Breast Cancer Prognosis Risk Estimation Using Integrated Gene Expression and Clinical Data

Figure 10

Kaplan-Meier survival graphs for good and poor prognosis groups in the Desmedt dataset for the IPRE algorithm and seven other existing algorithms. A log-rank test was performed to assess the 𝑃 value and to signify that the lower the 𝑃 value, the better the separation between the two prognosis groups. From this figure, the IPRE algorithm was able to achieve the lowest 𝑃 value and perform the best at separating the two prognosis groups.

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(h)