Recent advances in digital pathology and computational imaging have significantly enhanced our ability to analyze complex tissue structures in neuroscience and beyond. High-resolution whole-slide images (WSIs), combined with machine learning and deep learning techniques, are enabling more objective, reproducible, and scalable insights into disease mechanisms. Despite these advancements, traditional computational methods often suffer from limited generalizability, heavy annotation requirements, and a lack of interpretability in clinical contexts.

In parallel, the emergence of large-scale foundation models in pathology, particularly pre-trained vision transformers and self-supervised learning approaches, has opened new avenues for more robust and transferable analysis across different datasets and diseases. While their application in neuroscience is still emerging, these models offer valuable tools for accelerating biomarker discovery, improving clinical decision support, and understanding disease pathology at scale. There is a growing need to explore how these models and new algorithmic frameworks can be effectively applied to both general and neuroscience-specific pathological image analysis.

This Research Topic aims to bring together recent advances in computational pathology with a special focus on neuroscience applications, while remaining open to broader contributions that demonstrate methodological or clinical innovation. We welcome original research, reviews, and perspectives that cover the full spectrum of pathological image analysis from algorithm development to clinical translation and biological discovery.

We particularly encourage submissions addressing:

- Robust and interpretable deep learning methods for analyzing histopathological images

- Novel approaches for biomarker discovery and disease stratification

- Integration of computational pathology with clinical outcomes and multi-omics data

- Development and validation of foundation models or transfer learning strategies across diverse pathology domains

- New pipelines or tools for efficient annotation, data standardization, and deployment in real-world scenarios

We invite high-quality submissions in (but not limited to) the following areas:

- Computational methods for histopathological image analysis in neuroscience and general pathology

- Disease classification, prognosis, and subtype discovery using AI-driven approaches

- Biomarker discovery from pathology images and integration with genomics/clinical data

- Cross-modal learning: pathology images with radiology, omics, or text

- Dataset curation, annotation strategies, and reproducible pipelines

- Clinical validation, interpretability, and deployment of AI in pathology workflows

- Federated learning or privacy-preserving analysis for multi-center collaborations

All submissions should clearly articulate their relevance to pathology and demonstrate a contribution to advancing the field, whether through algorithmic novelty, clinical insight, or translational value.

Are you interested in contributing a manuscript to this Research Topic?

Advanced Computational Pathology in Neuroscience: From Algorithms to Clinical Applications