Explainable AI (xAI) for Anatomic Pathology.

  title={Explainable AI (xAI) for Anatomic Pathology.},
  author={Akif Burak Tosun and Filippo Pullara and Michael J. Becich and D. Lansing Taylor and Jeffrey L. Fine and S Chakra Chennubhotla},
  journal={Advances in Anatomic Pathology},
Pathologists are adopting whole slide images (WSIs) for diagnosis, thanks to recent FDA approval of WSI systems as class II medical devices. In response to new market forces and recent technology advances outside of pathology, a new field of computational pathology has emerged that applies artificial intelligence (AI) and machine learning algorithms to WSIs. Computational pathology has great potential for augmenting pathologists' accuracy and efficiency, but there are important concerns… 

Requirements for implementation of artificial intelligence in the practice of gastrointestinal pathology

The aim of this review is to present challenges in the process of AI development, validation, and regulation that should be overcome for its implementation in real-life GI pathology practice.

Mini Review: The Last Mile—Opportunities and Challenges for Machine Learning in Digital Toxicologic Pathology

The opportunities and challenges for applying ML (the study of computer algorithms that are able to learn from example data and extrapolate the learned information to unseen data) algorithms in toxicologic pathology and how regulatory bodies are navigating this rapidly evolving field are considered.

Whole-slide imaging, tissue image analysis, and artificial intelligence in veterinary pathology: An updated introduction and review

This review focuses on an updated introduction to the basics of digital pathology and image analysis and introduces emerging topics around artificial intelligence and machine learning.

Built to last? Reproducibility and Reusability of Deep Learning Algorithms in Computational Pathology

This review highlights candidates for reproducible and reusable algorithms in computational pathology, and provides a list of reusable data handling tools and a detailed overview of the publications together with the criteria for reproducibility and reusability.

Current Challenges and Future Opportunities for XAI in Machine Learning-Based Clinical Decision Support Systems: A Systematic Review

An overall distinct lack of application of XAI is found in the context of CDSS and, in particular, a lack of user studies exploring the needs of clinicians is found.

From What to Why, the Growing Need for a Focus Shift Toward Explainability of AI in Digital Pathology

In this review, some considerations that should be made in order to develop models with a focus on explainability are detailed.

Lizard: A Large-Scale Dataset for Colonic Nuclear Instance Segmentation and Classification

A multi-stage annotation pipeline is proposed to enable the collection of large-scale datasets for histology image analysis, with pathologist-in-the-loop refinement steps, and generates the largest known nuclear instance segmentation and classification dataset, containing nearly half a million labelled nuclei in H&E stained colon tissue.

CoNIC: Colon Nuclei Identification and Counting Challenge 2022

The CoNIC challenge encourages researchers to develop algorithms that perform segmentation, classification and counting of nuclei within the current largest known publicly available nuclei-level dataset in CPath, containing around half a million labelled nuclei.

Integrating artificial intelligence in pathology: a qualitative interview study of users’ experiences and expectations

Recent progress in the development of artificial intelligence (AI) has sparked enthusiasm for its potential use in pathology. As pathology labs are currently starting to shift their focus towards AI



Whole-slide Imaging: Routine Pathologic Diagnosis

It is evident that several areas still require attention and careful consideration before more widespread clinical adoption of WSI takes place, including regulatory issues, development of standards of practice and validation guidelines, workflow modifications, as well as defining situations where WSI technology will really improve practice in a cost-effective way.

Clinical-grade computational pathology using weakly supervised deep learning on whole slide images

A multiple instance learning-based deep learning system that uses only the reported diagnoses as labels for training, thereby avoiding expensive and time-consuming pixel-wise manual annotations, and has the ability to train accurate classification models at unprecedented scale.

Computational Pathology: A Path Ahead.

The scope and needs within the new discipline of computational pathology are defined, including clearly defining the field and articulating its value propositions, and it is decided that computational pathology should serve as a hub for data-related research in health care systems.

21st century workflow: A proposal

  • J. Fine
  • Computer Science
    Journal of pathology informatics
  • 2014
Digital pathology is rapidly developing, but early systems have been slow to gain traction outside of niche applications such as: Second-opinion telepathology, immunostain interpretation, and

Pathologists' Computer-Assisted Diagnosis: A Mock-up of a Prototype Information System to Facilitate Automation of Pathology Sign-out.

A prototype of pCAD for the sign-out of a breast cancer specimen is described, in detail, to spur discussion and to help usher in the next era of digitization for pathologists by providing desperately needed and long-awaited automation.

Computational Pathology: Challenges and Promises for Tissue Analysis

Clinical Applications of Whole-slide Imaging in Anatomic Pathology

The various niche applications that initiated awareness of whole-slide imaging, provide examples of clinical use cases, and discuss the requirements and challenges for full adoption in clinical diagnosis are reviewed.

Digital Imaging Applications in Anatomic Pathology

The future of digital imaging in pathology is very promising and new applications in the fields of automated quantification and interpretation are likely to have profound long-term influence on the practice of anatomic pathology.

Detecting Cancer Metastases on Gigapixel Pathology Images

This work presents a framework to automatically detect and localize tumors as small as 100 x 100 pixels in gigapixel microscopy images sized 100,000 x100,000 pixels and achieves image-level AUC scores above 97% on both the Camelyon16 test set and an independent set of 110 slides.