how to change color of censor tick and truncate KM curve if number at risk is < 5 using lifelines library?

Question

Is it possible to change the colour of the censor tag in lifelines? The code-chunk that I used was

kmf.plot_survival_function(
    show_censors=True,
    censor_styles={'ms': 5, 'marker': '|'},
    at_risk_counts=True,
    ci_show=False
)

And how to truncate when the number at risk falls below a certain limit?

That depends entirely on what library this is. – Silvio Mayolo Aug 02 '22 at 01:14 — Silvio Mayolo, Aug 02 '22 at 01:14
Oh, I forgot to mention. It's lifelines – Saheli Saha Aug 02 '22 at 01:27 — Saheli Saha, Aug 02 '22 at 01:27

score 1 · Accepted Answer · answered Aug 02 '22 at 01:52

The underlying class that is used in lifelines for drawing the markers is matplotlib.lines.Line2D, and censor_styles is passed to that class as kwargs when you call plot_survival_function.

You can use the property markeredgecolor (or mec) with a color code and it'll change the marker color.

Here's an example based on the KMF sample code from lifelines's documentation.

from lifelines import KaplanMeierFitter
from lifelines.datasets import load_waltons
waltons = load_waltons()

kmf = KaplanMeierFitter(label="waltons_data")
kmf.fit(waltons['T'], waltons['E'])

kmf.plot_survival_function(
    show_censors=True,
    censor_styles={'ms': 20, 'marker': '|', 'markeredgecolor': '#ff0000'}, 
    color='#abc3f0',
    at_risk_counts=True,
    ci_show=True
)

And the resulting image with my red markers:

@SaheliSaha sorry, didn't notice that part. If you just want to truncate your y-axis, you could add the keyword argument `ylim=(low_range)` - I _think_ that might be what you want? Like `kmf.plot_survival_function(..., ylim=(5,))` — wkl, Aug 02 '22 at 02:21

how to change color of censor tick and truncate KM curve if number at risk is < 5 using lifelines library?

1 Answers1