Computational framework for discriminating AITs from biological tubular microcavities using morphometric analysis and Fisher LDA
| Feature | Threshold | Accuracy | Effect Size | p-value |
|---|---|---|---|---|
| Branching Index | 0.000 | 1.000 | -3.131 | 3.10e-17 |
| Diameter CV | 0.023 | 0.983 | -1.835 | 3.09e-09 |
| Straightness | 0.159 | 0.983 | 38.330 | 3.97e-76 |
| Tortuosity | 1.000 | 0.983 | -3.024 | 1.28e-16 |
| Mineral Lining | 0.430 | 0.983 | 4.267 | 2.95e-23 |
| Organic Residue | 0.104 | 0.983 | -3.786 | 7.85e-21 |
| Polygonality | 0.189 | 0.967 | 2.862 | 1.16e-15 |
| Wall Roughness | 0.113 | 0.783 | -0.865 | 1.69e-03 |
| Terminal Shape | 0.009 | 0.767 | -1.056 | 1.70e-04 |
| Diameter Trend | 0.0001 | 0.750 | 0.086 | 7.43e-01 |
| Feature | AIT Mean | AIT Std | Bio Mean | Bio Std |
|---|---|---|---|---|
| Diameter CV | 0.021 | 0.001 | 0.202 | 0.140 |
| Straightness | 1.000 | <0.001 | 0.082 | 0.034 |
| Polygonality | 0.353 | 0.121 | 0.097 | 0.037 |
| Branching Idx | 0.000 | 0.000 | 2.811 | 1.270 |
| Wall Roughness | 0.101 | 0.008 | 0.127 | 0.041 |
| Tortuosity | 1.000 | <0.001 | 14.504 | 6.315 |
| Terminal Shape | 0.001 | 0.003 | 0.108 | 0.143 |
| Mineral Lining | 0.692 | 0.138 | 0.158 | 0.110 |
| Organic Residue | 0.048 | 0.027 | 0.556 | 0.188 |
| Diameter Trend | 0.0002 | <0.001 | -0.0002 | 0.008 |