دورية أكاديمية
أعرض في EDS

High-Risk HPV Cervical Lesion Potential Correlations Mining over Large-Scale Knowledge Graphs

التفاصيل البيبلوغرافية
العنوان: High-Risk HPV Cervical Lesion Potential Correlations Mining over Large-Scale Knowledge Graphs
المؤلفون: Tiehua Zhou, Pengcheng Xu, Ling Wang, Yingxuan Tang
المصدر: Applied Sciences, Vol 14, Iss 6, p 2456 (2024)
بيانات النشر: MDPI AG, 2024.
سنة النشر: 2024
المجموعة: LCC:Technology
LCC:Engineering (General). Civil engineering (General)
LCC:Biology (General)
LCC:Physics
LCC:Chemistry
مصطلحات موضوعية: semantic biomedical informatics computing, data mining, high-risk HPV cervical lesion, disease prediction, subgraph mining, Technology, Engineering (General). Civil engineering (General), TA1-2040, Biology (General), QH301-705.5, Physics, QC1-999, Chemistry, QD1-999
الوصف: Lesion prediction, a very important aspect of cancer disease prediction, is an important marker for patients before they become cancerous. Currently, traditional machine learning methods are gradually applied in disease prediction based on patient vital signs data. Accurate prediction requires a large amount and high quality of data, however, the difficulty in obtaining and incompleteness of electronic medical record (EMR) data leads to certain difficulties in disease prediction by traditional machine learning methods. Secondly, there are many factors that contribute to the development of cervical lesions, some risk factors are directly related to it while others are indirectly related to them. In addition, risk factors have an interactive effect on the development of cervical lesions; it does not occur in isolation, a large-scale knowledge graph is constructed base on the close relationships among risk factors in the literature, and new potential key risk factors are mined based on common risk factors through a subgraph mining method. Then lesion prediction algorithm is proposed to predict the likelihood of lesions in patients base on the set of key risk factors. Experimental results show that the circumvents the problems of large number of missing values in EMR data and discovered key risk factors that are easily ignored but have better prediction effect. Therefore, The method had better accuracy in predicting cervical lesions.
نوع الوثيقة: article
وصف الملف: electronic resource
اللغة: English
تدمد: 2076-3417
Relation: https://www.mdpi.com/2076-3417/14/6/2456; https://doaj.org/toc/2076-3417
DOI: 10.3390/app14062456
URL الوصول: https://doaj.org/article/05da8f91c98e44c59b02344b947a27bb
رقم الأكسشن: edsdoj.05da8f91c98e44c59b02344b947a27bb
قاعدة البيانات: Directory of Open Access Journals
الوصف
تدمد:20763417
DOI:10.3390/app14062456