REASON: Probability map-guided dual-branch fusion framework for gastric content assessment

TL;DR

REASON is a two-stage deep learning framework that uses probability maps and dual-view fusion to accurately assess gastric content from ultrasound, improving efficiency and precision over traditional methods.

Contribution

It introduces a novel two-stage framework combining probability map-guided segmentation and dual-view classification for gastric content assessment.

Findings

Outperforms state-of-the-art methods significantly

Demonstrates robustness and accuracy on a self-collected dataset

Offers potential for automated clinical preoperative assessment

Abstract

Accurate assessment of gastric content from ultrasound is critical for stratifying aspiration risk at induction of general anesthesia. However, traditional methods rely on manual tracing of gastric antra and empirical formulas, which face significant limitations in both efficiency and accuracy. To address these challenges, a novel two-stage probability map-guided dual-branch fusion framework (REASON) for gastric content assessment is proposed. In stage 1, a segmentation model generates probability maps that suppress artifacts and highlight gastric anatomy. In stage 2, a dual-branch classifier fuses information from two standard views, right lateral decubitus (RLD) and supine (SUP), to improve the discrimination of learned features. Experimental results on a self-collected dataset demonstrate that the proposed framework outperforms current state-of-the-art approaches by a significant margin. This framework shows great promise for automated preoperative aspiration risk assessment, offering a more robust, efficient, and accurate solution for clinical practice.