{
  "abstract": "Background: Postoperative sleep disturbance (POSD) is a common complication in elderly patients undergoing transurethral prostate surgery, which impairs recovery and increases disease burden. To date, prediction models for POSD in this population, particularly those based on machine learning techniques, remain scarce.\n\nObjective: To investigate the incidence and influencing factors of POSD in elderly patients undergoing transurethral prostate surgery, and to construct prediction models based on conventional logistic regression and seven machine learning algorithms.\n\nMethods: A total of 563 elderly patients who underwent elective transurethral prostate surgery at our hospital between March 2024 and December 2025 were prospectively enrolled and randomly divided into a training set (n = 338) and a validation set (n = 225) at a ratio of 6:4. LASSO regression (with lambda.1se) was used to screen predictive variables, and multivariate logistic regression to identify independent risk factors. Logistic regression, random forest (RF), extreme gradient boosting (XGBoost), naive Bayes, decision tree, support vector machine, k-nearest neighbors, and artificial neural network models were constructed. Model performance was evaluated using area under the receiver operating characteristic curve (AUC), accuracy, calibration curves, and decision curve analysis. SHAP (SHapley Additive exPlanations) was employed for interpretability analysis of the optimal model.\n\nResults: The incidence of POSD was 62.2%. Independent risk factors included preoperative anxiety (Hospital Anxiety and Depression Scale-Anxiety [HADS-A] score), nocturia frequency (≥3 times/night), elevated systemic immune-inflammation index (SII), preoperative noise score, and moderate-to-severe cough-related pain (P < 0.05). The random forest model demonstrated the best overall performance, with an AUC of 0.926 (95% CI: 0.89–0.96) in the training set and 0.964 (95% CI: 0.93–0.99) in the validation set, and exhibited superior calibration and clinical utility compared with the other models. SHAP analysis revealed that the HADS-A score contributed the most to the predictions.\n\nConclusions: Preoperative anxiety, increased nocturia frequency (≥3 times/night), elevated SII, high noise score, and moderate-to-severe cough-related pain are independent risk factors for POSD in elderly patients undergoing transurethral prostate surgery. The random forest prediction model exhibits optimal performance and can be used for early clinical risk identification and individualized intervention.\n\nClinical trial registration: Chinese Clinical Trial Registry ChiCTR2600124377.",
  "abstract_shown": true,
  "affiliations": [
    "Department of Anesthesiology, The First Affiliated Hospital of Guangxi Medical University, Nanning, China",
    "Department of Anesthesiology, The Third Affiliated Hospital of Guangxi Medical University, Nanning, China",
    "Department of Anesthesiology, Sichuan Clinical Research Center for Cancer, Sichuan Cancer Hospital & Institute, Sichuan Cancer Center, University of Electronic Science and Technology of China, Chengdu, China"
  ],
  "article_type": "research article",
  "arxiv": null,
  "author_total": 4,
  "authors": [
    {
      "affs": [
        0
      ],
      "name": "Jingxian Huang"
    },
    {
      "affs": [
        1
      ],
      "name": "Huageng Huang"
    },
    {
      "affs": [
        1
      ],
      "name": "Yuhua Zhou"
    },
    {
      "affs": [
        0,
        2
      ],
      "name": "Yubo Xie"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/bwf3d99t9vsea.bib",
  "cited_by": 0,
  "doi": "10.3389/fmed.2026.1962940",
  "fulltext_url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC13730954/",
  "id": "bwf3d99t9vsea",
  "issns": [
    "2296-858X"
  ],
  "landing_url": "https://doi.org/10.3389/fmed.2026.1962940",
  "language": "en",
  "licence": "CC-BY-4.0",
  "licence_label": "CC BY 4.0",
  "licence_source": "article",
  "licence_url": "https://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": null,
  "native_subjects": [
    [
      "heading",
      "Original Research"
    ],
    [
      "openalex",
      "2748|Urinary Bladder and Prostate Research"
    ]
  ],
  "openalex": "W7223118065",
  "pdf_url": "https://www.frontiersin.org/journals/medicine/articles/10.3389/fmed.2026.1962940/pdf",
  "pmcid": "PMC13730954",
  "pmid": "42984521",
  "published": "2026-10-21",
  "publisher": "Frontiers Media SA",
  "ris": "https://research.draconitic.com/paper/bwf3d99t9vsea.ris",
  "sources": [
    {
      "first_seen": "2026-11-07T12:23:55Z",
      "native_id": "PMC13730954",
      "source": "pmc",
      "url": "https://doi.org/10.3389/fmed.2026.1962940"
    }
  ],
  "subjects": [
    {
      "field": "medicine",
      "origin": "openalex:2748|Urinary Bladder and Prostate Research",
      "subfield": "general"
    }
  ],
  "title": "The influencing factors of postoperative sleep disturbance in elderly patients undergoing transurethral prostate surgery and the construction and validation of a machine learning prediction model",
  "title_text": "The influencing factors of postoperative sleep disturbance in elderly patients undergoing transurethral prostate surgery and the construction and validation of a machine learning prediction model",
  "type": "peer-reviewed",
  "updated_at": "2026-11-08T12:21:15Z",
  "url": "https://research.draconitic.com/paper/bwf3d99t9vsea",
  "venue": "Frontiers in Medicine",
  "versions": []
}