![[2602.17855] TopoGate: Quality-Aware Topology-Stabilized Gated Fusion for Longitudinal Low-Dose CT New-Lesion Prediction](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.17855] TopoGate: Quality-Aware Topology-Stabilized Gated Fusion for Longitudinal Low-Dose CT New-Lesion Prediction
Summary
The paper presents TopoGate, a model designed to enhance new-lesion prediction in longitudinal low-dose CT scans by integrating quality-aware gating mechanisms.
Why It Matters
This research addresses the challenges of noise and variability in CT imaging, which can lead to false alarms in lesion detection. By improving prediction accuracy, TopoGate has the potential to enhance patient outcomes and streamline diagnostic processes in medical imaging.
Key Takeaways
- TopoGate improves lesion prediction accuracy in low-dose CT scans.
- The model uses a quality-aware gate driven by CT appearance quality and registration consistency.
- Removing low-quality data pairs enhances the model's performance significantly.
- The approach aligns with radiologist practices by adapting to image quality variations.
- TopoGate is practical and interpretable, making it suitable for clinical use.
Electrical Engineering and Systems Science > Image and Video Processing arXiv:2602.17855 (eess) [Submitted on 19 Feb 2026] Title:TopoGate: Quality-Aware Topology-Stabilized Gated Fusion for Longitudinal Low-Dose CT New-Lesion Prediction Authors:Seungik Cho View a PDF of the paper titled TopoGate: Quality-Aware Topology-Stabilized Gated Fusion for Longitudinal Low-Dose CT New-Lesion Prediction, by Seungik Cho View PDF HTML (experimental) Abstract:Longitudinal low-dose CT follow-ups vary in noise, reconstruction kernels, and registration quality. These differences destabilize subtraction images and can trigger false new lesion alarms. We present TopoGate, a lightweight model that combines the follow-up appearance view with the subtraction view and controls their influence through a learned, quality-aware gate. The gate is driven by three case-specific signals: CT appearance quality, registration consistency, and stability of anatomical topology measured with topological metrics. On the NLST--New-Lesion--LongCT cohort comprising 152 pairs from 122 patients, TopoGate improves discrimination and calibration over single-view baselines, achieving an area under the ROC curve of 0.65 with a standard deviation of 0.05 and a Brier score of 0.14. Removing corrupted or low-quality pairs, identified by the quality scores, further increases the area under the ROC curve from 0.62 to 0.68 and reduces the Brier score from 0.14 to 0.12. The gate responds predictably to degradation, placing m...