Summary
We could make the main page of the HTML report a summary tab that contains summary statistics for the different steps of the pipeline (e.g., proportion of voxels with each adaptive mask value, proportion of voxels with curvefit failures, number of ICA components, percent variance explained by accepted and rejected components).
Additional Detail
Here is an example generated by ChatGPT. Some of the elements look useful, others look like noise.
┌──────────────────────────────────────────────────────────────────────┐
│ tedana QC summary │
├──────────────────────────────────────────────────────────────────────┤
│ Components: 67 total | 21 accepted | 43 rejected | 3 ignored │
│ Accepted variance: 56.4% | Rejected variance: 43.6% │
│ Kappa elbow: 12.7 | Rho elbow: 9.4 | Tree nodes with decisions: 8 │
│ Classification mask voxels: 118,204 | OC mask voxels: 139,812 │
│ Mean T2*: 38.2 ms | Median RMSE: 2.8 | Fit failures: 1.2% │
│ External regressors: task yes | nuisance yes | strongest hit: IC 12 │
│ Flags: high RMSE burst at vols 1–6; accepted IC near rho elbow │
├──────────────────────────────────────────────────────────────────────┤
│ [Open ICA review] [Open confounds-QC] [Open Rica] [Download JSON] │
└──────────────────────────────────────────────────────────────────────┘
Summary
We could make the main page of the HTML report a summary tab that contains summary statistics for the different steps of the pipeline (e.g., proportion of voxels with each adaptive mask value, proportion of voxels with curvefit failures, number of ICA components, percent variance explained by accepted and rejected components).
Additional Detail
Here is an example generated by ChatGPT. Some of the elements look useful, others look like noise.