Set Transformer Architectures and Synthetic Data Generation for Flow-Guided Nanoscale Localization

TL;DR

This paper introduces Set Transformer architectures combined with synthetic data generation to improve flow-guided nanoscale localization, enhancing adaptability and robustness in medical diagnostics.

Contribution

It proposes a permutation-invariant Set Transformer model for FGL and integrates deep generative models for synthetic data augmentation, addressing variability and data scarcity issues.

Findings

Set Transformer achieves comparable accuracy to GNNs.

Synthetic data improves model robustness under data scarcity.

Permutation-invariant models enhance generalization to anatomical variability.

Abstract

Flow-guided Localization (FGL) enables the identification of spatial regions within the human body that contain an event of diagnostic interest. FGL does that by leveraging the passive movement of energy-constrained nanodevices circulating through the bloodstream. Existing FGL solutions rely on graph models with fixed topologies or handcrafted features, which limit their adaptability to anatomical variability and hinder scalability. In this work, we explore the use of Set Transformer architectures to address these limitations. Our formulation treats nanodevices' circulation time reports as unordered sets, enabling permutation-invariant, variable-length input processing without relying on spatial priors. To improve robustness under data scarcity and class imbalance, we integrate synthetic data generation via deep generative models, including CGAN, WGAN, WGAN-GP, and CVAE. These models are trained to replicate realistic circulation time distributions conditioned on vascular region labels, and are used to augment the training data. Our results show that the Set Transformer achieves comparable classification accuracy compared to Graph Neural Networks (GNN) baselines, while simultaneously providing by-design improved generalization to anatomical variability. The findings highlight the potential of permutation-invariant models and synthetic augmentation for robust and scalable nanoscale localization.