Research Article

Machine Learning Analysis of Immune Cells for Diagnosis and Prognosis of Cutaneous Melanoma

Figure 6

Construction and validation of nomogram. (a–c) Nomogram for predicting 3- and 5-year overall survival (OS) for melanoma patients in training (a), testing (b), and GSE54467 (c) based on the risk score and clinical characteristics. (d–f) Nomograms’ calibration curves in terms of the correspondence between the predicted and observed 3 and 5 years of outcomes in training (d), testing (e), and GSE54467 (f). The 45° dashed line symbolizes perfect prediction, and the blue and red lines represent our nomogram’s actual performance. (g–i) The decision curves for the nomogram, tumor stage at 3 and 5 years in training (g), testing (h), and GSE54467 (i).
(a)
(b)
(c)
(d)
(e)
(f)
(g)
(h)
(i)