AI‐ and IoT‐Based Face Recognition Model for Identification of Human Diseases

Summary

Artificial intelligence (AI) technology is widely used in a variety of medical healthcare disciplines, including the diagnosis of various diseases based on facial characteristics. However, there is no evaluation or quantitative synthesis of AI performance on analysis of facial texture and pattern detection. Meta‐regression revealed that facial recognition intensity (FRI) plays an important role in varied diagnosis accuracy, and subgroup analysis revealed a similar finding. Even though no statistically significant association was found between accuracy and photographic resolution, training size, AI architecture, or the number of diseases, an appropriate increase in the training size and the use of deep learning models helped to improve diagnostic accuracy for diseases with low FRI. Furthermore, a fresh hypothesis for universal rules in AI performance is proposed, providing a new concept that could be explored in additional AI application. A new index, the facial recognition intensity (FRI), was established to describe the complexity of the association of diseases with facial phenotypes. Traditional recognition algorithms seem unable to identify specific facial traits, resulting in reduced facial recognition accuracy. Traditional face recognition systems are unable to predict precise facial traits, resulting in lower facial identification accuracy. The computer complexity of the false facial point detection technique is maximized. To anticipate and match the face from the database, an effective artificial intelligence internet of things (AIoT)‐based facial expression detection system is created in this study.