Determining dense feature points on fingerprints used in constructing deep fixed-length representations for accurate matching, particularly at the pixel level, is of significant interest. To explore the interpretability of fingerprint matching, we propose a multi-stage interpretable fingerprint matching network, namely Interpretable Fixed-length Representation for Fingerprint Matching via Vision Transformer (IFViT), which consists of two primary modules. The first module, an interpretable dense registration module, establishes a Vision Transformer (ViT)-based Siamese Network to capture long-range dependencies and the global context in fingerprint pairs. It provides interpretable dense pixel-wise correspondences of feature points for fingerprint alignment and enhances the interpretability in the subsequent matching stage. The second module takes into account both local and global representations of the aligned fingerprint pair to achieve an interpretable fixed-length representation extraction and matching. It employs the ViTs trained in the first module with the additional fully connected layer and retrains them to simultaneously produce the discriminative fixed-length representation and interpretable dense pixel-wise correspondences of feature points. Extensive experimental results on diverse publicly available fingerprint databases demonstrate that the proposed framework not only exhibits superior performance on dense registration and matching but also significantly promotes the interpretability in deep fixed-length representations-based fingerprint matching.

A recent study by Ooi and Ooi (EH Ooi, ET Ooi, Mass transport in biological tissues: Comparisons between single- and dual-porosity models in the context of saline-infused radiofrequency ablation, Applied Mathematical Modelling, 2017, 41, 271-284) has shown that single-porosity (SP) models for describing fluid transport in biological tissues significantly underestimate the fluid penetration depth when compared to dual-porosity (DP) models. This has raised some concerns on whether the SP model, when coupled with models of radiofrequency ablation (RFA) to simulate saline-infused RFA, could lead to an underestimation of the coagulation size. This paper compares the coagulation volumes obtained following saline-infused RFA predicted based on the SP and DP models for fluid transport. Results showed that the SP model predicted coagulation zones that are consistently 0.5 to 0.9 times smaller than that of DP model. This may be explained by the low permeability value of the tissue interstitial space, which causes the majority of the saline to flow through the vasculature. The absence of fluid flow tracking in the vasculature in the SP model meant that any flow of saline into the vasculature is treated as losses and do not contribute to the saline penetration depth of the tissue. Comparisons with experimental results from the literature revealed that the DP models predicted coagulation zone sizes that are closer to the experimental values than the SP models. This supports the hypothesis that the SP model is a poor choice for simulating the outcome of saline-infused RFA.