Assessment of artificial intelligence to detect gasoline in fire debris using HS‐SPME‐GC/MS and transfer learning

• Published in: Journal of forensic sciences Vol. 69; no. 4; pp. 1222 - 1234
• Main Authors: Huang, Ting‐Yu, Chung Yu, Jorn Chi
• Format: Journal Article
• Language: English
• Published: United States Wiley Subscription Services, Inc 01.07.2024
• Subjects: Accuracy; Artificial intelligence; Artificial neural networks; Chemical composition; convolutional neural network and CNN; Debris; fire debris analysis; Forensic chemistry; Gas chromatography; gas chromatography–mass spectrometry and GCMS; Gasoline; heatmaps; Image classification; Machine learning; solid phase microextraction and SPME; Substrates; transfer learning; Workflow; fire debris analysis; solid phase microextraction and SPME; convolutional neural network and CNN; transfer learning; heatmaps; gas chromatography–mass spectrometry and GCMS

Due to the complex nature of the chemical compositions of ignitable liquids (IL) and the interferences from fire debris matrices, interpreting chromatographic data poses challenges to analysts. In this work, artificial intelligence (AI) was developed by transfer learning in a convolutional neural ne...