Application and comparison of supervised learning strategies to classify polarity of epithelial cell spheroids in 3D culture
3D culture systems that allow generation of monolayered epithelial cell spheroids are widely used to study epithelial function in vitro. Epithelial spheroid formation is applied to address cellular consequences of (mono)-genetic disorders, i.e. ciliopathies, in toxicity testing, or to develop treatm...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article (Journal) |
| Language: | English |
| Published: |
27 March 2020
|
| In: |
Frontiers in genetics
Year: 2020, Volume: 11 |
| ISSN: | 1664-8021 |
| DOI: | 10.3389/fgene.2020.00248 |
| Online Access: | Verlag, lizenzpflichtig, Volltext: https://doi.org/10.3389/fgene.2020.00248 Verlag, lizenzpflichtig, Volltext: https://www.frontiersin.org/articles/10.3389/fgene.2020.00248/full 
|
| Author Notes: | Birga Soetje, Joachim Fuellekrug, Dieter Haffner and Wolfgang H. Ziegler |
| Summary: | 3D culture systems that allow generation of monolayered epithelial cell spheroids are widely used to study epithelial function in vitro. Epithelial spheroid formation is applied to address cellular consequences of (mono)-genetic disorders, i.e. ciliopathies, in toxicity testing, or to develop treatment options aimed to restore proper epithelial cell characteristics and function. With the potential of a high-throughput method, the main obstacle to efficient application of the spheroid formation assay so far is the laborious, time-consuming and bias-prone analysis of spheroid images by individuals. Hundredths of multidimensional fluorescence images are blinded, rated by 3 persons, and subsequently, differences in ratings are compared and discussed. Here, we apply supervised learning and compare strategies based on machine learning versus deep learning. While deep learning approaches can directly process raw image data, machine learning requires transformed data of features extracted from fluorescence images. We verify the accuracy of both strategies on a validation dataset, analyse an experimental dataset, and observe that different strategies can be very accurate. Deep learning, however, is less sensitive to overfitting and experimental batch-to-batch variations thus providing a rather powerful and easily adjustable classification tool. |
|---|---|
| Item Description: | Gesehen am 26.11.2020 |
| Physical Description: | Online Resource |
| ISSN: | 1664-8021 |
| DOI: | 10.3389/fgene.2020.00248 |