A GPT-4o-powered AI system enhances early-stage ovarian cancer diagnosis using pelvic CT exams.
Key Details
- 1Study published in Annals of Surgical Oncology; team led by Shimin Zhang, MD.
- 2GPT-4o trained to identify four key CT features linked to ovarian cancer malignancy.
- 3Dataset included 479 patients with confirmed benign or early malignant ovarian lesions.
- 4AI achieved diagnostic accuracies of 80.8%, 79.1%, and 93.3% across lesion identification, feature recognition, and malignancy status, respectively.
- 5Clinician reliability ratings for AI detection: 4.2–4.3 out of 5 across key features.
- 6AI assistance improved diagnosis accuracy for gynecologic oncologists (from 67.9% to 78.1%).
Why It Matters
This approach could reduce diagnostic variability and support less experienced clinicians by improving the consistency and accuracy of CT-based ovarian cancer diagnosis. Earlier detection enables better patient outcomes and survival rates.
Source
AuntMinnie
Related News
•AuntMinnie
AI Model Uses Ultrasound to Assess Fetal Lung Maturity
Researchers demonstrated an AI model's strong accuracy in measuring fetal lung maturity from ultrasound images.
•AuntMinnie
AI Model Predicts Dosimetry for Lu-177 PSMA Therapy Using PET/CT
A machine learning PET/CT model shows promise for predicting radiation dose prior to Lu-177 PSMA therapy in prostate cancer patients.
•AuntMinnie
LLM AI Significantly Boosts MRI Accuracy For Less Experienced Readers
AI LLMs notably improve diagnostic accuracy for less experienced brain MRI readers, with diminishing benefits for experts.