The presence of a positive deep surgical margin in tongue squamous cell carcinoma (TSCC) significantly elevates the risk of local recurrence. Therefore, a prompt and precise intraoperative assessment of margin status is imperative to ensure thorough tumor resection. In this study, we integrate Raman imaging technology with an artificial intelligence (AI) generative model, proposing an innovative approach for intraoperative margin status diagnosis. This method utilizes Raman imaging to swiftly and non-invasively capture tissue Raman images, which are then transformed into hematoxylin-eosin (H&E)-stained histopathological images using an AI generative model for histopathological diagnosis. The generated H&E-stained images clearly illustrate the tissue's pathological conditions. Independently reviewed by three pathologists, the overall diagnostic accuracy for distinguishing between tumor tissue and normal muscle tissue reaches 86.7%. Notably, it outperforms current clinical practices, especially in TSCC with positive lymph node metastasis or moderately differentiated grades. This advancement highlights the potential of AI-enhanced Raman imaging to significantly improve intraoperative assessments and surgical margin evaluations, promising a versatile diagnostic tool beyond TSCC.
Humans ; Spectrum Analysis, Raman/methods* ; Tongue Neoplasms/diagnostic imaging* ; Carcinoma, Squamous Cell/diagnostic imaging* ; Artificial Intelligence ; Margins of Excision

Accurate prediction of drug-induced adverse drug reactions (ADRs) is crucial for drug safety evaluation, as it directly impacts public health and safety. While various models have shown promising results in predicting ADRs, their accuracy still needs improvement. Additionally, many existing models often lack interpretability when linking molecular structures to specific ADRs and frequently rely on manually selected molecular fingerprints, which can introduce bias. To address these challenges, we propose ToxBERT, an efficient transformer encoder model that leverages attention and masking mechanisms for simplified molecular input line entry system (SMILES) representations. Our results demonstrate that ToxBERT achieved area under the receiver operating characteristic curve (AUROC) scores of 0.839, 0.759, and 0.664 for predicting drug-induced QT prolongation (DIQT), rhabdomyolysis, and liver injury, respectively, outperforming previous studies. Furthermore, ToxBERT can identify drug substructures that are closely associated with specific ADRs. These findings indicate that ToxBERT is not only a valuable tool for understanding the mechanisms underlying specific drug-induced ADRs but also for mitigating potential ADRs in the drug discovery pipeline.