About this Research Topic
The rise of digital pathology has resulted in the development of hardware, software, processes, workflows, and quality control measures for scanning, storing, analyzing, and viewing pathology slides. Moreover, whole slide images serve as the substrates for exciting possibilities in computer vision applications, including decision support for pathologists and automated standalone diagnostics. There is even hope that artificial intelligence applied to whole slide images can help fulfil the promises of precision medicine by extracting clinically actionable knowledge not seen through the pathologist’s eye. Indeed, early studies have shown that image analysis can provide prognostic information and predict therapy responses from digital slides. Despite these successes, numerous challenges remain in fully leveraging digital pathology images for medical applications and new methods are needed to overcome these hurdles.
The main objective of this Research Topic is to identify the artificial intelligence algorithms and methods needed to address the challenges that come with gigapixel-size whole-slide pathology images, marked histologic stain variation, artefacts introduced by tissue handling, and the sparsity of high-quality expert annotations. While the applications of the tools may include pathologic diagnosis, prognostication, or biomarker discovery, artificial intelligence algorithms should be the centrepiece of the manuscript.
• Methods of utilizing gigapixel-size digital pathology images, including methods not requiring tiling whole slide images;
• Methods of Data Augmentation or stain normalization tailored for handling the known problem of histologic staining variation;
• Methods to combat sparse training data such as generative adversarial networks or the use of “weak” slide-level diagnosis labels for training image segmentation tasks;
• Methods to detect and remove tissue artefacts which can degrade model performance.
The main objective of this Research Topic is to identify the artificial intelligence algorithms and methods needed to address the challenges that come with gigapixel-size whole-slide pathology images, marked histologic stain variation, artefacts introduced by tissue handling, and the sparsity of high-quality expert annotations. While the applications of the tools may include pathologic diagnosis, prognostication, or biomarker discovery, artificial intelligence algorithms should be the centrepiece of the manuscript.
• Methods of utilizing gigapixel-size digital pathology images, including methods not requiring tiling whole slide images;
• Methods of Data Augmentation or stain normalization tailored for handling the known problem of histologic staining variation;
• Methods to combat sparse training data such as generative adversarial networks or the use of “weak” slide-level diagnosis labels for training image segmentation tasks;
• Methods to detect and remove tissue artefacts which can degrade model performance.
Keywords: Machine Learning, Computer Vision, Image Analysis, Image Processing, Digital Pathology, Computational Pathology, Histopathology, Artificial Intelligence
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.
