AI-Powered Ultrasound Detection Enhances Endometrial Cancer Diagnosis in Postmenopausal Women

Strategic Insights -

Research conducted by a team of doctors from the Department of Gynecology at the Hospital of Traditional Chinese Medicine of Qiqihar has led to the development of an artificial intelligence-based ultrasound detecting model that significantly improves the diagnosis of endometrial cancer in postmenopausal women. The study aimed to develop a deep learning model to analyze comprehensive gynecological ultrasound images, specifically targeting the identification of endometrial cancer based on ultrasound features.

Key Takeaways:

  • The AI-powered model was developed to analyze 877 ultrasound images of patients diagnosed with endometrial cancer between January 1, 2020, and December 31, 2024.
  • The model was trained on 614 images, validated on 175 images, and tested on 88 images, demonstrating high accuracy and robustness.
  • The area under the receiver operating characteristic (ROC) curve for the training set was 0.844, while in the validation set, the AUC for predicting endometrial cancer was 0.811, and in the testing set, the AUC reached 0.858.
  • The study concluded that the AI-based model significantly enhances the ability to provide diagnostic assistance for endometrial cancer through ultrasound in postmenopausal women.
  • This research provides substantial clinical value, especially by enabling less experienced physicians in primary care settings to effectively detect endometrial cancer lesions, ensuring patients receive timely diagnosis and treatment.

Statistics:

  • 877 patients with endometrial cancer were enrolled in the study, with 877 ultrasound images analyzed.
  • The AI-powered model demonstrated high accuracy and robustness, with an area under the ROC curve of 0.844 for the training set, 0.811 for the validation set, and 0.858 for the testing set.
  • The study focused on patients diagnosed with endometrial cancer between January 1, 2020, and December 31, 2024.

Sources:

  • NewsRx. Research from Qiqihar Yields New Data on Artificial Intelligence (Improving the diagnosis of endometrial cancer in postmenopausal women in primary care settings using an artificial intelligence-based ultrasound detecting model). OBGYN & Reproduction Week. September 22, 2025; p 680.
  • Improving the diagnosis of endometrial cancer in postmenopausal women in primary care settings using an artificial intelligence-based ultrasound detecting model. Frontiers in Oncology, 2025,15. (Frontiers in Oncology - http://www.frontiersin.org/oncology)
  • Frontiers in Oncology. Publisher: Frontiers Media S.A. DOI: 10.3389/fonc.2025.1646826