Artificial Intelligence in Breast Cancer Screening in Region Östergötland Linkoping

Completed

• Conditions: Breast Cancer
• Interventions: Other: AI cancer detection system

• Registration Number: NCT05048095
• Lead Sponsor: Ostergotland County Council, Sweden

Brief Summary

The purpose of this observational study is to assess whether the use of AI (Transpara®) can lead to an improved quality of a double reading mammography screening program. This is investigated by performing AI as a third reader and as a decision support during the consensus meeting, compared with conventional mammography screening (double reading and consensus without AI).

Detailed Description

The AI cancer detection system will act as a 3rd reader and will recall additional cases to the consensus conference: the exams that were not recalled by double reading but are classified as the 3% most suspicious exams, based on AI derived cancer-risk scores. Secondly, AI is used as a decision support during consensus. AI risk scores and Computer-Aided Detection (CAD)-marks of suspicious calcifications and soft tissue lesions are provided to the reader(s).

The hypothesis of this study is that the use of AI has the potential to improve the quality of the screening program by increasing the cancer detection rate without affecting the recall rate.

Study Timeline

• Study First Submitted: 2021-09-08
• Study First Submitted QC: 2021-09-08
• Study First Posted: 2021-09-17
• Study Start: 2021-10-15
• Primary Completion: 2022-02-15
• Study Completion: 2022-02-15
• Status Verified: 2022-04
• Last Update Submitted: 2022-04-19
• Last Update Posted: 2022-04-20

Recruitment & Eligibility

• Status: COMPLETED
• Sex: Female
• Target Recruitment: 15500

Inclusion Criteria

• Women participating in the regular Breast Cancer Screening Program in Region Östergötland Linkoping

Exclusion Criteria

• Women with breast implants or other foreign implants in the mammogram
• Women with symptoms or signs of suspected breast cancer

Study & Design

• Study Type: OBSERVATIONAL
• Study Design: Not specified

Arm && Interventions

Group	Intervention	Description
Screened women in Region Östergötland Linkoping	AI cancer detection system	-

Primary Outcome Measures

Name	Time	Method
Cancer Detection rate	After 4 months of inclusion	Proportion of women diagnosed with breast cancer among those recalled after consensus
Recall or referral rate	After 4 months of inclusion	Proportion of women who are referred for further diagnostic workup after consensus
Positive predictive value of referrals	After 4 months of inclusion	Proportion of women diagnosed with breast cancer among those referred

Secondary Outcome Measures

Name	Time	Method
Positive predictive value of Transpara® scores	After 4 months of inclusion	Proportion of breast cancers diagnosed among women with a given AI score

Trial Locations

Locations (1)

Region Östergötland; 🇸🇪 Linköping, Östergötland, Sweden