Researchers have developed SurgicalMamba, an algorithm for online surgical phase recognition. It uses a dual-path SSD approach with two mechanisms: state regramming and intensity-modulated stepping. State regramming rotates the carried state at each chunk boundary based on the content of the chunk, while intensity-modulated stepping increases decay at annotated phase transitions. This allows the algorithm to maintain high accuracy even in long procedures with varying phase lengths. SurgicalMamba has been tested on seven public benchmarks and achieves state-of-the-art online accuracy and phase-level Jaccard scores. --- Why it matters: SurgicalMamba matters because it addresses a significant challenge in online surgical phase recognition: maintaining high accuracy over long periods without increasing computational cost. Its ability to handle varying phase lengths and maintain performance makes it a valuable tool for medical professionals and researchers. Source: https://arxiv.org/abs/2605.14889