Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Minimally invasive implantation of scalable high-density cortical microelectrode arrays for multimodal neural decoding and stimulation.

Nature biomedical engineering·2025

Same author

A self-supervised framework for laboratory data imputation in electronic health records.

Communications medicine·2025

Same author

The Role of PSMA PET Imaging in Prostate Cancer: Current Applications and Future Directions.

Current urology reports·2025

Same author

Triaging mammography with artificial intelligence: an implementation study.

Breast cancer research and treatment·2025

Same author

First-in-human experience performing high-resolution cortical mapping using a novel microelectrode array containing 1024 electrodes.

Journal of neural engineering·2025

Same author

Efficient and Generalizable Prediction of Molecular Alterations in Multiple-Cancer Cohorts Using Hematoxylin and Eosin Whole Slide Images.

Modern pathology : an official journal of the United States and Canadian Academy of Pathology, Inc·2024

Search research articles

Related Experiment Video

Updated: Dec 31, 2025

Introduction of an Integrated Pathology Image Management, Artificial Intelligence, and Reporting System

Introduction of an Integrated Pathology Image Management, Artificial Intelligence, and Reporting System

Published on: July 11, 2025

Whole-Slide Image Focus Quality: Automatic Assessment and Impact on AI Cancer Detection.

Timo Kohlberger¹, Yun Liu¹, Melissa Moran¹

¹Google Health, Palo Alto, CA, USA.

Journal of Pathology Informatics

|January 11, 2020

Summary

A new deep learning tool, ConvFocus, accurately identifies and quantifies out-of-focus (OOF) regions in digital pathology slides. This technology can prevent workflow delays and errors caused by OOF artifacts in digitized slides.

Keywords:

Computer-aided diagnostics digital pathology focus quality out-of-focus quality control

More Related Videos

Automated Slide Scanning and Segmentation in Fluorescently-labeled Tissues Using a Widefield High-content Analysis System

Automated Slide Scanning and Segmentation in Fluorescently-labeled Tissues Using a Widefield High-content Analysis System

Published on: May 3, 2018

Artificial Intelligence Approaches to Assessing Primary Cilia

Artificial Intelligence Approaches to Assessing Primary Cilia

Published on: May 1, 2021

Related Experiment Videos

Last Updated: Dec 31, 2025

Introduction of an Integrated Pathology Image Management, Artificial Intelligence, and Reporting System

Introduction of an Integrated Pathology Image Management, Artificial Intelligence, and Reporting System

Published on: July 11, 2025

Automated Slide Scanning and Segmentation in Fluorescently-labeled Tissues Using a Widefield High-content Analysis System

Automated Slide Scanning and Segmentation in Fluorescently-labeled Tissues Using a Widefield High-content Analysis System

Published on: May 3, 2018

Artificial Intelligence Approaches to Assessing Primary Cilia

Artificial Intelligence Approaches to Assessing Primary Cilia

Published on: May 1, 2021

Area of Science:

Digital pathology
Computational pathology
Medical image analysis

Background:

Digital pathology offers remote access and advanced image analysis but is susceptible to out-of-focus (OOF) artifacts.
OOF artifacts can cause workflow delays due to manual review and potential rescanning.
Manual screening for partial OOF is impractical, necessitating automated solutions.

Purpose of the Study:

To develop and evaluate ConvFocus, a convolutional neural network for localizing and quantifying OOF regions in digitized slides.
To assess the generalization of ConvFocus across various tissue types, stains, and scanners.
To investigate the impact of OOF on downstream image analysis algorithms.

Main Methods:

Developed ConvFocus using a semi-synthetic OOF data generation process.
Evaluated ConvFocus on seven slides across three tissue and three stain types, digitized by two scanners.
Compared ConvFocus predictions with pathologist annotations on 514 regions and 21 z-stack WSIs.

Main Results:

ConvFocus achieved high correlation with pathologist grading (Spearman coefficients 0.81 and 0.94) on two scanners.
The algorithm accurately reproduced OOF patterns from z-stack data.
Increasing OOF levels consistently decreased the performance of a metastatic breast cancer detector.

Conclusions:

Automated OOF detection can enable pre-review rescans, mitigating digitization focus issues.
ConvFocus, trained on semi-synthetic data, generalizes well to real-world OOF across diverse conditions.
Quantitative OOF maps can prevent misclassifications by image analysis algorithms, improving diagnostic accuracy.