{
  "id": 9444080,
  "title": "Tumor digital twins took months to build—AI now drafts them in minutes",
  "url": "https://urgent.news/2026/09/23/tumor-digital-twins-took-months-to-build-ai-now-drafts-them-in-minutes",
  "topic": "ai",
  "section": "AI",
  "published": "2026-09-23T23:40:06.000Z",
  "source": {
    "name": "Medical Xpress",
    "slug": "medical-xpress",
    "url": "https://medicalxpress.com/news/2026-09-tumor-digital-twins-months-ai.html"
  },
  "original_language": "en",
  "account": "Researchers at the Barcelona Supercomputing Center—Centro Nacional de Supercomputación (BSC-CNS) have developed a groundbreaking technology that leverages artificial intelligence (AI) to create digital models of tumors in minutes, rather than months of specialized work. Traditionally, building a digital tumor model required extensive literature review, training in specialized software, and proficiency in multiple programming languages, often taking years to complete.\n\nThe BSC-CNS's innovative approach involves connecting AI agents with specialized tools, enabling researchers to construct digital tumor models by conversing with an AI agent equipped with advanced capabilities. This eliminates the need for coding expertise, allowing researchers to focus on the biological problem they wish to explore rather than the technical details of model creation.\n\nBy using this new system, researchers can describe their biological inquiry to an AI agent, which then interacts with tools like NeKo, MaBoSS, and PhysiCell to generate a preliminary digital twin of the tumor. This process reduces setup time dramatically, with researchers able to draft an initial model in under 10 minutes, provided they have sufficient background knowledge in the field.\n\nThe study, published in the journal npj Systems Biology and Applications, also highlights the importance of iterative interaction with AI agents to ensure consistent, reliable results. While different language models may yield varied outputs, consistent refinement through repeated questioning ultimately leads to accurate and scientifically valid findings.\n\nThe open-source MCP servers developed by the BSC researchers are now available to the scientific community, making advanced biological modeling tools accessible to a wider range of researchers. This democratization of technology promises to accelerate treatment development and enhance our understanding of complex diseases like cancer. By integrating AI agents with domain expertise, the researchers believe this breakthrough will significantly reduce the time and effort required to create digital tumor models, ultimately contributing to more efficient and effective biomedical research.",
  "summary": "Until now, building a digital model of a tumor required scientists to review extensive literature, train in specialized software and master multiple programming languages. From start to finish, the process often took months, if not years, of specialized work.",
  "key_points": [],
  "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."
}