Artificial Intelligence (AI) is transforming healthcare, and a new study shows it may revolutionize breast cancer screening by detecting cancers years before they are clinically diagnosed. Commercially available AI algorithms, designed to help radiologists detect breast cancer in mammograms, are proving capable of identifying women at risk years in advance. These findings, published in JAMA Network Open, suggest that AI could pave the way for more personalized and earlier breast cancer diagnoses.
How AI Detects Breast Cancer Early
The study, conducted by Solveig Hofvind, PhD, and her team from the Norwegian Institute of Public Health, investigated whether AI scores from the INSIGHT MMG algorithm could predict the development of breast cancer in women who had undergone routine mammograms. The research found that women who later developed breast cancer had consistently higher AI scores four to six years before the cancer was detected by traditional methods.
The AI algorithm assigns a score between 0 and 100, indicating the likelihood of cancer based on the current mammogram. Higher scores suggest a greater risk. The researchers analyzed data from 116,495 women, aged 50 to 69, who participated in a national screening program in Norway. These women had undergone at least three consecutive biennial mammography screenings.
Key Findings of the Study
The study revealed significant differences in AI scores between women who developed breast cancer and those who remained cancer-free:
- Mean absolute differences in AI scores were 21.3, 30.7, and 79.0 during the first, second, and third screening rounds, respectively, for women who developed screening-detected cancer.
- At an AI score threshold of 91.3, the AI algorithm identified 4.5% of cancers at the first screening, 8.6% at the second, and 52.9% by the third screening round.
- For women who did not develop cancer, only 0.7% had a false-positive AI score at each screening round.
These findings suggest that AI technology could identify subtle imaging features linked to future breast cancer cases, enabling earlier and potentially less aggressive treatment options.
The Future of Personalized Screening
With AI algorithms like INSIGHT MMG showing promise for early breast cancer detection, the possibility of personalized screening programs becomes more feasible. The ability of AI to estimate short-term cancer risk could lead to tailored preventive measures, such as:
- More frequent screenings for those with high AI scores
- Supplemental imaging techniques to monitor suspicious areas
- Early diagnosis of cancers that might otherwise go undetected until later stages
These proactive approaches could improve outcomes by catching breast cancer earlier, when it is easier to treat.
Commercial AI in Breast Cancer Detection
Several AI algorithms have already received regulatory approval to assist radiologists in detecting breast cancer in mammograms. These algorithms flag areas of concern and provide scores indicating the likelihood of malignancy. The current study shows that these same algorithms can also identify future cancer risks, which could change how screening programs are structured.
What This Means for Breast Cancer Patients
For women at risk of breast cancer, the development of AI-based screening methods offers a promising new approach. Early detection is critical for reducing the need for more aggressive treatments, and with AI providing more accurate risk assessments, screening programs can become more individualized and effective.
AI is not only enhancing immediate cancer detection, but also playing a role in estimating future cancer development. With continued research and advances in AI, the goal of earlier breast cancer diagnosis may soon become a reality for more women.
Conclusion
The study highlights the potential of AI algorithms in breast cancer detection—both in identifying cancers at the time of screening and predicting risks years in advance. As AI technology continues to evolve, it may lead to more personalized screening and treatment approaches, offering new hope for early intervention and better outcomes.
