Electrochemical-assisted scattering imaging system for lymphoma cell classification using machine learning.

Lymphoma is one of the most common malignancies globally, making early diagnosis crucial for improving survival. This study introduces an electrochemical-assisted scattering imaging system (ESIS) for lymphoma cell classification. The system integrates scattering imaging with electrochemical measurements, using a fiber-optic probe for scattering excitation and a 3D rGO-TiC-MWCNTs composite electrode to simultaneously monitor HO release. Data from these modalities are combined with an SVM algorithm, improving classification performance significantly, with the AUC for HMy2.CIR cells increased from 0.79 to 0.97. The dual-modality approach achieved 90% accuracy, outperforming scattering imaging alone. This method enhances lymphoma subtype differentiation and shows promise for personalized cancer therapies.