融合形态特征的基于GRU的介入机器人导丝轨迹预测建模OA

We present a causality-preserving sequential estimator for guidewire trajectory reconstruction during in-terventional navigation.Unlike generic recurrent baselines,the proposed model time-broadcasts sequence-level con-stants(including guidewire stiffness,insertion angle,and an effective friction descriptor)and concatenates them with dynamic geometric tokens(centerline coordinates and local diameter)before a two-stage feature encoder and a unidirectional gated recurrent unit decoder that emits 2D positions stepwise.To cope with variable sequence lengths,we adopt a time-step length classification training strategy with mask-based loss function,which limits pad-ding-induced invalid gradients and improves training and inference efficiency without altering the network architec-ture.On a phantom platform covering multiple guidewire types and insertion angles,the method achieves a 0.40～0.54 mm position-error range(mean 0.46 mm)while preserving strict causality;relative to a baseline without the time-step classification strategy,it reduces epochs-to-convergence by 42%,training time by 52%,and per-inference latency by 51%.These results indicate a deployable,real-time basis for guidewire trajectory estimation and intraop-erative navigation.