Comparison of AI-integrated pathways with human-AI interaction in population mammographic screening for breast cancer
Artificial intelligence (AI) readers of mammograms compare favourably to individual radiologists in detecting breast cancer. However, AI readers cannot perform at the level of multi-reader systems used by screening programs in countries such as Australia, Sweden, and the UK. Therefore, implementation demands human-AI collaboration. Here, we use a large, high-quality retrospective mammography dataset from Victoria, Australia to conduct detailed simulations of five potential AI-integrated screening pathways, and examine human-AI interaction effects to explore automation bias. Operating an AI reader as a second reader or as a high confidence filter improves current screening outcomes by 1.9-2.5% in sensitivity and up to 0.6% in specificity, achieving 4.6-10.9% reduction in assessments and 48-80.7% reduction in human reads. Automation bias degrades performance in multi-reader settings but improves it for single-readers. This study provides insight into feasible approaches for AI-integrated screening pathways and prospective studies necessary prior to clinical adoption.
Authors
- Ravishankar Karthik (ORCID: https://orcid.org/0000-0002-4709-9058)
- Enes Makalic (ORCID: https://orcid.org/0000-0003-3017-0871)
- John L. Hopper (ORCID: https://orcid.org/0000-0002-8567-173X)
- Jocelyn Lippey (ORCID: https://orcid.org/0000-0001-9977-1646)
- Gustavo Carneiro (ORCID: https://orcid.org/0000-0002-5571-6220)
- Brendan Hill (ORCID: https://orcid.org/0000-0001-9255-7479)
- Davis J. McCarthy (ORCID: https://orcid.org/0000-0002-2218-6833)
- Michael S. Elliott (ORCID: https://orcid.org/0000-0002-2406-4522)
- Carlos A. Peña‐Solórzano (ORCID: https://orcid.org/0000-0002-1734-7284)
- Yuanhong Chen (ORCID: https://orcid.org/0000-0002-8983-2895)
- Osamah M. Al-Qershi (ORCID: https://orcid.org/0000-0002-1796-2009)
- Daniel F. Schmidt (ORCID: https://orcid.org/0000-0002-9603-9529)
- Peter Brotchie (ORCID: https://orcid.org/0000-0003-2313-9067)
- Tuong L. Nguyen (ORCID: https://orcid.org/0000-0002-6597-8363)
- Helen Frazer (ORCID: https://orcid.org/0000-0001-9399-8102)
- Prue C. Weideman (ORCID: https://orcid.org/0000-0002-3672-5201)
- Chun Fung Kwok (ORCID: https://orcid.org/0000-0002-0716-3879)
- Samantha Fox (ORCID: https://orcid.org/0000-0001-5748-8869)
- Katrina Kunicki (ORCID: https://orcid.org/0000-0001-6579-3150)
- Prabhathi Basnayake Ralalage (ORCID: https://orcid.org/0000-0003-2771-5316)
- Chong Wang (ORCID: https://orcid.org/0000-0003-0022-0217)
- Shuai Li (ORCID: https://orcid.org/0000-0002-8696-8594)
Institutions
- The University of Melbourne (AU)
- University of Surrey (GB)
- St Vincent's Hospital Melbourne (AU)
- Australian Centre for Robotic Vision (AU)
- Torrens University Australia (AU)
- Melbourne Health (AU)
- Breastscreen Victoria (AU)
- Melbourne Genomics Health Alliance (AU)
- Monash University (AU)
- The University of Adelaide (AU)
- St Vincents Institute of Medical Research (AU)
Publication Details
- Journal
- Nature Communications
- Published
- 2024-08-30
- DOI
- https://doi.org/10.1038/s41467-024-51725-8
- Citations
- 33
- Primary Topic
- AI in cancer detection
- Type
- article
- Field-Weighted Citation Impact
- 11.12
Funders
- Ramaciotti Foundations
- Australian Government
- Department of Health and Aged Care, Australian Government
- St Vincent's Hospital Melbourne
- Medical Research Council
- National Health and Medical Research Council