Machine learning-based ability to classify psychosis and early stages of disease through parenting and attachment-related variables is associated with social cognition

Table 4 Validated classification performance of individual classifiers in all study cohorts

Stydy Cohort	True Negatives	True Positives	False Negatives	False Positives	Sensitivity	Specificity	Balanced Accuracy	Area Under the Curve	Positive Predictive Value	Negative Predictive Value	Positive Likelihood Ratio	Permutation Test, p value
Discovery Sample (HC vs. PSY)	44	28	6	27	82.4	62.0	72.2	0.71	50.9	88.0	3.5	< 0.001
Validation Clinical Sample (HC vs. ESD)	34	19	11	26	67.7	59.3	63.2	0.63	46.6	77.7	1.8	n.a
Validation Familial Risk Sample (HC vs. FHR)	9	7	6	17	53.8	34.6	44.2	0.46	29.1	60.0	0.8	n.a

HC = Healthy Controls; PSY = Patients with Psychosis; ESD = Early Stages of Disease; FHR = Familial High Risk

ISSN: 2050-7283