Rapid progress of AI technology in the life science area is observed in recent years. Convolutionalneural network (CNN) models were successfully applied for the localization and classification of cellson microscopic images. Induced pluripotent stem cells are one of the most important innovations inbiomedical research and are widely used, e.g. in regenerative medicine, drug screening, and diseasemodeling. However, assessment of cell cultures’ quality requires trained personnel, is timeconsumingand hence expensive. Fluorescence microscope images of human induced pluripotentstem‐hepatocytes (hiPS‐HEPs) derived from three human induced pluripotent stem cell (hiPSC) lineswere taken daily from day 1 until day 22 of differentiation. The cells from day 1 to 14 were classifiedas ´Early differentiation´, and above day 16 as ´Late differentiation´. In this study, it wasdemonstrated that a CNN‐based model can be trained with simple fluorescence microscope imagesof human induced pluripotent stem‐hepatocytes, and then used to predict with high accuracy(96.4%) the differentiation stage of an independent new set of images.
| Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:his-19420 |
| Date | January 2020 |
| Creators | Marzec-Schmidt, Katarzyna |
| Publisher | Högskolan i Skövde, Institutionen för biovetenskap |
| Source Sets | DiVA Archive at Upsalla University |
| Language | English |
| Detected Language | English |
| Type | Student thesis, info:eu-repo/semantics/bachelorThesis, text |
| Format | application/pdf |
| Rights | info:eu-repo/semantics/openAccess |
Page generated in 0.0021 seconds