Data set creation for supervised deep learning-based analysis of microscopic images: Review of important considerations and recommendations.

Christof A Bertram¹, Viktoria Weiss¹, Jonas Ammeling², F Maria Schabel¹, Taryn A Donovan³, Frauke Wilm^4,5, Christian Marzahl⁶, Katharina Breininger⁷, Marc Aubreville⁸

¹University of Veterinary Medicine Vienna, Vienna, Austria.

²Technische Hochschule Ingolstadt, Ingolstadt, Germany.

³The Schwarzman Animal Medical Center, New York, NY.

⁴Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany.

⁵Mira Vision Microscopy GmbH, Göppingen, Germany.

⁶Gestalt Diagnostics, Spokane, WA.

⁷Julius-Maximilians-Universität Würzburg, Würzburg, Germany.

⁸Flensburg University of Applied Sciences, Flensburg, Germany.

PubMed DOI

Abstract

Supervised deep learning (DL) receives great interest for automated analysis of microscopic images with an increasing body of literature supporting its potential. The development and testing of those DL models rely heavily on the availability of high-quality, large-scale data sets. However, creating such data sets is a complex and resource-intensive process, often hindered by challenges such as ti…

Create a free account to keep reading

Free members get 10 full research views every month across publications, clinical trials, FDA clearances, adverse events, and NIH grants. No credit card required.

Create Free Account Sign In

Want unlimited research access? See Pro plans

Data Accuracy Notice: Research intelligence on Health AI Central is aggregated from public sources (PubMed, ClinicalTrials.gov, FDA, NIH, CMS, and others) and refreshed nightly. Classifications and derived metrics are produced by automated methods described in our Methodology. We recommend verifying critical data points against the primary sources before making decisions.