Scholarlink

<p>In the treatment of urinary stones, surgical intervention is crucial. Urinary stones composition and type directly affect surgical planning. However, research on preoperative stone composition analysis is limited. This paper aimed to predict urinary stones types preoperatively using clinical data. Data from 1020 patients, including stone composition, clinical biochemical indicators, and demographic information, were collected. A stone composition graph network was constructed using cosine similarity, with stone composition as nodes and biochemical/demographic data as node features. The <i>Louvain</i> community detection algorithm was utilized to divide the network into distinct communities for the classification of stone types, with the effectiveness of the partitioning evaluated by the <i>Modularity</i> score. Stone types were classified, and their distribution across genders and age groups was described. Clinical feature averages were calculated for each community, and patients were assigned to the most similar community. Six machine learning algorithms (<i>RandomForest</i>, <i>GradientBoosting</i>, <i>SVM</i>, <i>KNN</i>, <i>Logistic Regression</i>, <i>XGBoost</i>) were trained to predict stone types. Model performance was evaluated, and the importance of clinical features for prediction was ranked. Six stone types were identified (<i>Modularity</i> = 0.828), namely common COM (Class I), COM with minor AU (Class II), COM with high UA (Class III), COM containing MAP (Class IV), high CAP-MAP (Class V), and high COM-CAP containing DCPD (Class VI). Among males, Class III and Class I were most prevalent; among females, Class V and Class III were most prevalent (<i>χ</i><sup><i>2</i></sup> = 95.066,<i> P</i> &lt; 0.001). Patients with Class IV stones were significantly older than those with Class I stones (<i>P</i> = 0.038). GradientBoosting showed the best prediction performance, with an <i>Accuracy</i> of 0.837, <i>Precision</i> of 0.840, <i>Recall</i> of 0.8366, <i>F1</i> Score of 0.8368, and <i>ROC-AUC</i> area of 0.941. Significant clinical features for prediction included urine specific gravity, white blood cells, pH, and crystals. This paper first analyzed stone categories using a community detection algorithm and then predicted types using machine learning, providing a reference for preoperative surgical planning in urinary stones.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Preoperative classification of urinary stones based on community detection

  • 期刊论文

  • 发表日期：

    2025年3月7日

查看全文
  • Danhui Mao,
  • Hao Liu,
  • Qianshan Wang,
  • Mingyan Ma,
  • Mohan Zhang,
  • Juanjuan Zhao,
  • Xin Wang

摘要

In the treatment of urinary stones, surgical intervention is crucial. Urinary stones composition and type directly affect surgical planning. However, research on preoperative stone composition analysis is limited. This paper aimed to predict urinary stones types preoperatively using clinical data. Data from 1020 patients, including stone composition, clinical biochemical indicators, and demographic information, were collected. A stone composition graph network was constructed using cosine similarity, with stone composition as nodes and biochemical/demographic data as node features. The Louvain community detection algorithm was utilized to divide the network into distinct communities for the classification of stone types, with the effectiveness of the partitioning evaluated by the Modularity score. Stone types were classified, and their distribution across genders and age groups was described. Clinical feature averages were calculated for each community, and patients were assigned to the most similar community. Six machine learning algorithms (RandomForest, GradientBoosting, SVM, KNN, Logistic Regression, XGBoost) were trained to predict stone types. Model performance was evaluated, and the importance of clinical features for prediction was ranked. Six stone types were identified (Modularity = 0.828), namely common COM (Class I), COM with minor AU (Class II), COM with high UA (Class III), COM containing MAP (Class IV), high CAP-MAP (Class V), and high COM-CAP containing DCPD (Class VI). Among males, Class III and Class I were most prevalent; among females, Class V and Class III were most prevalent (χ2 = 95.066, P < 0.001). Patients with Class IV stones were significantly older than those with Class I stones (P = 0.038). GradientBoosting showed the best prediction performance, with an Accuracy of 0.837, Precision of 0.840, Recall of 0.8366, F1 Score of 0.8368, and ROC-AUC area of 0.941. Significant clinical features for prediction included urine specific gravity, white blood cells, pH, and crystals. This paper first analyzed stone categories using a community detection algorithm and then predicted types using machine learning, providing a reference for preoperative surgical planning in urinary stones.