Chapter 10 Table 10.2 Classification performance of selected Si‐sensor features LOO CV TP TN Sensitivity Specificity PPV NPV Accuracy Parathyroid – Adipose 15/21 16/18 71 (48‐88) 89 (64‐98) 88 (62‐98) 73 (50‐88) 79 Parathyroid – Thyroid 15/21 20/23 71 (48‐88) 87 (65‐97) 83 (58‐96) 77 (56‐90) 80 TP = true positive; TN = true negative numbers indicate identified tissue spots. A positive test is defined as the tissue observed being parathyroid gland; a negative test is defined as the tissue observed being adipose tissue / thyroid. Sensitivity; specificity; PPV = positive predictive value; NPV = negative predictive value; accuracy numbers are percentages; numbers in parentheses indicate 95% confidence interval. LOO CV = leave‐one‐out cross–validation. Classification of spectral data in InGaAs‐sensor range From the 36 extracted features 25 were located within the spectral detection range of InGaAs (see Figure 10.2). After binary logistic regression, gradients Ft12 (W1 – F4) and Ft17 (F4 – F5) were selected as the most promising combination for differentiation of parathyroid from surrounding adipose tissue. Gradients Ft12 (W1 – F4) and Ft14 (F2 – F1) were identified as most best distinctive feature combination for differentiation of parathyroid from thyroid tissue. Figure 10.5A and 10.5B show scatter plots for these InGaAs‐sensor based features, extracted for respectively parathyroid versus adipose tissue and parathyroid versus thyroid tissue. The quantitative results of LOO CV classification performance are listed in Table 10.3. LOO CV is solely based on InGaAs‐sensor range data from thyroid and parathyroid surgery for train and test purposes. Figure 10.5A Parathyroid versus adipose tissue: Scatter plot of two selected features within InGaAs‐range Scatter plot showing two selected features (gradients Ft12 and Ft17) 156
proefschrift_Schols_SLV
To see the actual publication please follow the link above