Artificial Intelligence for Diabetic Foot Screening Based on Digital Image Analysis: A Systematic Review

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Diabetic Complications

Background:

• Diabetic foot complications require early detection for effective management.
• Artificial intelligence (AI) using digital image analysis presents a noninvasive screening method.
• This systematic review focuses on AI model development for diabetic foot screening.

Purpose of the Study:

• To systematically review studies on AI model development for diabetic foot screening.
• To identify the effectiveness of AI in analyzing digital foot images for early detection of complications.

Main Methods:

• Systematic review of articles published between 2018-2023 from PubMed, ProQuest, and ScienceDirect.
• Inclusion criteria applied to 2214 initially identified articles, resulting in nine selected studies.
• Quality assessment using QUADAS; data extraction and analysis via NVivo.

Main Results:

• Thermal imagery (foot thermograms) and plantar temperature patterns are key data sources.
• Deep learning methods, including artificial neural networks (ANNs) and convolutional neural networks (CNNs), are prevalent.
• An ANN model achieved 97.5% accuracy in classifying macula types, demonstrating high performance.

Conclusions:

• AI holds significant potential to enhance the accuracy and efficiency of diabetic foot screening.
• Future research should address clinical applicability, ethical considerations, and data security.
• Development of comprehensive datasets is crucial for advancing AI in diabetic foot care.