P Value. Note.—Table shows regressions performed by using only tumor ADC and by using tumor and normal PZ median ADC. The b parameters are regression parameters and their value and significance are shown respectively for each regression. Subscripts T and N = tumor and normal PZ tissue, respectively. C = regression constant. and the model combining tumor and normal PZ ADCs can be expressed as z = 0.126 − 18.82ADCT + 13.43ADCN (4). In combination with Equation (2), these models result in a probability that a given sample is a high-grade cancer. The model incorporating normal PZ ADC (Eq [4]), together with the data used in the regres- sion, is shown in Figure 3. This plot in- dicates that a relatively high tumor ADC might still constitute a high-grade tumor if the normal PZ ADC is high. In addition, one can appreciate that using a static threshold on tumor ADC (a vertical line/ contour in Fig 3) to determine cancer aggressiveness could result in incorrect diagnosis in some patients. Including normal PZ significantly (P = .0401) improved diagnostic accuracy. The ROC curves for the regression models in Equations 3 and 4 are shown in Figure 4. The area under the curve increases from 0.91 to 0.96. We have also included flow charts detailing the diagnostic accuracy of both tests in Figure 1.
Appears in 3 contracts
Sources: End User Agreement, End User Agreement, End User Agreement