Pathology, research and practice · 2025
PESI-MS combined with AI to build a prediction model for lymph node metastasis of papillary thyroid cancer.
СтатьяНазвание и аннотация: Перевод пока не подготовлен
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Huang Qixin, Chen Zhenhe, Zhao Lingyu, Jiang Lichao, Wang Ye, Feng Qianqian, Lei Yajuan, Li Xiaodong, Zhong Dingrong
Аннотация
OBJECTIVE: Construct a prediction model for lymph node metastasis (LNM) in papillary thyroid carcinoma (PTC) using Probe Electrospray Ionization Mass Spectrometry (PESI - MS) combined with artificial intelligence (AI), to assist in the preoperative prediction of lymph node metastasis in thyroid carcinoma by intraoperative frozen pathology.
METHODS: A total of 78 fresh tissue samples of PTC and their adjacent normal tissues were collected. After proper processing, these samples were subjected to detection and analysis using PESI - MS. Subsequently, a classification prediction model was established based on the mass spectrometry test results integrated with AI algorithms. Support vector machine (SVM), random forest (RF), multi - layer perceptron (MLP), and Gradient boosting classifier (GBC) were employed for model building. Employing Support Vector Machine (SVM), Random Forest (RF), and Multilayer Perceptron (MLP) to conduct a single-blinded test on 10 independent PTC samples with unknown lymph node metastasis status.
RESULTS: The SVM, and MLP algorithms achieved an accuracy of 100 % in differentiating PTC with or without LNM, while the RF and GBC algorithm reached an accuracy of 92 %. All four algorithms demonstrated an accuracy of 100 % in distinguishing PTC from adjacent normal tissues.
CONCLUSION: The combination of PESI - MS and AI exhibits high accuracy in predicting LNM in PTC and performs remarkably well in the rapid diagnosis of PTC. This approach can effectively assist in the rapid diagnosis of intraoperative pathology, assist in determining the surgical scope of thyroid lymph node dissection, and offer more precise treatment for patients.
METHODS: A total of 78 fresh tissue samples of PTC and their adjacent normal tissues were collected. After proper processing, these samples were subjected to detection and analysis using PESI - MS. Subsequently, a classification prediction model was established based on the mass spectrometry test results integrated with AI algorithms. Support vector machine (SVM), random forest (RF), multi - layer perceptron (MLP), and Gradient boosting classifier (GBC) were employed for model building. Employing Support Vector Machine (SVM), Random Forest (RF), and Multilayer Perceptron (MLP) to conduct a single-blinded test on 10 independent PTC samples with unknown lymph node metastasis status.
RESULTS: The SVM, and MLP algorithms achieved an accuracy of 100 % in differentiating PTC with or without LNM, while the RF and GBC algorithm reached an accuracy of 92 %. All four algorithms demonstrated an accuracy of 100 % in distinguishing PTC from adjacent normal tissues.
CONCLUSION: The combination of PESI - MS and AI exhibits high accuracy in predicting LNM in PTC and performs remarkably well in the rapid diagnosis of PTC. This approach can effectively assist in the rapid diagnosis of intraoperative pathology, assist in determining the surgical scope of thyroid lymph node dissection, and offer more precise treatment for patients.