{
  "abstract": "Background Raman spectroscopy has emerged as a technology that can precisely associate tissue biomolecular alterations with histopathologic changes. We developed a novel fiber-optic probe Raman spectroscopy-based AI system (Spectra IMDx) that can provide real-time diagnosis of gastric lesions in vivo ( IDDF2026-ABS-0089 Figure 1. Lesion diagnosis based on WLE spectra IMDx and histopathology). This prospective two-center study compared Spectra IMDx’s performance in diagnosing gastric cancer (GC) and precancerous lesions with that of white light endoscopy (WLE).Methods Patients with suspected precancerous lesions or GC who were scheduled to undergo gastroscopy at the two centers were recruited. They underwent WLE, followed by Spectra IMDx of the identified lesions, and mucosal biopsies were paired with Raman spectral readings. With histopathological diagnosis as the gold standard, the sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of gastric lesion diagnosis using the IMDx were calculated.Results The sensitivity, specificity, PPV, NPV, and accuracy of the Spectra IMDx in diagnosing cancerous lesions were 88.00% (95% CI, 75.00-95.03), 83.5% (95% CI, 77.25-88.36), 58.67% (95% CI, 46.71-69.73), 96.32% (95% CI, 91.80-98.49), and 84.45% (95% CI, 79.30-55.50), respectively. Spectra IMDx performed significantly better than WLE in diagnosing gastric cancerous lesions in terms of sensitivity (88.00% vs. 56.00%, p<0.01), specificity (83.51% vs. 62.77%, p<0.01), accuracy (84.45% vs. 61.34%, p<0.01), PPV (58.67% vs. 28.57%, p<0.01), and NPV (96.32% vs. 84.29%, p<0.01).Conclusions In this proof-of-concept study, we showed that spectra IMDx exhibited excellent performance in diagnosing GC and precancerous lesions, with accuracy, sensitivity, and specificity all exceeding 80%. Spectra IMDx performed significantly better than WLE in diagnosing gastric lesions and EGC. While these results are encouraging, they need to be validated in larger-scale studies in the future (IDDF2026-ABS-0089 ­Figure 2. Value of IMDx in real-time diagnosis of multiple lesions in the gastric antrum).Abstract IDDF2026-ABS-0089 Figure 1Abstract IDDF2026-ABS-0089 Figure 2",
  "authors": [
    {
      "affiliations": [
        "Department of Gastroenterology, The National Key Clinical Specialty, Zhongshan Hospital of Xiamen University, School of Medicine, Xiamen University, Xiamen, Fujian Province, P.R. China, China"
      ],
      "name": "Kun Xia"
    },
    {
      "affiliations": [
        "Department of Gastroenterology, The National Key Clinical Specialty, Zhongshan Hospital of Xiamen University, School of Medicine, Xiamen University, Xiamen, Fujian Province, P.R. China, China"
      ],
      "name": "Xiaoning Yang"
    },
    {
      "affiliations": [
        "Department of Gastroenterology, The National Key Clinical Specialty, Zhongshan Hospital of Xiamen University, School of Medicine, Xiamen University, Xiamen, Fujian Province, P.R. China, China"
      ],
      "name": "Shuntian Cai"
    },
    {
      "affiliations": [
        "Department of Medicine, National University of Singapore, Singapore"
      ],
      "name": "Khek Yu Ho"
    },
    {
      "affiliations": [
        "Department of Medicine, National University of Singapore, Singapore"
      ],
      "name": "Calvin Jian-Yi Koh"
    },
    {
      "affiliations": [
        "Department of Gastroenterology, The National Key Clinical Specialty, Zhongshan Hospital of Xiamen University, School of Medicine, Xiamen University, Xiamen, Fujian Province, P.R. China, China"
      ],
      "name": "Jianlin Ren"
    },
    {
      "affiliations": [
        "Department of Gastroenterology, The National Key Clinical Specialty, Zhongshan Hospital of Xiamen University, School of Medicine, Xiamen University, Xiamen, Fujian Province, P.R. China, China"
      ],
      "name": "Hongzhi Xu"
    }
  ],
  "title": "IDDF2026-ABS-0089 The application of a novel raman spectroscopic probe-based AI in the real-time diagnosis of gastric neoplasia diagnosis: a proof-of-concept, multicenter, prospective cohort study",
  "uid": "fbb47f3d-1b86-5bd9-bfe2-3c4e4f122498"
}