Image Smart Segmentation Analysis Against Diabetic Foot Ulcer Using Internet of Things with Virtual Sensing.

Area of Science:

• Medical Imaging
• Computer Vision
• Internet of Things (IoT)

Background:

• Diabetic foot ulcer (DFU) poses a global health challenge, necessitating effective prevention strategies.
• Accurate image segmentation is vital for DFU identification, but current methods face issues like imprecision and incompleteness.
• Existing segmentation techniques often lack the reliability needed for clinical application.

Purpose of the Study:

• To introduce an advanced image segmentation analysis method for Diabetic Foot Ulcers (DFU) utilizing the Internet of Things (IoT) and virtual sensing.
• To address limitations of current DFU segmentation methods, aiming for higher accuracy, reliability, and deeper image analysis.
• To evaluate the proposed method's performance against existing techniques using multimodal datasets.

Main Methods:

• Implementation of a novel image segmentation analysis for DFU incorporating IoT and virtual sensing.
• Analysis of four range segmentation levels: region-based, edge-based, image-based, and CAD-based.
• Utilizing object co-segmentation for multimodal data compression and semantic segmentation.

Main Results:

• The proposed model demonstrated superior segmentation accuracy for DFU, achieving scores of 90.85% and 89.03% with virtual sensing.
• Significant improvements were observed compared to existing methodologies, with error rates reduced.
• The system showed a 59.1% improvement over deep segmentation techniques in live DFU studies, with notable gains in smart segmentation.

Conclusions:

• The developed method offers enhanced validity and reliability for DFU image segmentation analysis.
• The approach provides a more efficient and accurate tool for DFU identification and monitoring.
• The range-based segmentation achieved substantial interobserver reliability (73.9%) with minimal parameters.