Enhancing Confidence in Microplastic Spectral Identification via Conformal Prediction.

ABSTRACT: Microplastics are an emerging pollutant of concern, with environmental observations recorded across the world. Identifying the type of microplastic is challenging due to spectral similarities among the most common polymers, necessitating methods that can confidently distinguish plastic identities. In practice, a researcher chooses the reference vibrational spectrum that is most like the unknown spectrum, where the likeness between the two spectra is expressed numerically as the hit quality index (HQI). Despite the widespread use of HQI thresholds in the literature, acceptance of a spectral label often lacks any associated confidence. To address this gap, we apply a machine-learning framework called conformal prediction to output a set of possible labels that contain the true identity of th

SUBMITTER: Clough ME 

PROVIDER: S-EPMC11636256 | biostudies-literature | 2024 Dec

REPOSITORIES: biostudies-literature