# pip install pandas
# pip install numpy
# pip install lifelines
# pip install matplotlib
from lifelines import KaplanMeierFitter
import pandas as pd
import numpy as np
from matplotlib import pyplot as plt
gbsg2 = pd.read_csv("/home/pgr16/Documents/Data_Analysis/German Breast Cancer/GBSG2.csv")
# Create Age Group
bins = [0, 40, 81]
labels = ["0-40", "40 and older"]
gbsg2['Age Group'] = pd.cut(gbsg2['age'], bins = bins, labels= labels, right = False)
kmf = KaplanMeierFitter()
T = gbsg2["time"]
E = gbsg2["cens"]
kmf.fit(T, event_observed=E)
plt.clf() # clear plot
kmf.plot_survival_function()
plt.title("Time to Recurrence of German Breast Cancer Patients")
print(kmf.median_survival_time_)
import matplotlib.pyplot as plt
from lifelines import KaplanMeierFitter
age = (gbsg2["Age Group"] == "40 and older")
plt.clf()
ax = plt.subplot(111)
kmf = KaplanMeierFitter()
kmf.fit(T[age], event_observed=E[age], label="Over 40 years old")
kmf.plot_survival_function(ax=ax)
kmf.fit(T[~age], event_observed=E[~age], label="40 years or younger")
kmf.plot_survival_function(ax=ax)
plt.title("Kaplan-Meier Survival Curves by Age Group")
plt.xlabel("Time (days)")
plt.ylabel("Survival Probability")
plt.legend()
grade1 = gbsg2["tgrade"] == "I"
grade2 = gbsg2["tgrade"] == "II"
grade3 = gbsg2["tgrade"] == "III"
plt.clf()
ax = plt.subplot(111)
kmf = KaplanMeierFitter()
# Grade 1
kmf.fit(T[grade1], event_observed=E[grade1], label=" Tumor Grade 1")
kmf.plot_survival_function(ax=ax, at_risk_counts=True)
# Grade 2
kmf.fit(T[grade2], event_observed=E[grade2], label="Tumor Grade 2")
kmf.plot_survival_function(ax=ax, at_risk_counts=True)
#Grade 3
kmf.fit(T[grade3], event_observed=E[grade3], label="Tumor Grade 3")
kmf.plot_survival_function(ax=ax, at_risk_counts=True)
plt.title("Time to Recurrence by Tumor Grade")
plt.xlabel("Time (days)")
plt.ylabel("Survival Probability")
plt.legend()
