AI-based Strategies in Breast Mass ≤ 2cm classification with Mammography and Tomosynthesis: A Retrospective Evaluation
Bibliographic record
Abstract
Abstract Purpose To evaluate the efficiency of digital mammography (DM) and combined digital breast tomosynthesis (DBT) on AI-based strategies for breast mass ≤ 2cm classification. Methods DM and DBT images in 483 patients including 512 breast masses were acquired from November 2018 to November 2019. The radiomics and deep learning methods were employed to extract the breast mass features in images and finally for benign and malignant classification. The DM and combined DBT (DM+DBT) images were fed into radiomics and deep learning models to construct corresponding models, respectively. The area under the receiver operating characteristic curve (AUC) was estimated models performance. A comprehensive comparison of the subgroups AUCs of the best optimal model was calculated on age, tumor size, and breast density category. Results In the testing dataset, the AUC of DM combined DBT by radiomics and deep learning models were 0.869 and 0.908, respectively. Compared with the DM model, the combined DBT models based on radiomics and deep learning both showed statistically significant higher AUCs (0.869 vs. 0.810, P<0.001, by radiomics; 0.908 vs. 0.867, P<0.001, by deep learning). The deep learning models present superior than the radiomics models in the experiments with only DM (P<0.001) and DM+DBT (P<0.003). The advantage of the deep learning model is especially prominent in patients with small masses less than 1cm, 20 to 40 years old, and dense breast. Conclusions Deep learning model based on DM+DBT has a best diagnostic efficiency. AI-based stragies will play a major role in detecting early breast cancer in screening.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.005 | 0.013 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".