Intelligent Reconstruction Algorithm-Based Computed Tomography Images for Automatic Detection of Gastric Tumor

Area of Science:

• Medical Imaging
• Oncology
• Radiology

Background:

• Gastric tumors pose a significant diagnostic challenge.
• Accurate imaging is crucial for timely diagnosis and treatment planning.
• Traditional CT imaging may have limitations in visualizing subtle gastric lesions.

Purpose of the Study:

• To evaluate the efficacy of computed tomography (CT) images analyzed with the intelligent reconstruction algorithm (IRA), specifically the Feldkamp-Davis-Kress (FDK) algorithm, for diagnosing gastric tumors.
• To assess the detection rates and diagnostic accuracy of CT scans in differentiating benign and malignant gastric lesions.

Main Methods:

• 120 gastric cancer patients underwent CT scanning.
• CT images were analyzed using the Feldkamp-Davis-Kress (FDK) intelligent reconstruction algorithm (IRA).
• Imaging features of gastric lesions were evaluated and correlated with biopsy or surgical pathology results.

Main Results:

• The study identified three pathological types of benign tumors (polyps, leiomyomas, mesenchymomas) and three of malignant tumors (mesenchymomas, adenomas, lymphomas).
• CT scans achieved high detection rates: 94.2% for different stomach orientations, 90.7% for benign tumors, and 90.9% for malignant tumors.
• Significant differences in CT imaging features were observed between benign and malignant gastric tumors.

Conclusions:

• CT images processed with the FDK IRA demonstrate high detection rates and accurate localization of gastric lesions.
• This method effectively displays lesion characteristics and surrounding tissue metastasis, aiding in differentiating benign from malignant tumors.
• The application of CT imaging with FDK IRA is valuable for clinical diagnosis and promotion in gastric tumor management.