Skeletal scintigraphy image enhancement based neutrosophic sets and salp swarm algorithm

Area of Science:

• Medical Imaging
• Computer Science
• Artificial Intelligence

Background:

• Existing methods for enhancing dark regions in skeletal scintigraphy images often suffer from performance issues.
• Effective enhancement is crucial for accurate diagnosis and interpretation of skeletal scintigraphy.

Purpose of the Study:

• To develop an efficient adaptive scheme for enhancing dark regions in skeletal scintigraphy images.
• To address the limitations of current image enhancement techniques in medical imaging.

Main Methods:

• An adaptive scheme combining the Salp Swarm Algorithm (SSA) and a neutrosophic set (NS) under multi-criteria was proposed.
• The SSA algorithm optimizes image enhancement parameters, while the NS algorithm calculates similarity scores with adaptive weights.
• The method was applied to a no-reference medical dataset from Menoufia University Hospital.

Main Results:

• The proposed SSA-NS algorithm demonstrated superior performance across various metrics, including fitness function, entropy, edge count, naturalness image quality, sharpness, and contrast enhancement.
• Comparative analysis against Gamma Correction, NS algorithm, and local enhancement algorithms confirmed the effectiveness of the integrated approach.
• The study validated the utility of integrating neutrosophic set falsity membership with SSA for skeletal scintigraphy enhancement.

Conclusions:

• The integration of the Salp Swarm Algorithm and neutrosophic set provides an effective solution for enhancing dark regions in skeletal scintigraphy.
• The proposed method significantly improves image quality, potentially aiding in more accurate medical diagnoses.
• This research highlights the potential of utilizing falsity membership within neutrosophic sets for advanced image enhancement applications in the medical field.