# TRACE: A Concept Bottleneck Model for Longitudinal 3D Glioblastoma Response Assessment

arXiv: https://arxiv.org/abs/2606.30313 · HTML: https://mlsanity.com/paper/2606.30313/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 21 of its 48 references that are in the map.
- First posted on arXiv 29 Jun 2026.

## Builds on (most cited first, 21 of 21)

- [Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization](https://mlsanity.com/paper/1610.02391/) (2016)
- [Concept Bottleneck Models](https://mlsanity.com/paper/2007.04612/) (2020)
- [Med3D: Transfer Learning for 3D Medical Image Analysis](https://mlsanity.com/paper/1904.00625/) (2019)
- [Label-Free Concept Bottleneck Models](https://mlsanity.com/paper/2304.06129/) (2023)
- [Explaining Classifiers with Causal Concept Effect (CaCE)](https://mlsanity.com/paper/1907.07165/) (2019)
- [Automated brain extraction of multi-sequence MRI using artificial neural networks](https://mlsanity.com/paper/1901.11341/) (2019)
- [A Closer Look at the Intervention Procedure of Concept Bottleneck Models](https://mlsanity.com/paper/2302.14260/) (2023)
- [Robust and Interpretable Medical Image Classifiers via Concept Bottleneck Models](https://mlsanity.com/paper/2310.03182/) (2023)
- [Eliminating Information Leakage in Hard Concept Bottleneck Models with Supervised, Hierarchical Concept Learning](https://mlsanity.com/paper/2402.05945/) (2024)
- [Causally Reliable Concept Bottleneck Models](https://mlsanity.com/paper/2503.04363/) (2025)
- [DCBM: Data-Efficient Visual Concept Bottleneck Models](https://mlsanity.com/paper/2412.11576/) (2024)
- [Towards Interpretable Radiology Report Generation via Concept Bottlenecks using a Multi-Agentic RAG](https://mlsanity.com/paper/2412.16086/) (2024)
- [Constructing Concept-based Models to Mitigate Spurious Correlations with Minimal Human Effort](https://mlsanity.com/paper/2407.08947/) (2024)
- [Sample-efficient Learning of Concepts with Theoretical Guarantees: from Data to Concepts without Interventions](https://mlsanity.com/paper/2502.06536/) (2025)
- [Directed Acyclic Graphs and causal thinking in clinical risk prediction modeling](https://mlsanity.com/paper/2002.09414/) (2020)
- [A Concept-based Interpretable Model for the Diagnosis of Choroid Neoplasias using Multimodal Data](https://mlsanity.com/paper/2403.05606/) (2024)
- [Learning a Clinically-Relevant Concept Bottleneck for Lesion Detection in Breast Ultrasound](https://mlsanity.com/paper/2407.00267/) (2024)
- [Actionable Interpretability Must Be Defined in Terms of Symmetries](https://mlsanity.com/paper/2601.12913/) (2026)
- [CBM-RAG: Demonstrating Enhanced Interpretability in Radiology Report Generation with Multi-Agent RAG and Concept Bottleneck Models](https://mlsanity.com/paper/2504.20898/) (2025)
- [GLOMIA-Pro: A Generalizable Longitudinal Medical Image Analysis Framework for Disease Progression Prediction](https://mlsanity.com/paper/2507.12500/) (2025)
- [Concepts in Motion: Temporal Concept Bottleneck Model for Interpretable Video Classification](https://mlsanity.com/paper/2509.20899/) (2025)

## Abstract (arXiv)

Longitudinal glioblastoma response assessment requires comparing subtle tumor changes across MRI time points using structured clinical criteria such as RANO. However, most deep learning methods predict response labels directly from imaging features, which limits clinical inspection, verification, and correction. We introduce TRACE, a RANO 2.0-aligned concept bottleneck model for interpretable 4-class glioblastoma response classification on longitudinal 3D MRI. TRACE processes paired baseline and follow-up multimodal MRI scans with a shared 3D vision encoder, predicts clinically meaningful tumor measurements as root concepts, computes downstream RANO-derived concepts through deterministic rules, and incorporates scan interval and new-lesion information as passthrough concepts. This design frames response assessment as structured concept reasoning rather than direct image-to-label prediction. Using 5-fold patient-wise cross-validation on the LUMIERE dataset, TRACE achieves a 4-class macro F1 of 0.4769 and a binary progression-versus-non-progression macro F1 of 0.7085. It improves over a concept bottleneck baseline and remains within the range of published non-interpretable deep learning approaches. Ablation studies show that the expert RANO graph and intervention-consistency training are important for performance, while intervention experiments demonstrate that correcting concepts can improve downstream predictions. These results suggest that structured concept bottlenecks offer a transparent and clinically aligned direction for longitudinal glioblastoma response assessment, while highlighting the need for larger protocol-aligned datasets and external validation.