# Improving early gastric cancer detection in upper gastrointestinal series adding digital tomosynthesis (DTS): a preliminary study

## TL;DR

This study shows that adding digital tomosynthesis to upper gastrointestinal radiography improves early gastric cancer detection, especially for less experienced radiologists.

## Contribution

Digital tomosynthesis significantly enhances early gastric cancer detection in UGI radiography, particularly for specific lesion types and less experienced readers.

## Key findings

- DTS improved detection rates for early gastric cancer from 71.0% to 83.9% for experienced readers and from 53.4% to 79.0% for less experienced readers.
- DTS significantly improved detection for 2–3 cm lesions and non-ulcerated elevated lesions, especially for less experienced readers.
- Inter-reader agreement improved from 0.71 with conventional radiography to 0.90 with DTS.

## Abstract

The aim of this study is to evaluate the effectiveness of digital tomosynthesis (DTS) for improving the detection rate of early gastric cancer (EGC) in upper gastrointestinal (UGI) radiography. We retrospectively analyzed 57 patients with 62 pathologically confirmed EGC lesions. Imaging accuracy was compared between conventional double-contrast UGI and UGI with DTS, using endoscopy as the gold standard.

In experienced readers, the detection rate improved from 71.0% with conventional radiography to 83.9% with DTS. For less experienced readers, the detection rate significantly increased from 53.4% to 79.0%. For lesions < 2 cm, DTS significantly improved detection only for less experienced readers (p < 0.05). For 2–3 cm lesions, both readers showed significant improvement (p < 0.05 and p < 0.01, respectively). No significant difference was found for lesions ≥ 3 cm. DTS also significantly improved detection for non-ulcerated lesions, particularly those with elevated structures, in the less experienced reader (p < 0.05).

AUC analysis showed a slight improvement for experienced readers (0.86 → 0.98, p = 0.0604), while less experienced readers demonstrated significant improvement (0.76 → 0.96, p < 0.01). Inter-reader agreement for conventional radiography was 0.71, which improved to 0.90 with DTS. The total combined radiation dose for UGI and DTS was 4.7–5.3 ± 0.5 mSv, within acceptable limits.

UGI with DTS significantly enhances EGC detection, particularly for lesions 2–3 cm and non-ulcerated elevated lesions, improving diagnostic accuracy and consistency, especially for less experienced radiologists. Given its minimal additional examination time (143 s) and cost-effectiveness compared to EGD, DTS may serve as a practical supplementary tool for gastric cancer screening and preoperative evaluation.