Additional file 1 of Conditional generative adversarial network driven radiomic prediction of mutation status based on magnetic resonance imaging of breast cancer
Bibliographic record
Abstract
Additional file 1: Figure S1. Top-down MRI view. Full 32 slices of top-down MRI view shown in Fig. 1. Figure S2. Side MRI view. Full 32 slices of side MRI view shown in Fig. 1. Figure S3. cGAN loss curves. The loss curves of the cGAN trained for 1200 epochs using the mean squared error loss function. Blue curve depicts the generator loss while the orange curve represents the loss for the discriminator. Figure S4. Real MRI for the patient TCGA-AO-A12E. Full 32 slices of the real patient MRI shown in panel A of Fig. 3. Figure S5. cGAN generated MRI for the patient TCGA-AO-A12E. Full 32 slices of the cGAN generated MRI shown in panel B of Fig. 3. Figure S6. Resnet 18 autoencoder generated MRI for the patient TCGA-AO-A12E. Full 32 slices of the Resnet 18 autoencoder generated MRI shown in C of Fig. 3. Figure S7. Traditional autoencoder generated MRI for the patient TCGA-AO-A12E. Full 32 slices of the autoencoder generated MRI shown in D of Fig. 3. Figure S8. CNN MSE loss curve, ROC and PR curves for PIK3CA. Top panel depicts CNN trained using real patient MRIs, middle panel represents CNN trained on cGAN predicted MRIs, and bottom panel for CNN trained with both real and cGAN generated MRIs. Figure S9. CNN MSE loss curve, ROC and PR curves for CDH1. Top panel depicts CNN trained using real patient MRIs, middle panel represents CNN trained on cGAN predicted MRIs, and bottom panel for CNN trained with both real and cGAN generated MRIs. Figure S10. ROC AUC and PR AUC for the chosen genes. Logistic regression with L1 regularization was trained to predict the mutation status of the 3 chosen genes. ROC AUC and PR AUC were calculated and plotted A TP53, B PIK3CA, C CDH1. Table S1. ROC AUC and PR AUC scores of CNN trained with cGAN predicted images for TP53, PIK3CA and CDH1 with various portions of the testing set. Table S2. CLAIMs checklist for artificial intelligence in medical imaging.
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 | 0.001 | 0.019 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.002 | 0.001 |
| Insufficient payload (model declined to judge) | 0.806 | 0.153 |
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; the direct Gemma label and the distilled Codex classifier agree on what is shown here.
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".