Clinical evaluation of Computer Aided Diagnosic support for ultrasound assessment of ovarian tumours
| Reference number | |
| Coordinator | Region Stockholm - Södersjukhuset Kvinnokliniken |
| Funding from Vinnova | SEK 1 500 000 |
| Project duration | September 2024 - August 2026 |
| Status | Completed |
| Venture | Medtech4Health innovators |
| Call | Medtech4Health: Implementation of medical technology in healthcare in 2024 |
Important results from the project
The main objectives were met. In the Usability & Safety study, 20 physicians tested the AI support on stored patient cases, and it proved safe and easy to use. In the Feasibility study, the AI support was used in routine clinical care and 120 patients were included (target: 60), with no adverse events for patients. Parts of the evaluation are ongoing. A follow-up interview study led to an improved tool with a risk scale and a roadmap for new AI tools in gynaecology, starting with endometriosis.
Expected long term effects
AI support can provide more women with ovarian tumors with an expert-level assessment at the first visit. It can reduce unnecessary referrals and surgeries, shorten waiting times and detect cancer earlier. The results are included in the clinical basis for CE marking according to the MDR and are followed by an Clinical Benefit-studie and a health economic analysis. The roadmap identifies the next step: AI-support for more gynecological diagnoses, primarily endometriosis.
Approach and implementation
The project was carried out in three stages: a Usability & Safety with saved patient cases, a clinical Feasibility study at Södersjukhuset, and analysis and regulatory documentation. The activities were correct and the collaboration worked well. The project took slightly longer than planned, as the start-up was more complex and it took time to include enough patients. The connection to the ultrasound device was more difficult than expected, but the AI support worked well via the workstation.