Publications by authors named "Evangelos Tzoras"
Developing AI models for digital pathology has traditionally relied on single-scale analysis of histopathology slides. However, a whole slide image is a rich digital representation of the tissue, captured at various magnification levels. Limiting our analysis to a single scale overlooks critical information, spanning from intricate high-resolution cellular details to broad low-resolution tissue structures.
View Article and Find Full Text PDFMultiresolution Self-Supervised Feature Integration via Attention Multiple Instance Learning for Histopathology Analysis.
Annu Int Conf IEEE Eng Med Biol Soc
July 2023
Article Synopsis
- Digital histopathology image analysis is gaining attention for automating diagnostics and creating new prognostic biomarkers, but current models usually focus on either high or low-resolution features.
- Most existing models lose important tumor diversity information during training because they are patch-based and weakly supervised.
- This study introduces a new multiresolution framework that effectively captures both cellular and contextual features, outperforming traditional models in breast cancer grading, with the best performance achieved using a multiplication-based approach (AUC=0.864).
Purpose: High levels of tumor-infiltrating lymphocytes (TILs) are associated with better outcomes in early breast cancer and higher pathological response rates to neoadjuvant chemotherapy especially in the triple-negative (TNBC) and HER2+ subtypes. However, the dynamic changes in TILs levels after neoadjuvant treatment (NAT) are less studied. This systematic review and meta-analysis aimed to investigate the patterns and role of TILs dynamics change in early breast cancer patients receiving NAT.
View Article and Find Full Text PDFThe tumor immune microenvironment (TIME) is an important player in breast cancer pathophysiology. Surrogates for antitumor immune response have been explored as predictive biomarkers to immunotherapy, though with several limitations. Immunohistochemistry for programmed death ligand 1 suffers from analytical problems, immune signatures are devoid of spatial information and histopathological evaluation of tumor infiltrating lymphocytes exhibits interobserver variability.
View Article and Find Full Text PDFArticle Synopsis
- Although classical Hodgkin lymphoma (cHL) is generally treatable, 20-30% of patients still face treatment failure, leading to further relapse or progression, especially after therapies like salvage chemotherapy and autologous stem cell transplantation (autoSCT).
- Recent immunotherapy advancements, including drugs like Brentuximab vedotin (BV) and immune checkpoint inhibitors (CPIs) such as nivolumab and pembrolizumab, have shown promising results for patients with relapsed/refractory cHL, providing prolonged remissions or disease stabilization even after multiple previous treatments.
- These newer therapies are being explored for use earlier in treatment protocols, with BV now approved for initial treatment of advanced stages of cHL and ongoing trials