Artificial intelligence (AI) and machine learning (ML) are rapidly advancing fields within computer science, driving significant progress in cancer diagnostics. Various ML models have been developed to assist diagnosis, guide therapy decisions, and facilitate early disease detection. In this review, we discuss diverse AI and ML approaches and critically evaluate their applications and limitations in pancreatic cancer histopathology, diagnostics, and biomarker discovery.

• Klíčová slova
• Artificial intelligence, Biomarker discovery, Machine learning, Multimodal learning, Pancreatic cancer, Pancreatic ductal adenocarcinoma,
• Publikační typ
• časopisecké články MeSH
• přehledy MeSH

BACKGROUND: Multiple myeloma (MM) represents the second most common hematological malignancy characterized by the infiltration of the bone marrow by plasma cells that produce monoclonal immunoglobulin. While the quality and length of life of MM patients have significantly increased, MM remains a hard-to-treat disease; almost all patients relapse. As MM is highly heterogenous, patients relapse at different times. It is currently not possible to predict when relapse will occur; numerous studies investigating the dysregulation of non-coding RNA molecules in cancer suggest that microRNAs could be good markers of relapse. RESULTS: Using small RNA sequencing, we profiled microRNA expression in peripheral blood in three groups of MM patients who relapsed at different intervals. In total, 24 microRNAs were significantly dysregulated among analyzed subgroups. Independent validation by RT-qPCR confirmed changed levels of miR-598-3p in MM patients with different times to relapse. At the same time, differences in the mass spectra between groups were identified using matrix-assisted laser desorption/ionization time of flight mass spectrometry. All results were analyzed by machine learning. CONCLUSION: Mass spectrometry coupled with machine learning shows potential as a reliable, rapid, and cost-effective preliminary screening technique to supplement current diagnostics.

• Klíčová slova
• Liquid biopsy, MALDI-TOF MS, Machine learning, Multiple myeloma, Relapse, Small RNA seq, microRNA,
• Publikační typ
• časopisecké články MeSH