Skip to content

Approaches — dual-path late fusion vs early fusion

Approach A — dual-path late fusion (primary)

Each input gets its own Features + Mamba SSM; the two models converge at the output.

flowchart LR
    SIG["signal\nraw ECG waveform"] --> F1["features A"] --> M1["Model Mamba SSM"]
    SCAL["scalogram\nCWT heatmap"] --> F2["features B"] --> M2["Model Mamba SSM"]
    M1 --> OUT["output\nN / V / S / F / Q"]
    M2 --> OUT

Note: the sketch output reads N / V / S / F / S / Q (S twice, six entries). Assumed typo for N / V / S / F / Q. Flag if the sixth output is intentional.

Approach B — early fusion (alternative)

Signal and scalogram encoded separately, concatenated per time step, one Mamba SSM, one 5-logit head. Fewer params, joint representation. Risk: stronger branch dominates, needs aligned time axes.

Comparison

  • Params/compute: B ~1 model; A ~2 models, roughly double train cost.
  • Ablation: A gives per-modality scores (signal-only vs scalogram-only) for free; B needs extra runs to prove the scalogram helps.
  • Robustness: A survives a noisy modality (one branch can outvote); B can collapse if fusion is poorly balanced.
  • Fusion rule (A, undecided): average logits, learned per-class gate, or majority vote. Default: average logits, then per-class threshold tuning.

Decision — A is primary, built first

Build A first: train signal-only and scalogram-only branches, record per-branch macro F1 and V recall, then fuse. Build B second as the ablation. Keep whichever wins on DS2 macro F1 + V recall; report both.

Effect on plan

  • plan.md is research-phased: literature review first, then baselines, Approach A experiments, Approach B ablation, final write-up.
  • Loss/metrics in model.md apply unchanged to each branch.