Automation of hemocompatibility analysis using image segmentation and supervised classification

Abstract The hemocompatibility of blood-contacting medical devices remains one of the major challenges in biomedical engineering and makes research in the field of new and improved materials inevitable. However, current in-vitro test and analysis methods are still lacking standardization and comparability, which impedes advances in material design. For example, the optical platelet analysis of material in-vitro hemocompatibility tests is carried out manually or semi-manually by each research group individually. As a step towards standardization, this paper proposes an automation approach for the optical platelet count and analysis. To this end, fluorescence images are segmented using Zach’s convexification of the multiphase-phase piecewise constant Mumford–Shah model. The non-background components then need to be classified as platelet or no platelet. For this purpose, a supervised random forest is applied to feature vectors derived from the components using features like area, perimeter and circularity. With an overall high accuracy ( >  93 %) and low error rates ( ≤ 5 %), the random forest achieves reliable results. This is supported by high areas under the receiver–operator characteristic curve ( ≥ 0.94) and the prediction–recall curve ( ≥ 0.77), respectively. We developed a novel method for a fast, user-independent and reproducible analysis of material hemocompatibility tests. The automatized analysis method overcomes the current obstacles in the way of standardized in-vitro material testing and is therefore a unique and powerful tool for advances in biomaterial research.