Early diagnosis is critical for improving outcomes in cancer patients; however, the application of diagnostic markers derived from serum proteomic screening remains challenging. Artificial intelligence (AI), encompassing deep learning and machine learning (ML), has gained increasing prominence across various scientific disciplines. In this study, we utilized cervical cancer (CC) as a model to develop an AI-driven pipeline for the identification and validation of serum biomarkers for early cancer diagnosis, leveraging mass spectrometry-based proteomics data. By processing and normalizing serum polypeptide differential peaks from 240 patients, we employed eight distinct ML algorithms to classify and analyze these differential polypeptide peaks, subsequently constructing receiver operating characteristic (ROC) curves and confusion matrices. Key performance metrics, including accuracy, precision, recall, and F1 score, were systematically evaluated. Furthermore, by integrating feature importance values, Shapley values, and local interpretable model-agnostic explanation (LIME) values, we demonstrated that the diagnostic area under the curve (AUC) achieved by our multi-dimensional learning models approached 1, significantly outperforming the diagnostic AUC of single markers derived from the PRIDE database. These findings underscore the potential of proteomics-driven integrated machine learning as a robust strategy to enhance early cancer diagnosis, offering a promising avenue for clinical translation.

Primary cutaneous CD4⁺ small/medium-sized pleomorphic T-Cell lymphoproliferative disorder (PC-SMTLD) has been considered as a controversial dermatological disease that has been included in cutaneous T-cell lymphoma group, presenting most commonly as a solitary nodule and/or plaque with a specific and characteristic head and neck predilection. Due to the considerable overlap between PC-SMTLD and pseudolymphoma (PL), the differential diagnosis is often challenging. Methylation of DNA at position 5 of cytosine, and the subsequent reduction in intracellular 5-hydroxymethylcytosine (5-hmC) levels, is a key epigenetic event in several cancers, including systemic lymphomas. However, it has rarely been studied in cutaneous lymphomas. The authors aimed to explore the role of differential 5-hmC immunostaining as a useful marker to distinguish PC-SMTLD from PL. Retrospective case series study with immunohistochemical and immunofluorescence analysis of 5-hmC was performed in PL and PC-SMTLD. Significant decrease of 5-hmC nuclear staining was observed in PC-SMTLD when compared with PL (p < 0.0001). By semi-quantitative grade integration, there were statistical differences in the final 5-hmC scores in the two study groups. The IF co-staining of 5-hmC with CD4 revealed a decrease of 5-hmC in CD4⁺ lymphocytes of PC-SMTLD. The small clinical sample size of the study. The immunorreactivity of 5-hmC in CD4⁺ lymphocytes was highly suggestive of a benign process as PL. Furthermore, the decrease of 5-hmC nuclear staining in PC-SMTLD indicated its lymphoproliferative status and helped to make the differential diagnosis with PL.