JoVE - Journal of Visualized Experiments
JoVE Visualize
Contact Us

PESI-MS combined with AI to build a prediction model for lymph node metastasis of papillary thyroid cancer

  • 0Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing 100006, China; Department of Pathology, China-Japan Friendship Hospital, Beijing 100029, China.
Pathology, Research and Practice +

|

April 24, 2025

|

April 24, 2025

View abstract on PubMed

Summary

This summary is machine-generated.

This study developed a Probe Electrospray Ionization Mass Spectrometry (PESI-MS) and artificial intelligence (AI) model to predict lymph node metastasis (LNM) in papillary thyroid carcinoma (PTC). The AI model achieved high accuracy, aiding rapid intraoperative diagnosis and surgical planning for PTC patients.

Area Of Science

  • Oncology
  • Analytical Chemistry
  • Computational Biology

Background

  • Accurate preoperative prediction of lymph node metastasis (LNM) in papillary thyroid carcinoma (PTC) is crucial for effective surgical planning and patient treatment.
  • Intraoperative frozen pathology assessments can be time-consuming, necessitating faster diagnostic methods.
  • Probe Electrospray Ionization Mass Spectrometry (PESI-MS) offers a rapid analytical technique for tissue characterization.

Purpose Of The Study

  • To construct a prediction model for LNM in PTC using PESI-MS combined with artificial intelligence (AI).
  • To evaluate the model's efficacy in assisting preoperative prediction of LNM via intraoperative frozen pathology.
  • To assess the diagnostic performance of AI algorithms in differentiating PTC with or without LNM.

Main Methods

  • Collected 78 fresh PTC tissue samples and adjacent normal tissues.
  • Analyzed samples using PESI-MS.
  • Developed classification prediction models using AI algorithms (SVM, RF, MLP, GBC) integrated with mass spectrometry data.

Main Results

  • Support Vector Machine (SVM) and Multilayer Perceptron (MLP) algorithms achieved 100% accuracy in differentiating PTC with or without LNM.
  • Random Forest (RF) and Gradient Boosting Classifier (GBC) algorithms reached 92% accuracy for LNM prediction.
  • All four AI algorithms demonstrated 100% accuracy in distinguishing PTC from adjacent normal tissues.

Conclusions

  • The combination of PESI-MS and AI demonstrates high accuracy in predicting LNM in PTC.
  • This approach facilitates rapid intraoperative diagnosis of PTC.
  • The model assists in determining the scope of thyroid lymph node dissection and enables more precise patient treatment.

Keywords: