Table 1 Detailed dataset splitting specification with sample counts.

From: Novel metaheuristic optimized latent diffusion framework for automated oral disease detection in public health screening

Pathology class

Training count

Validation count

Test count

Total

Patient ID ranges

Stratification key

Dental caries (early)

1050

225

225

1500

DC_E_001-1500

Age + severity

Dental caries (advanced)

1260

270

270

1800

DC_A_001-1800

Age + location

Periodontal disease

1470

315

315

2100

PD_001-2100

Severity + site

Apical periodontitis

1190

255

255

1700

AP_001-1700

Tooth type + stage

Root resorption

1120

240

240

1600

RR_001-1600

Type + severity

Dental fractures

1330

285

285

1900

DF_001-1900

Location + type

Impacted teeth

1400

300

300

2000

IT_001-2000

Position + angle

Bone pathology

1050

225

225

1500

BP_001-1500

Location + type

Cysts/tumors

980

210

210

1400

CT_001-1400

Size + location

Orthodontic issues

1540

330

330

2200

OI_001-2200

Type + severity

TMJ disorders

1190

255

255

1700

TMJ_001-1700

Classification

Developmental anomalies

1260

270

270

1800

DA_001-1800

Type + severity

Total

17,500

3750

3750

25,000

Complete range

Multi-factor