{
  "id": 6344237,
  "title": "AI helping doctors to detect intestinal disease",
  "url": "https://urgent.news/2026/09/08/ai-helping-doctors-to-detect-intestinal-disease",
  "topic": "ai",
  "section": "AI",
  "published": "2026-09-08T23:20:08.000Z",
  "source": {
    "name": "Medical Xpress",
    "slug": "medical-xpress",
    "url": "https://medicalxpress.com/news/2026-09-ai-doctors-intestinal-disease.html"
  },
  "original_language": "en",
  "account": "This article explores the use of artificial intelligence (AI) to assist doctors in detecting intestinal polyps, which could potentially lead to early detection and treatment of bowel cancer. Researchers at NTNU in Gjøvik collaborated with doctors and researchers at Gjøvik Hospital to develop and test AI-generated images of intestinal polyps.\n\nA significant challenge in detecting polyps is the large volume of data produced, even with capsule endoscopy, which involves patients swallowing a small camera that takes tens of thousands of images over six to seven hours. AI is needed to process this data, but sufficient traditional images are required for effective training. To overcome this, the researchers used diffusion models and the CLIP machine-learning method to generate artificial images of polyps.\n\nThese synthetic images were compared to real images obtained from colonoscopies and capsule endoscopies of the same patients. The doctors were able to distinguish between real and artificial images 54% of the time, suggesting that the AI-generated images are of comparable quality. This finding indicates that AI can potentially help reduce the number of examinations and analysis work required by doctors, making the process more efficient and consistent across different specialists.\n\nThe ultimate goal of using AI-generated images is not to replace current examinations but to streamline the process and help doctors prioritize which images to examine. This could potentially reduce the time required for image analysis from three to four days per patient to just one day. Additionally, the pill camera used for capsule endoscopy is less uncomfortable for patients compared to traditional colonoscopies. Furthermore, the AI-generated images could be particularly useful in detecting bowel cancer, which is the most common form of cancer in Norway, with around 5,000 new diagnoses each year.",
  "summary": "One of the above images is real; the other is AI-generated. Can you tell which is which? The images depict intestinal polyps, often benign growths on the inside of the colon or rectum. Polyps are usually removed because some can develop into bowel cancer over time.",
  "key_points": [
    "Researchers at NTNU and Gjøvik Hospital developed AI-generated images of intestinal polyps.",
    "AI aims to streamline process, reduce time for image analysis to one day per patient."
  ],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}