Table 1 Performance comparison of clustering methods for breast cancer detection
From: Fuzzy technique for microcalcifications clustering in digital mammograms
Author | Clustering method | N. of mammograms (clusters) | Sensitivity | Accuracy | FP/Im | Precision |
|---|---|---|---|---|---|---|
Nishikawa [[12]] | Spatial clustering | 78 (41) | 85% | - | - | - |
Cihan [[16]] | Subtractive clustering | 34 (72) | 93% | - | - | - |
Riyahi [[17]] | Wavelet transform and Fuzzy clustering | 47 (47) | 87% | - | 0.5% | - |
Cordella [[18]] | Graph-theoretical cluster analysis | 40 (102) | 94% | - | - | 100% |
Wang [[19]] | EFCM | 180 | - | 99.7% | - | - |
Quintanilla [[20]] | PFCM and ANN | - | 98.2% | 99.5% | - | - |
Malar [[21]] | Wavelet and ELM | 55 | - | 94% | Â | - |
Cheng [[22]] | FLSS | 40 (105) | - | > 97% | 14% | - |
Our method (Private database) | FCM-WF | 39 (39) | 93% | 95% | 4% | 62% |
Our method (MIAS database) | FCM-WF | 20(25) | 82% | 94% | 4% | 65% |