Oral Disease Breath VOC Profiles from GC-MS and Electronic Nose

Breath samples from 346 participants across five diagnostic groups (caries, gingivitis, periodontitis, peri-implantitis, healthy controls) were analyzed for volatile organic compound (VOC) profiles. The dataset contains VOC measurements, electronic nose sensor array responses, and diagnostic labels used to evaluate noninvasive disease detection.

Use Cases

• Train machine-learning models to classify diagnostic labels using VOC measurement patterns.
• Analyze correlations between specific electronic nose sensor array responses and oral disease states.
• Identify disease-specific volatile organic compound (VOC) signatures by comparing profiles across the five diagnostic groups.

Strengths

• Includes data from 346 participants, providing a substantive sample for analysis.
• Combines two analytical methods: gas chromatography–mass spectrometry (GC-MS) and a chemiresistive electronic nose sensor array.
• Covers five distinct diagnostic groups for comparative study of oral diseases.

Limitations

• The dataset is small at 2.1 MB, limiting the scale for complex model training.
• Geographic and demographic details of the 346 participants are unknown, which may affect generalizability.

Provenance

Source

figshare, authored by maisa Haiek.

Collection Method

Breath samples analyzed using GC-MS and an electronic nose sensor array; multivariate statistical and machine-learning methods applied.

Time Range

null

Freshness

null

Geography

null