Deep learning reconstruction in biparametric prostate MRI: Impact on qualitative and radiomics analyses
Bibliographic record
Abstract
Objective To assess the impact of a commercially available deep learning reconstruction (DLR) algorithm on qualitative and radiomics analyses in prostate MRI. Methods This single-centre retrospective cohort included 25 consecutive patients who underwent a prostate MRI (1.5 T) in 2022. T2-weighted (T2WI), diffusion-weighted (DWI; b = 50, 1000, extrapolated 2000 s/mm 2 ) and apparent diffusion coefficient (ADC) images were reconstructed using DLR and standard (non-DLR) techniques. The two sets were mixed and blind-reviewed independently by six radiologists. Images were qualitatively scored according to PI-QUAL score, overall image quality, diagnostic confidence, anatomical conspicuity, artifact, and noise. Transition and peripheral zones were segmented and radiomics features extracted from region-of-interests using Pyradiomics package. Qualitative criteria and radiomics were compared using a pairwise Wilcoxon signed-rank test. Results PI-QUAL score was not significantly different ( p = 0.32). Overall image quality was not significantly different ( p = 0.21 on T2WI and 0.56 on DWI/ADC). Noise was lower on DLR images for T2WI ( p < 0.01) and DWI/ADC ( p = 0.04). Diagnostic confidence in excluding clinically significant cancer (PI-RADS ≥ 3) in the transition zone was lower with DLR images (p = 0.02). In the transition zone, 89/93 (96 %) of the radiomics features were significantly different between non-DLR and DLR images on T2WI, 68/93 (73 %) on DWI b-2000 s/mm 2 , and 55/93 (59 %) on ADC images. In the peripheral zone, 91/93 (98 %) were significantly different on T2WI, 50/93 (54 %) on DWI b-2000 s/mm 2 , and 70/93 (75 %) on ADC images. Conclusion Radiomics features were significantly different on DLR images which should encourage caution for clinical and research purposes. DLR algorithm decreases noise while preserving overall image quality.
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 | 0.008 | 0.026 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.000 |
| 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".