About this Research Topic
Deep learning has recently revolutionized the field of computational pathology, offering significant advancements in health informatics applications. It has become the standard in digital pathology, aiding in diagnosis, prognosis, and treatment planning. Despite these advancements, a critical challenge persists: deep neural networks require vast amounts of data for effective model training, which is often unavailable due to privacy concerns and budget constraints. This limitation hinders the application of cutting-edge AI and deep learning algorithms in both research and clinical settings. Furthermore, the integration of multi-modal data—encompassing medical images, omics data, clinical notes, and lab results—remains under-explored, with many existing methods failing to leverage the comprehensive information available across these modalities. Recent studies have begun to address data scarcity and heterogeneous data integration, yet numerous challenges remain, particularly in digital pathology. Unique issues such as high image dimensions, batch effects from inconsistent imaging procedures, and the high cost of data collection and annotation exacerbate the difficulties of training deep models with limited data and integrating multi-modal data.
This Research Topic aims to provide a platform for computational pathology researchers to present innovative methods and practical applications, fostering collaboration among diverse groups. The primary objective is to address the challenges of data scarcity and heterogeneity in computational pathology by exploring data/annotation-efficient modeling and the integration of multi-modal data. Key questions include how to effectively train models with limited data and how to integrate diverse data types to enhance diagnostic and prognostic capabilities. The research will test hypotheses related to the efficacy of various AI approaches in overcoming these challenges, ultimately aiming to advance the field of computational pathology.
To gather further insights in the realm of computational pathology, we welcome Original Research, Review, and Case Report articles addressing, but not limited to, the following themes:
• Weakly, active, semi-, and self-supervised learning in digital pathology.
• Domain adaptation and generalization for cross-domain pathology data.
• Data synthesis, including pathology image generation and translation.
• Federated learning to bridge pathology data silos.
• Multi-modal learning with different types of medical data.
• Geometric deep learning, including graph neural networks for computational pathology.
This Research Topic aims to provide a platform for computational pathology researchers to present innovative methods and practical applications, fostering collaboration among diverse groups. The primary objective is to address the challenges of data scarcity and heterogeneity in computational pathology by exploring data/annotation-efficient modeling and the integration of multi-modal data. Key questions include how to effectively train models with limited data and how to integrate diverse data types to enhance diagnostic and prognostic capabilities. The research will test hypotheses related to the efficacy of various AI approaches in overcoming these challenges, ultimately aiming to advance the field of computational pathology.
To gather further insights in the realm of computational pathology, we welcome Original Research, Review, and Case Report articles addressing, but not limited to, the following themes:
• Weakly, active, semi-, and self-supervised learning in digital pathology.
• Domain adaptation and generalization for cross-domain pathology data.
• Data synthesis, including pathology image generation and translation.
• Federated learning to bridge pathology data silos.
• Multi-modal learning with different types of medical data.
• Geometric deep learning, including graph neural networks for computational pathology.
Keywords: Deep learning, Weakly-supervised learning, Self-supervised learning, Federated learning, Multi-modal learning, Generative models, Graph neural networks
Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.
