Differentiating Small (≤1 cm) Focal Liver Lesions as Metastases or Cysts by means of Computed Tomography: A Case-Study to Illustrate a Fuzzy Logic-Based Method to Assess the Impact of Diagnostic Confidence on Radiological Diagnosis
Table 1
Examples of how to enter DC fuzzy levels in the 2 × 2 table cells using the operator (e.g., PM(d) = min[P(d), M(d)]). In the example (a), correct diagnosis of metastasis (M(d) = 1) is done with a DC level of 0.7, whereas in the example (b), correct diagnosis of cyst (B(d) = 1) is done with a DC level of 0.8. Of note, each couple of complementary diagnosis fills a column only of 2 × 2 table cells.
(a)
M(d) = 1
B(d) = 0
P(d) = 0.7
min(0.7, 1) = 0.7
0.7 − [min(0.7, 1)] = 0
N(d) = 0.3
0.3 − [min(0.3, 0)] = 0.3
min(0.3, 0) = 0
(b)
M(d) = 0
B(d) = 1
P(d) = 0.8
min(0.8, 0) = 0
0.8 − [min(0.8, 0)] = 0.8
N(d) = 0.2
0.2 − [min(0.2, 1)] = 0
min(0.2, 1) = 0.2
P(d): fuzzy diagnosis of malignancy; N(d): fuzzy diagnosis of benignancy; M(d): standard of reference diagnosis of malignancy; B(d): standard of reference diagnosis of benignancy.
