(a) Original slice, (b) binary mask computed by thresholding with μsup k), (c) object which contains the pixels belonging to the spine, (d) binary mask computed as the negative of the morphological dilation, (e) binary mask computed by thresholding with μ k), (f) ROI before the deletion of the spine, (g) application of mask (d) to the ROI, (h) restoration of the aorta, and (i) masked slice.
ΔF was computed as the negative of the slope of the regression of ln 1-Ft) on t for F6-F12 where F tis the ln 1-Ftf inbreeding at generation t, ΔG was computed as the diforrence in genetic level between G12 and G6 divided by 6.
Since different targets have different activity value ranges, the minimal activity threshold to locate an active set was computed as the average of the negative logarithmic activity values (i.e., ({log}} {Ki}) or ({log} {IC50})) plus one standard deviation above that activity value for each target.
The test score was computed as the ratio of the negative to the total number of sentences formed.
In our experiment, the term E I is computed as the sum of the negative Normalized Cross-Correlation of each corresponding image patch, the term E s is computed as Euclidean distance between two translations, and the choices of weight coefficients α and β are empirically decided.
In this analysis, under the assumption that CDSs, introns, 5'-UTR and randomized poly(A) regions do not contain real CS s), false positives and true negatives were computed as number of sequences with at least one site predicted or without any putative site, respectively.
For example, the precision of the negative class is computed as: (P mathit{neg}) = i/(c+f+i)); its recall, as: (R(mathit{neg}) = i/(g+h+i)); and the F1 measure is the harmonic mean between both precision and recall.
Similarly, negative predictive values were computed as the number of patients without a process of care as the numerator and the total number of patients registered without a process of care in the DNIP as the denominator.
Given the true positives (TP), the true negatives (TN), the false positives (FP) and the false negatives (FN), the MCC is computed as (3) This results in a value between +1.0 for perfect predictions and -1.0 for maximal false predictions.
Motif enrichment in foreground sets against background sets was measured using classification relative error rate (err), where relative error rate is computed as the average of the false positive and false negative rates [11].
