ، Consciousness ، Electroencephalogram ، Consciousness disorders