{
  "abstract": "Background: Accurate short-term glucose forecasting is important for diabetes self-management, yet dietary information remains difficult to quantify and integrate into forecasting models in free-living settings, especially for complex Chinese meals. Methods: We investigated whether large language models (LLMs) can convert free-text Chinese dietary records into structured nutritional features for short-term continuous glucose forecasting. Five Chinese-capable LLMs were evaluated as nutrient estimators, and the resulting nutritional features were integrated with continuous glucose monitoring histories in a seasonal-patch transformer for glucose forecasting. Performance was assessed in 2 real-world cohorts (ShanghaiT1DM and ShanghaiT2DM) for both full-period and postprandial glucose forecasting. Results: LLM-derived nutrient estimates significantly outperformed semantic embeddings as dietary representations for glucose forecasting (P < 0.001). In the primary comparison setting, the best-performing LLM-based model achieved a postprandial 3-h root mean square error (RMSE) of 34.668 mg/dl, reducing the error by 1.967 mg/dl compared to non-LLM settings. In the full-period setting, the optimal LLM-based model achieved an RMSE of 27.035 mg/dl, reducing the error by 0.42 mg/dl relative to non-LLM settings. Across both forecasting tasks, direct nutrient estimation by DeepSeek-v3.1 achieved the best average RMSE. Forecasting gains were positively associated with nutrient-estimation quality, particularly for carbohydrate, dietary fiber, and fat. Architecture ablations further showed that patch-based encoding was the most influential component, with additional contributions from temporal decomposition and side branch design. Conclusion: These findings show that LLM-based nutrient estimation provides a practical and physiologically interpretable way to convert free-text dietary records into structured meal-related inputs for glucose forecasting.",
  "abstract_shown": true,
  "affiliations": [
    "Institute of Medical Information, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing, China."
  ],
  "article_type": "research article",
  "arxiv": null,
  "author_total": 5,
  "authors": [
    {
      "affs": [
        0
      ],
      "name": "Leyao Ma",
      "orcid": "0000-0003-2645-6754"
    },
    {
      "affs": [
        0
      ],
      "name": "Jie Hao",
      "orcid": "0009-0002-7208-7443"
    },
    {
      "affs": [
        0
      ],
      "name": "Lin Yang",
      "orcid": "0000-0001-9211-0079"
    },
    {
      "affs": [
        0
      ],
      "name": "Xiaowei Xu"
    },
    {
      "affs": [
        0
      ],
      "name": "Jiao Li",
      "orcid": "0000-0001-6391-8343"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/378nkxpz2ehz4.bib",
  "cited_by": 0,
  "doi": "10.34133/hds.0471",
  "fulltext_url": "https://europepmc.org/articles/PMC13714320",
  "id": "378nkxpz2ehz4",
  "issns": [
    "2097-1095",
    "2765-8783"
  ],
  "landing_url": "https://europepmc.org/article/MED/42953316",
  "language": "eng",
  "licence": "CC-BY",
  "licence_label": "CC BY",
  "licence_source": "article",
  "licence_url": "https://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": null,
  "native_subjects": [
    [
      "heading",
      "Research Article"
    ],
    [
      "openalex",
      "2712|Diabetes Management and Research"
    ]
  ],
  "openalex": "W7155367875",
  "pdf_url": "https://europepmc.org/articles/PMC13714320?pdf=render",
  "pmcid": "PMC13714320",
  "pmid": "42953316",
  "published": "2026-10-28",
  "publisher": "AAAS Science Partner Journal Program",
  "ris": "https://research.draconitic.com/paper/378nkxpz2ehz4.ris",
  "sources": [
    {
      "first_seen": "2026-10-31T00:20:40Z",
      "native_id": "MED:42953316",
      "source": "europepmc",
      "url": "https://europepmc.org/article/MED/42953316"
    },
    {
      "first_seen": "2026-11-01T12:21:50Z",
      "native_id": "PMC13714320",
      "source": "pmc",
      "url": "https://doi.org/10.34133/hds.0471"
    }
  ],
  "subjects": [
    {
      "field": "medicine",
      "origin": "openalex:2712|Diabetes Management and Research",
      "subfield": "endocrinology-metabolism"
    }
  ],
  "title": "Short-Term Continuous Glucose Forecasting with Large-Language-Model-Derived Nutrient Estimates from Real-World Chinese Dietary Records.",
  "title_text": "Short-Term Continuous Glucose Forecasting with Large-Language-Model-Derived Nutrient Estimates from Real-World Chinese Dietary Records.",
  "type": "peer-reviewed",
  "updated_at": "2026-11-01T12:21:50Z",
  "url": "https://research.draconitic.com/paper/378nkxpz2ehz4",
  "venue": "Health data science",
  "versions": []
}