Preeclampsia case study: enhancing reproductive disease research with AI

Written by Aiforia | Sep 1, 2020 2:43:30 PM

Preeclampsia, a hypertensive pregnancy disorder, is a major cause of maternal and perinatal mortality and morbidity.¹ Satu Wedenoja, MD, PhD, a Specialist in Obstetrics and Gynecology at Helsinki University Hospital, and her team studied the link between preeclampsia, fetal human leukocyte antigen G (HLA-G), and other genes regulating maternal immune responses.

Using AI for image analysis

The team decided to use Aiforia software "to confirm earlier findings of HLA-G down-regulation in the preeclamptic placenta and show that the observed low level of expression is not related to the number of trophoblasts in the samples", as Satu explains.

Two convolutional neural network-based algorithms were trained using Aiforia^® Create to identify anti-CK7 and anti-HLA-G immunostained tissue areas separately. Each algorithm consists of two algorithms in sequence, the first segmenting the extravillous tissue areas and the second identifying trophoblasts that exhibit a strong HLA-G or CK7 expression, respectively.

Read Satu's interview below, and read their full publication here.

What did Aiforia allow you to do that you could not do with traditional tools?

"Aiforia was excellent in providing HLA-G/CK7 ratios for protein expression, and we were able to show a nice correlation between HLA-G/CK7 protein and RNA expression levels. With Aiforia analysis, we obtained novel data by demonstrating that HLA-G protein down-regulation in the preeclamptic placentas was not explained by the lower number of trophoblasts in the placental samples.

Quantitative analysis of protein expression also helped us demonstrate the association between HLA-G protein expression and fetal sex, which supported the overall conclusions of our study."

What is the biggest benefit of using Aiforia?

"Aiforia allowed us to count the number of HLA-G positive cells on the placental tissue sections. Moreover, we were able to calculate the proportion of HLA-G positive cells in comparison with all placental epithelial cells on the slides, stained by the trophoblast marker CK7."