Related Experiment Video
Updated: Sep 27, 2025
Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
Published on: August 16, 2020
A Weakly Supervised Deep Learning Method for Guiding Ovarian Cancer Treatment and Identifying an Effective Biomarker
Ching-Wei Wang1,2, Yu-Ching Lee2, Cheng-Chang Chang3,4
1Graduate Institute of Biomedical Engineering, National Taiwan University of Science and Technology, Taipei 106335, Taiwan.
Abstract:
Ovarian cancer is a common malignant gynecological disease. Molecular target therapy, i.e., antiangiogenesis with bevacizumab, was found to be effective in some patients of epithelial ovarian cancer (EOC). Although careful patient selection is essential, there are currently no biomarkers available for routine therapeutic usage. To the authors’ best knowledge, this is the first automated precision oncology framework to effectively identify and select EOC and peritoneal serous papillary carcinoma (PSPC) patients with positive therapeutic effect. From March 2013 to January 2021, we have a database, containing four kinds of immunohistochemical tissue samples, including AIM2, c3, C5 and NLRP3, from patients diagnosed with EOC and PSPC and treated with bevacizumab in a hospital-based retrospective study. We developed a hybrid deep learning framework and weakly supervised deep learning models for each potential biomarker, and the experimental results show that the proposed model in combination with AIM2 achieves high accuracy 0.92, recall 0.97, F-measure 0.93 and AUC 0.97 for the first experiment (66% training and 34%testing) and high accuracy 0.86 ± 0.07, precision 0.9 ± 0.07, recall 0.85 ± 0.06, F-measure 0.87 ± 0.06 and AUC 0.91 ± 0.05 for the second experiment using five-fold cross validation, respectively. Both Kaplan-Meier PFS analysis and Cox proportional hazards model analysis further confirmed that the proposed AIM2-DL model is able to distinguish patients gaining positive therapeutic effects with low cancer recurrence from patients with disease progression after treatment (p < 0.005).
Insights
This study introduces an AI framework to identify ovarian cancer patients who will benefit from bevacizumab therapy using AIM2 biomarkers. The AI model accurately predicts positive therapeutic effects, aiding in personalized cancer treatment.
Area of Science:
- Oncology
- Biomarkers
- Artificial Intelligence
Background:
- Ovarian cancer (EOC) is a prevalent gynecological malignancy.
- Bevacizumab (antiangiogenesis) shows efficacy in some EOC patients, but lacks reliable biomarkers for selection.
- Accurate patient selection is crucial for optimizing bevacizumab therapy.
Purpose of the Study:
- To develop and validate an automated precision oncology framework for identifying epithelial ovarian cancer (EOC) and peritoneal serous papillary carcinoma (PSPC) patients who will respond positively to bevacizumab treatment.
- To investigate the potential of AIM2, c3, C5, and NLRP3 as immunohistochemical biomarkers for predicting bevacizumab efficacy.
- To establish a novel deep learning model for biomarker-based patient stratification.
Main Methods:
- A retrospective study analyzed immunohistochemical tissue samples (AIM2, c3, C5, NLRP3) from EOC and PSPC patients treated with bevacizumab.
- A hybrid deep learning framework and weakly supervised models were developed for biomarker analysis.
- Model performance was evaluated using training/testing splits and five-fold cross-validation.
Main Results:
- The AIM2-DL model achieved high performance metrics, including accuracy up to 0.92 and AUC of 0.97 in initial experiments.
- Cross-validation yielded strong results: accuracy 0.86 ± 0.07, precision 0.9 ± 0.07, recall 0.85 ± 0.06, F-measure 0.87 ± 0.06, and AUC 0.91 ± 0.05.
- Kaplan-Meier and Cox analyses confirmed the AIM2-DL model's ability to differentiate patients with positive outcomes and low recurrence from those with disease progression (p < 0.005).
Conclusions:
- The developed AIM2-DL framework represents the first automated precision oncology approach for selecting EOC and PSPC patients for bevacizumab therapy.
- AIM2 is identified as a promising biomarker for predicting therapeutic response to bevacizumab in ovarian cancer.
- This AI-driven approach can significantly improve patient stratification and personalize treatment strategies, potentially reducing cancer recurrence.
More Related Videos
08:55Murine Model for Non-invasive Imaging to Detect and Monitor Ovarian Cancer Recurrence
Published on: November 2, 2014
04:09Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
Published on: October 10, 2018