Cancer Cases from ACRIN Digital Mammographic Imaging Screening Trial: Radiologist Analysis with Use of a Logistic Regression Model
Bibliographic record
Abstract
PURPOSE: To determine which factors contributed to the Digital Mammographic Imaging Screening Trial (DMIST) cancer detection results. MATERIALS AND METHODS: This project was HIPAA compliant and institutional review board approved. Seven radiologist readers reviewed the film hard-copy (screen-film) and digital mammograms in DMIST cancer cases and assessed the factors that contributed to lesion visibility on both types of images. Two multinomial logistic regression models were used to analyze the combined and condensed visibility ratings assigned by the readers to the paired digital and screen-film images. RESULTS: Readers most frequently attributed differences in DMIST cancer visibility to variations in image contrast--not differences in positioning or compression--between digital and screen-film mammography. The odds of a cancer being more visible on a digital mammogram--rather than being equally visible on digital and screen-film mammograms--were significantly greater for women with dense breasts than for women with nondense breasts, even with the data adjusted for patient age, lesion type, and mammography system (odds ratio, 2.28; P < .0001). The odds of a cancer being more visible at digital mammography--rather than being equally visible at digital and screen-film mammography--were significantly greater for lesions imaged with the General Electric digital mammography system than for lesions imaged with the Fischer (P = .0070) and Fuji (P = .0070) devices. CONCLUSION: The significantly better diagnostic accuracy of digital mammography, as compared with screen-film mammography, in women with dense breasts demonstrated in the DMIST was most likely attributable to differences in image contrast, which were most likely due to the inherent system performance improvements that are available with digital mammography. The authors conclude that the DMIST results were attributable primarily to differences in the display and acquisition characteristics of the mammography devices rather than to reader variability.
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 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 | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| 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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, 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".