Cardiac　scarring　and　edema　are　critical　pathological　features　of　heart　diseases.　Accurate　segmentation　of　these　features　in　Cardiac　Magnetic　Resonance　(CMR)　imaging　is　crucial　for　understanding　the　pathological　changes　associated　with　cardiac　diseases.　In　the　field　of　myocardial　scar　and　edema　segmentation,　it　is　of　significant　importance　to　study　the　C0,　T2,　and　LGE　modalities.　These　modalities　offer　different　perspectives　on　myocardial　tissue　characteristics,　aiding　in　the　more　accurate　diagnosis　and　assessment　of　cardiac　diseases.　However,　the　high-intensity　features　of　scars　and　edema　cannot　be　directly　obtained　from　individual　CMR　imaging　sequences,　making　simultaneous　accurate　segmentation　challenging.　To　address　this,　we　propose　a　multi-modal,　multi-channel　fusion　interactive　progressive　segmentation　strategy　that　leverages　the　distinctive　properties　of　each　modality　and　the　surrounding　tissue　characteristics　for　the　segmentation　of　myocardial　scars　and　edema.　We　have　designed　a　multi-channel　fusion　interactive　progressive　segmentation　model,　suitable　for　scar　and　myocardial　segmentation,　which　incorporates　an　attention　mechanism　that　enhances　channel　information　interaction　within　a　U-Net　structure　to　extract　features　across　different　modalities.　On　the　MyoPS++　2024　public　dataset,　our　method　achieved　an　average　Dice　score　of　0.5486　for　scar　segmentation　and　0.6081　for　the　segmentation　of　both　scars　and　edema.　©　The　Author(s),　under　exclusive　license　to　Springer　Nature　Switzerland　AG　2025.