Ovarian Cancer Identification on CT Using Deep Learning

Development and Validation of a Deep Learning Model for Ovarian Cancer Identification on CT: a Nationwide Population-Based and International Study

Ovarian cancer remains the deadliest gynecologic malignancy, with poor survival rates largely due to late-stage diagnosis. Early detection is crucial, yet no universally accepted screening method exists. Current imaging techniques and biomarkers, such as CA-125, have limitations in specificity and sensitivity. This study aims to develop and evaluate a deep learning-based computer-aided diagnosis tool (CAT-OV), for ovarian cancer detection using CT imaging. The system integrates a Body Part Regression (BPR) model for pelvic localization and a Multiple Instance Learning (MIL) ensemble classifier for cancer prediction. The model was trained and validated using retrospective datasets from Taiwan, the United States, and a nationwide real-world cohort. Stringent preprocessing and quality control measures were implemented to enhance model accuracy. Results highlight the potential of AI-driven CT screening in improving early detection, though further validation is needed for clinical adoption.

Study Overview

• Status: Active, not recruiting
• Conditions: Ovarian Cancer

Detailed Description

Ovarian cancer is the deadliest gynecologic malignancy, with 20,890 new cases and 12,730 deaths expected in the U.S. in 2025. Despite accounting for only 2.1% of female cancers, it causes 4.3% of cancer-related deaths. The lifetime risk of developing ovarian cancer is 1 in 78, with a mortality risk of 1 in 108. Optimal treatment involves complete surgical resection followed by chemotherapy; however, survival rates have remained largely unchanged over the past two decades. Early detection significantly improves survival, but only 31% of cases are diagnosed at an early stage due to the disease's asymptomatic nature. No universal screening test exists, and current methods, including CA-125 and transvaginal ultrasound, have limitations in reducing mortality.

Computed tomography (CT) is commonly used for ovarian cancer detection, but its effectiveness is limited by nonspecific symptoms. AI-driven CT screening has gained interest, with deep learning showing promise in cancer detection. However, challenges remain in ensuring model generalizability and optimizing technical parameters. Effective screening must minimize unnecessary surgeries, as previous trials reported high false-positive rates and surgical complications. To address this, the study developed CAT-OV, an AI-based tool for ovarian cancer detection using CT imaging. The system integrates a Body Part Regression (BPR) model for pelvic localization and a Multiple Instance Learning (MIL) ensemble classifier with five convolutional neural networks (CNNs) to predict cancer presence. CAT-OV was evaluated on three test sets: an internal dataset, an international dataset from the U.S., and a nationwide multi-institutional dataset from Taiwan.

This retrospective study was approved by the institutional review board, and informed consent was waived. The dataset was constructed from CT scans of patients aged ≥20 who underwent ovarian surgery between 2010 and 2020 at Chang Gung Memorial Hospital. Malignant cases included various histopathological subtypes, while controls consisted of benign ovarian tumors and an enriched dataset of cancer-free individuals. The final dataset comprised 5,680 cases, split into a training/validation set (n=4,554) and an internal test set (n=1,126), including 173 cancer and 953 control cases. The international dataset included 40 cancer and 47 control cases from Brigham and Women's Hospital. The nationwide dataset consisted of 447 ovarian cancer cases and 1,131 controls from Taiwan's National Health Insurance database.

The BPR model, modified from ResNet50, localized pelvic regions on CT scans through unsupervised learning. Training involved preprocessing, augmentation, and regression-based subvolume selection. The MIL classification model treated each 3D subvolume as a "bag" of 2D slices, using EfficientNetV2-S as the backbone and an attention-based aggregation module for final prediction. Training involved preprocessing, augmentation, and a five-fold cross-validation strategy. The final ensemble model determined classification based on averaged logits and optimized thresholds. Visualization was performed using the Per-Sample Bottleneck technique to enhance interpretability.

Surgical histopathology served as the reference standard, reviewed by an experienced pathologist using the 2020 WHO classification. Immunohistochemical analysis was conducted to distinguish primary ovarian cancer from metastases. Tumors were staged according to the 8th edition FIGO system. This study aims to improve ovarian cancer detection and screening efficacy through AI-driven CT analysis.

• Study Type: Observational
• Enrollment (Actual): 12578

Contacts and Locations

Study Locations

• Taiwan
  • Guishan District
    • Taoyuan City, Guishan District, Taiwan, 333
      • Chang Gung Memorial Hospital

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: Yes

• Sampling Method: Probability Sample

Study Population

Women who have undergone a CT scan.

Description

Inclusion Criteria:

1. Age ≥ 20 years old.
2. Female
3. undergone a CT scan
4. undergone a CT scan within 180 days prior to ovarian surgery for histopathological evaluation.

Exclusion Criteria:

1. Age < 20 years old.
2. Non-female
3. Non-CT imaging
4. Incorrect image orientation
5. Number of slices < 10
6. Slice thickness >10 mm or < 1 mm
7. Unsuccessful DICM-to-NIfTI
8. Pelvic subvolume extraction failed
9. Non-contrast CT scans
10. Metallic artifacts
11. Inconclusive cases

Study Plan

How is the study designed?

Number of groups / cohorts

2

Cohorts and Interventions

Group / Cohort
control group; The control group included both benign ovarian tumors and an enriched dataset.
case group; ovarian cancer

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Performance of a deep learning-based computer-aided diagnosis tool (CAT-OV) for identification of primary ovarian cancer on CT	Sensitivity, Specificity, Accuracy, PPV, NPV, AUC	Perioperative/Periprocedural 180 days

Collaborators and Investigators

• Sponsor: Chang Gung Memorial Hospital

Study record dates

Study Major Dates

• Study Start (Actual): September 1, 2022
• Primary Completion (Actual): February 7, 2025
• Study Completion (Estimated): February 28, 2025

Study Registration Dates

• First Submitted: February 24, 2025
• First Submitted That Met QC Criteria: February 27, 2025
• First Posted (Actual): March 25, 2025

Study Record Updates

• Last Update Posted (Actual): March 25, 2025
• Last Update Submitted That Met QC Criteria: February 27, 2025
• Last Verified: February 1, 2025

More Information

Terms related to this study

• Additional Relevant MeSH Terms: Urogenital Diseases; Genital Diseases; Endocrine System Diseases; Urogenital Neoplasms; Neoplasms by Site; Neoplasms; Female Urogenital Diseases; Female Urogenital Diseases and Pregnancy Complications; Neoplasms by Histologic Type; Genital Diseases, Female; Endocrine Gland Neoplasms; Neoplasms, Glandular and Epithelial; Ovarian Diseases; Adnexal Diseases; Genital Neoplasms, Female; Gonadal Disorders; Carcinoma; Carcinoma, Ovarian Epithelial; Ovarian Neoplasms
• Other Study ID Numbers: 202201271B0

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No
• product manufactured in and exported from the U.S.: No