Application of an Interpretable Machine Learning Model to Predict Lymph Node Metastasis in Patients with Laryngeal Carcinoma

Table 3

Multivariate logistic regression analysis of variables related to LNM.

Variables	OR	95% CI	value
Age at diagnosis
≥65	Reference
50–65	1.354	0.931–1.970	0.113
<50	1.248	1.080–1.442	0.003
Race
White	Reference
Black	1.064	0.886–1.277	0.510
Other	1.538	1.109–2.133	0.010
Sex
Male	Reference
Female	0.939	0.784–1.123	0.489
Primary site
Supraglottis	Reference
Glottis	0.289	0.252–0.354	<0.001
Subglottis	0.254	0.169–0.381	<0.001
Larynx	0.522	0.408–0.669	<0.001
Other	0.511	0.390–0.670	<0.001
Number of tumours
1	Reference
>1	0.744	0.642–0.863	<0.001
Tumour size
≤3 cm	Reference
>3 cm	1.742	1.498–2.026	<0.001
Grade
Grade I	Reference
Grade II	2.438	1.818–3.267	<0.001
Grade III and IV	4.609	3.407–6.235	<0.001
T-stage
T1	Reference
T2	2.923	2.231–3.829	<0.001
T3	3.925	3.030–5.084	<0.001
T4	5.890	4.571–7.590	<0.001

Notes: represents a statistically significant difference. Abbreviations: LNM, lymph node metastasis.