Table 1 Description of the 7 features that change monotonically (decrease/increase) with diabetes stage/progression and P values of changes

From: Dermal features derived from optoacoustic tomograms via machine learning correlate microangiopathy phenotypes with diabetes stage

Feature number

Feature description

Feature scale

Feature change

P values

A vs B

B vs C

A vs C

3

Number of junction-to-junction branches in the dermal layer (2D)

Mesoscale

0.047*

0.153

0.01**

5

Thickness of the dermal layer (3D)

Macroscale

0.464

0.188

0.252

6

Total area of vessels in the dermal layer (3D)

Macroscale

0.317

0.326

0.477

8

Average vessel length in the dermal layer (2D)

Microscale

0.156

0.336

0.315

13

Number of vessels in the dermal layer (3D)

Mesoscale

0.395

0.364

0.276

20

Number of junctions in the dermal layer (3D)

Mesoscale

0.387

0.373

0.276

21

Number of junction-to-junction branches in the dermal layer (3D)

Mesoscale

0.349

0.221

0.331

1

Number of vessels in the epidermal/SVP layer (2D)

Mesoscale

0.257

0.386

0.323

2

Number of junctions in the epidermal/SVP layer (2D)

Mesoscale

0.263

0.444

0.377

4

Number of junction-to-junction branches in the epidermal/SVP layer (2D)

Mesoscale

0.447

0.266

0.377

25

Average vessel area in the epidermal/SVP layer (2D)

Microscale

0.109

0.354

0.09*

  1. *P < 0.05, **P ≤ 0.01. All P values were calculated using a two-sided Student’s t-test.