[Research status of automatic localization of acupoint based on deep learning]

Area of Science:

• Biomedical Engineering
• Artificial Intelligence
• Acupuncture Research

Background:

• Traditional acupoint localization relies on anatomical landmarks, which can be subjective and imprecise.
• Deep learning offers a promising approach for automating and standardizing acupoint localization.
• Recent advancements have focused on applying deep learning techniques to this field.

Purpose of the Study:

• To review and summarize recent deep learning applications in automatic acupoint localization.
• To identify key challenges and limitations in current deep learning-based acupoint localization methods.
• To outline future research directions for enhancing acupoint localization accuracy and personalization.

Main Methods:

• Systematic review of published literature on deep learning for acupoint localization.
• Analysis of studies focusing on dataset construction, neural network model design, and accuracy evaluation.
• Identification of trends and gaps in the application of deep learning in this domain.

Main Results:

• Significant progress has been achieved in deep learning-based acupoint localization.
• Current methods face challenges in dataset scale, precision, generalization, and real-time performance.
• Key areas for improvement include standardized datasets and integration of 3D modeling and multimodal data fusion.

Conclusions:

• Deep learning shows great potential for accurate and personalized acupoint localization.
• Further research is needed to address current limitations and enhance model performance.
• Standardized datasets and advanced techniques like 3D modeling are crucial for future development.