Prenatal prediction of neonatal respiratory morbidity: a radiomics method based on imbalanced few-shot fetal lung ultrasound images

Table 3 Feature names and means of the features selected

Feature name	Mean ± std
	Normal	NRM
Energy	0.543 ± 0.070	0.551 ± 0.063
Inverse difference moment normalized	0.999 ± 0.0004	0.998 ± 0.0005
High grey-level run emphasis	298 ± 62.5	279 ± 57.0
Run-length variance	(2.04 ± 0.996) × 10⁻⁵	(2.30 ± 0.844) × 10⁻⁵
Inverse difference moment normalized of approximation	0.801 ± 0.113	0.773 ± 0.114
Information measure of correlation 1 of approximation	0.989 ± 0.002	0.990 ± 0.002
Energy of horizontal	0.362 ± 0.036	0.374 ± 0.042
Sum entropy of vertical		(4.81 ± 2.87) × 10⁴
Long-run high grey-level emphasis of vertical	432 ± 78.4	462 ± 95.3
Energy of diagonal	(1.40 ± 0.724) × 10³	(1.20 ± 0.841) × 10³

Approximate, horizontal, vertical, and diagonal were decomposed from the image by wavelet transform (first-level decomposition)

ISSN: 1471-2342