An Effective Occipitomental View Enhancement Based on Adaptive Morphological Texture Analysis

Area of Science:

• Medical Imaging
• Radiology
• Computer-Aided Diagnosis

Background:

• Accurate visualization of maxillary sinuses in occipitomental radiographs is crucial for diagnosis.
• Existing methods may lack sufficient detail for precise interpretation.
• Computed tomography (CT) is the gold standard but is costly and involves radiation exposure.

Purpose of the Study:

• To develop and evaluate a novel contrast enhancement algorithm for maxillary sinuses in skull X-rays (SXR).
• To improve diagnostic accuracy using adaptive morphological texture analysis.
• To provide a cost-effective alternative to CT for sinus evaluation.

Main Methods:

• Skull X-ray (SXR) decomposition into rotational blocks (RBs).
• Morphological kernel processing to extract dark and bright features.
• Gradient-based block segmentation and feature block (FB) creation.
• Local histogram equalization and overlay onto the input SXR.

Main Results:

• The proposed method demonstrated significant enhancement of maxillary sinus visibility.
• Evaluation on 145 occipitomental view SXR images showed improved diagnostic accuracy.
• Achieved an 83.45% increase in diagnosis accuracy compared to CT.

Conclusions:

• The novel adaptive morphological texture analysis algorithm effectively enhances maxillary sinuses in SXR.
• This technique offers a promising approach to improve diagnostic accuracy in radiology.
• The method presents a viable, high-accuracy alternative to CT for sinus imaging.