Detection of gastrointestinal foreign bodies in pets using single-grid-based dark-field X-ray imaging

Abstract

Gastrointestinal foreign bodies (GI-FBs) occur when pets consume non-digestible items that do not readily pass through the stomach or intestines. While some GI-FBs pass through, many become lodged along the gastrointestinal tract and cause discomfort, often leading to sickness. Conventional absorption-based radiography is widely used to detect GI-FBs in pets. However, detecting low-density FBs, such as wood, plastic, clothing, and sticks, is typically difficult using conventional radiography. This study presents a novel approach for detecting low-density GI-FBs in pets by using single-grid-based dark-field X-ray imaging (SG-DFXI). It obtains microstructural information from small-angle X-ray scattering (SAXS) of the sample using a conventional X-ray source and grid. SG-DFXI requires minimal exposure and system setup and is specifically utilized to detect low-density materials that are invisible when using conventional radiography. To validate the efficacy of the proposed method, an experiment was conducted on a mouse phantom attached to a wooden chopstick as a low-density GI-FB. Quantitative evaluation was performed using image quality metrics of the contrast-to-noise ratio (CNR) and relative contrast gain (RCG). The CNR value measured in the dark-field image obtained with an autocorrelation length of ξG = 294 nm was 4.59, approximately 5.4 times larger than that of the absorption image obtained in the same imaging setup. In addition, the RCG characteristics improved as the autocorrelation length of the system increased; the RCG value for ξG = 294 nm was 5.4, approximately 3.2 times larger than that for ξG = 194 nm. Thus, increasing the autocorrelation length of the system is critical to improve its ability to detect low-density FBs. Consequently, SG-DFXI appears to be a promising method that can be used effectively and easily to detect GI-FBs in pets, which are barely visible in absorption images.