Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

< Back to Article

Fig 1.

Convexity feature was developed to quantify tumor shape.

Convexity is computed as a ratio of tumor border (blue) to convex hull (red) (a). Convexity feature tracks the change in tumor morphology (b). Convexity is predictive of patient overall survival when dichotomized at the median value (c).

More »

Fig 1 Expand

Fig 2.

Entropy ratio was developed to quantify intensity variations across the tumor.

While some tumors present with consistent mean entropy across the core and the boundary (a), others have a distinct difference in the values (b).

More »

Fig 2 Expand

Table 1.

Distribution of study population demographics and imaging parameters by imaging biomarkers in Cohort 1.

More »

Table 1 Expand

Fig 3.

Entropy ratio between the core and border regions of the tumor is predictive of patient survival.

The tumors in the two prognostic groups (a) did not appear significantly different in the CT scans (b).

More »

Fig 3 Expand

Table 2.

Cox Proportional Hazards Models for Overall Survival.

More »

Table 2 Expand

Fig 4.

Histogram of the two imaging features across cohorts.

Convexity (a) shows similar range across cohorts (training-green, test-blue) However, training cohort is enriched with round tumors. The range of values for entropy ratio feature (b) is larger in training cohort. Both convexity (a) and entropy ratio (b) consistently capture targeted tumor characteristics in both cohorts.

More »

Fig 4 Expand