Breast Imaging Reporting and Data System Lexicon for US: Interobserver Agreement for Assessment of Breast Masses
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
PURPOSE: To retrospectively evaluate the interobserver agreement of radiologists who used the Breast Imaging Reporting and Data System (BI-RADS) lexicon to characterize and categorize ultrasonographic (US) features of breast masses. MATERIALS AND METHODS: No institutional review board approval or patient consent was required. Five breast radiologists retrospectively independently evaluated 267 breast masses (113 benign and 154 malignant masses in 267 patients) by using the BI-RADS US lexicon. Reviewers were blinded to mammographic images, medical history, and pathologic findings. Interobserver agreement was assessed with the Aickin revised kappa statistic. RESULTS: Interobserver agreement varied from fair for evaluation of mass margins (kappa = 0.36) to moderate for evaluation of lesion boundary (kappa = 0.48), echo pattern (kappa = 0.58), and posterior acoustic features (kappa = 0.47) to substantial for evaluation of mass orientation (kappa = 0.70) and shape (kappa = 0.64). For small (< or =0.7 cm; n = 49) or malignant (n = 154) masses, low concordance was noted for margin descriptors (kappa = 0.30 and 0.28, respectively) and BI-RADS category (kappa = 0.21 and 0.26, respectively). Overall, only fair agreement was obtained for BI-RADS category (kappa = 0.30). Agreement for subdivisions 4a, 4b, and 4c of BI-RADS category 4 was fair (kappa = 0.33), fair (kappa = 0.32), and poor (kappa = 0.17), respectively. CONCLUSION: Reproducibility of US BI-RADS terminology is good except for margin evaluation. A trend toward lower concordance was noted for the evaluation of small masses and malignant lesions. Classification into subdivisions 4a, 4b, and 4c was poorly reproducible.
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.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 it