According to the investigators at the RadNet subsidiary DeepHealth, it was found that an artificial intelligence algorithm helped in the early detection of breast cancer when screened on mammography when compared to a conventional radiography model.
This study is published in the Nature Medicine.
Much progress has been made in applying AI algorithms in early detection fo breast cancers but there is always a scope for improvement particularly in developing methods for DBT and for demonstrating widespread generalizability, according to the researchers.
They sought to address the challenges of producing a high-performing AI model by progressively training an algorithm on a series of increasingly difficult tasks.
Hence, Bill Lotter, the lead author and DeepHealth co-founder and Chief Technology Officer describes the results that point to the clinical utility of AI for mammography in facilitating earlier breast cancer detection, as well as an ability to develop AI with similar benefits for other medical imaging applications.
By leveraging prior information learned in each successive training stage, this strategy results in AI that detects cancer accurately while also relying less on highly-annotated data. The approach and validation encompass DBT, which is particularly important given the growing use of DBT and the significant challenges it presents from an AI perspective, he further added.
The researchers compared the algorithm`s performance with five expert radiologists on index exams (mammograms acquired up to three months prior to biopsy-proven cancer) and "pre-index" studies (mammograms acquired 12-24 months prior to the index exam and were interpreted as negative in the clinical setting). These cases were gathered from a different population than was used to train the model.
The following results were seen-
a. On the index set (131 cancer exams and 154 confirmed negative exams), the algorithm yielded higher performance- 14.2% absolute increase in sensitivity (at the average reader specificity) and 24% absolute increase in specificity (at average reader sensitivity).
b. The algorithm also outperformed the readers on the non-index set (120 cancer cases exams and 154 confirmed negative cases)- 17.5% absolute increase in sensitivity (at the average reader specificity) and 16.2% absolute increase in specificity (at the average reader sensitivity)
c. The algorithm -- at 90% specificity -- flagged 45.8% of the missed cancer cases for follow-up.
Hence, the authors concluded that altogether, the results show great promise towards earlier cancer detection and improved access to screening mammography using deep learning.