Theranostic applications: Non-ionizing cellular and molecular imaging through innovative nanosystems for early diagnosis and therapy

Figure 3 Example of a 3D automatic phantom image segmentation simulating liver vessel trees. The optimal algorithm configuration resulted in a vessel detection sensitivity of 100% for vessels of more than 1 mm in diameter, 50% in the range 0.5-1 mm and 14% in the 0.2-0.5 mm range. An average area overlap of 94.9% was obtained between automatically and manually segmented vessel sections, with an average difference of 0.06 mm². The average values of corresponding false positive and false negative ratios were approximately 8% and 3%, respectively.