Reconstruction Algorithm-Based CT Imaging for the Diagnosis of Hepatic Ascites

Area of Science:

• Radiology
• Medical Imaging
• Artificial Intelligence

Background:

• Hepatic ascites diagnosis can be challenging, requiring effective etiological evaluation.
• Computed Tomography (CT) imaging is crucial for diagnosing abdominal effusion.
• Novel CT reconstruction algorithms may improve diagnostic accuracy.

Purpose of the Study:

• To evaluate the diagnostic value of an artificial intelligence reconstruction algorithm (AIHT) combined with CT image parameters for hepatic ascites.
• To provide a reference for the etiological evaluation of clinical abdominal effusion.
• To compare CT parameters between hepatic ascites and cancerous peritoneal effusion.

Main Methods:

• Proposed an adaptive iterative hard threshold (AIHT) algorithm for CT image reconstruction.
• Selected 100 patients with peritoneal effusion, divided into hepatic ascites (S1 group, n=50) and cancerous peritoneal effusion (D0 group, n=42) after exclusions.
• Performed Gemstone energy spectrum CT scanning and analyzed CT parameters including CT values, effective atomic number, water (iodine) and water (calcium) concentrations, and spectral curve slope.

Main Results:

• CT values (mixed energy, 60-100 KeV), water (calcium) concentration, water (iodine) concentration, and spectral curve slope were significantly lower in the hepatic ascites group (S1) compared to the cancerous effusion group (D0).
• Effective atomic number was significantly higher in the S1 group than in the D0 group.
• AIHT algorithm demonstrated reduced root mean square error (RMSE) and improved peak signal-to-noise ratio (PSNR) in reconstructed CT images compared to traditional algorithms.
• Water (iodine) and water (calcium) showed high diagnostic performance with sensitivities of 0.927 and 0.863, and specificities of 0.836 and 0.887, respectively.

Conclusions:

• AIHT-based CT images enhance the visualization of hepatic ascites.
• CT parameters such as CT value, effective atomic number, water (iodine), water (calcium), and spectral curve slope aid in identifying hepatic ascites.
• Water (iodine) and water (calcium) concentrations are particularly effective in the diagnosis of hepatic ascites, offering high diagnostic performance.