Improved Prediction of Trimodality Therapy Response in Muscle-Invasive Bladder Cancer Using Deep Learning and Pathology-Optimized Expansion Microscopy
Bibliographic record
Abstract
Trimodality therapy (TMT) is considered one of the first-line treatment options for muscle-invasive bladder cancer (MIBC). Regardless, a significant portion of patients with MIBC do not respond to this treatment, highlighting an urgent need for improved diagnostic tools [1,2]. Our study introduces a novel approach combining fluorescent in-situ hybridization (FISH), expansion microscopy, and state of the art deep learning models to enhance the prediction of MIBC treatment responsiveness [3-8]. In this pilot study involving a small cohort of MIBC patients, categorized as responders or non-responders based on clinical local response status at follow-up; patient tissue slides were stained with Hematoxylin-Eosin (H&E), and expanded tissue samples underwent fluorescent labelling using DAPI nuclear stain, wheat germ agglutinin, FISH staining with TelC and CENP-B box molecular probes. Preliminary results from the model that was trained with Magnify-processed samples showed 49% increase in area under curve (AUC) value compared to that of the model trained with images of H&E staining, along with significantly improved sensitivity and specificity metrics. The outcomes revealed that models trained on expansion microscopy images significantly outperformed those trained on conventional H&E images, with the expansion-trained model showing prominent evaluation results, indicating that the detailed cellular and molecular information captured through expansion microscopy substantially enhances the model's ability to predict treatment outcomes. Moreover, we used saliency maps and Gradient-weighted Class Activation Mapping (Grad-CAM) to elucidate the features and attention of the model in predicting treatment responses, providing insights into the model's prediction capability. In conclusion, our findings suggest that fluorescent staining of multiple markers and nuclear structures, coupled with expansion microscopy and deep learning, can potentially enhance the performance of models predicting TMT responsiveness in MIBC patients. The innovative method offers a cost-effective, accessible diagnostic tool for pathologists to enhance patient care and treatment decisions.
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 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| 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".