> ## Content Index
> Fetch the complete content index at: https://nordisk-ai.ghost.io/llms.txt
> Use this file to discover other available public pages before exploring further.

# AI Outperforms Experts at Detecting Ovarian Cancer
- URL: https://nordisk-ai.ghost.io/ai-outperforms-experts-at-detecting-ovarian-cancer/
- Published: 2025-01-13T08:15:00.000Z
- Updated: 2026-08-27T15:54:44.000Z
- Description: AI has been developed to diagnose ovarian cancer using ultrasound images - with better accuracy than human experts. Could this revolutionize healthcare?
- Author: Anders Hagen
- Tags: Application Areas, AI, AI Sweden, research, Karolinska Institute, medical diagnostics, nordiskai, Sweden, Sweden, #Migrated-1787841634512, #wp, #wp-post, #Import 2026-08-27 14:41, #Import 2026-09-07 13:41

Researchers at Karolinska Institutet have, in a groundbreaking international study, developed AI models that can identify ovarian cancer with high precision by analyzing ultrasound images. The results show that the AI models can not only distinguish between benign and malignant tumors but also outperform human experts in diagnosis.

### AI Models Trained on Thousands of Ultrasound Images

The study used a comprehensive dataset consisting of 17,000 ultrasound images from 3,652 patients at 20 hospitals in eight different countries. The AI models achieved an average accuracy of 86.3 percent, higher than the 82.6 percent achieved by human experts and significantly better than the 77.7 percent achieved by less experienced examiners. Elisabeth Epstein, professor at the Department of Clinical Science and Education at Karolinska Institutet and senior consultant at the Women's Clinic at Södersjukhuset, commented on the results: "This suggests that AI models can offer valuable support in diagnosing ovarian cancer, especially in difficult-to-diagnose cases and in situations where there is a shortage of ultrasound experts."

### A Solution to Global Diagnostic Challenges

The background to the research is the need to complement human expertise, especially since a shortage of ultrasound experts is a major problem in many parts of the world. Ovarian tumors are often discovered by chance, which can lead to unnecessary surgical procedures or delayed cancer diagnosis. An important application of the AI models in the study was to simulate a patient prioritization process, known as triage. With AI support, referrals could be reduced by as much as 63 percent, and the rate of incorrect diagnoses was reduced by 18 percent.

### Next Step: Clinical Trials

The research team is now continuing work to test the AI models in real clinical settings at Södersjukhuset. The goal is to evaluate both the safety and practical usability of AI support in everyday care. The results so far point to a promising future where AI could become a crucial tool in medical diagnostics.