Liver microvascular flow evaluation by texture analysis and quantitative contrast-enhanced ultrasound in pediatric chronic liver disease.

Abstract

The liver's intricate vascular network is pivotal to its function, requiring a comprehensive understanding for improved diagnostic and treatment strategies in liver diseases. In this pilot study we explore the efficacy of contrast-enhanced ultrasound in identifying microvascular flow in pediatric chronic liver disease (CLD), highlighting the need for objective tools to gain deeper insights. We focused on two novel methods: microvasculature texture analysis and quantitative microvascular flow analysis. CEUS liver images were acquired from 8 patients with CLD related to different etiologies. These images underwent texture analysis, incorporating techniques like first-order histogram analysis, gray-level co-occurrence matrices, and run-length matrix analysis. We also employed a delta projection method to assess microvascular perfusion dynamics. For comparative purposes, liver CEUS images from four healthy individuals were also analyzed. Our texture analysis showed distinct vessel arrangement, and heterogeneity patterns with different liver diseases. For example, patients with primary sclerosing cholangitis exhibited average heterogeneity values of 1411.9. In contrast, those with biliary atresia had different echogenicity heterogeneity (1149.3). The delta projection technique further differentiated microvascular perfusion patterns across liver diseases, with each condition displaying unique perfusion index (PI) and peak enhancement (PE) values. Patients with primary sclerosing cholangitis exhibited comparable perfusion index (PI) values (3.2-3.5) and peak enhancement (PE) values (49.8-53.5). Biliary atresia cases perfusion values were (Pl of 1.3-1.7 and a PE of 40-45.4). Microvasculature texture analysis use is feasible, and with quantitative CEUS analysis, can provide valuable insights into microvascular patterns, enhancing diagnostic precision for pediatric CLD.