Application of an artificial intelligence for quantitative analysis of endothelial capillary beds in vitro.

Objective: BACKGROUND: The use of endothelial cell cultures has become fundamental to study angiogenesis. Recent advances in artificial intelligences (AI) offer opportunities to develop automated assessment methods in medical research, analyzing larger datasets. Objective: OBJECTIVE: The aim of this study was to compare the application of AI with a manual method to morphometrically quantify in vitro angiogenesis. Objective: METHODS: Co-cultures of human microvascular endothelial cells and fibroblasts were incubated mimicking endothelial capillary-beds. An AI-software was trained for segmentation of endothelial capillaries on anti-CD31-labeled light microscope crops. Number of capillaries and branches and average capillary diameter were measured by the AI and manually on 115 crops. Objective: RESULTS: The crops were analyzed faster by the AI than manually (3 minutes vs 1 hour per crop). Using the AI, systematically more capillaries (mean 48/mm2 vs 27/mm2) and branches (mean 23/mm2 vs 11/mm2) were counted than manually. Both methods had a strong linear relationship in counting capillaries and branches (r-capillaries = 0.88, r-branches = 0.89). No correlation was found for measurements of the diameter (r-diameter = 0.15). Objective: CONCLUSIONS: The present AI reduces the time required for quantitative analysis of angiogenesis on large datasets, and correlates well with manual analysis.

Microvascular changes following exposure to iodinated contrast media in vitro. A qualitative comparison to serum creatinine concentrations in post-cardiac catheterization patients.

INTRODUCTION: Contrast-associated acute kidney injury (CA-AKI) is characterized as a loss of renal function following radiological contrast media administration. While all contrast media induce variable changes in microvascular endothelial cells in vitro, only few studies report clinical significance of their findings. A comprehensive assessment of the effect of iodinated contrast media on the renal function in vitro and in vivo is essential. The aim of our study was to morphometrically quantify the effect of two different contrast media (Iobitridol and Iodixanol) on vascular endothelial capillaries in vitro and to analyze their effect on the renal function of patients who underwent cardiac catheterization including the intra-arterial administration of contrast media, by measuring serum creatinine concentration (SCr), a byproduct of muscle metabolism, primarily excreted by the kidneys. Our hypothesis suggests that conducting a qualitative comparison of both outcomes will enable identification of differences and similarities between in vitro and in vivo exposure. MATERIAL AND METHODS: In vitro, co-cultures of human dermal fibroblasts and human dermal microvascular endothelial cells forming capillary beds were exposed to a mixture of phosphate buffered saline and either Iobitridol, Iodixanol, or one of their supplements EDTA or Trometamol for 1.5 or 5 min. Negative control co-cultures were exposed exclusively to phosphate buffered saline. Co-cultures were either directly fixed or underwent a regeneration time of 1, 3 or 7 days. An artificial intelligence software was trained for detection of labeled endothelial capillaries (CD31) on light microscope images and measurements of morphometric parameters. In vivo, we retrospectively analyzed data from patients who underwent intra-arterial administration of contrast media and for whom SCr values were available pre- and post-contrast exposition (1, 3, and 7 days following procedure). Temporal development of SCr and incidence of CA-AKI were assessed. Both exposure types were qualitatively compared. RESULTS: In vitro, Iobitridol, Iodixanol and EDTA induced a strong decrease of two morphometric parameters after 3 days of regeneration. In vivo, a significant increase of SCr and incidence of CA-AKI was observed 3 days following procedure in the post-contrast media patients. No difference was observed between groups. DISCUSSION: Two of the morphometric parameters were inversely proportional to the SCr of the patients. If the endothelial damages observed in vitro occur in vivo, it may result in renal hypoxia, inducing a loss of kidney function clinically translated into an increase of SCr. Further development of our in vitro model could allow closer replication of the internal structure of a kidney and bridge the gap between in vitro studies and their clinical findings.