Search

Background: Cardiac MRI derived biventricular mass and function parameters, such as end-systolic volume (ESV), end-diastolic volume (EDV), ejection fraction (EF), stroke volume (SV), and ventricular mass (VM) are clinically well established. Image segmentation can be challenging and time-consuming, due to the complex anatomy of the human heart. Objectives: This study introduces $��$-net (/nju:n$\varepsilon$t/) -- a deep learning approach allowing for fully-automated high quality segmentation of right (RV) and left ventricular (LV) endocardium and epicardium for extraction of cardiac function parameters. Methods: A set consisting of 253 manually segmented cases has been used to train a deep neural network. Subsequently, the network has been evaluated on 4 different multicenter data sets with a total of over 1000 cases. Results: For LV EF the intraclass correlation coefficient (ICC) is 98, 95, and 80 % (95 %), and for RV EF 96, and 87 % (80 %) on the respective data sets (human expert ICCs reported in parenthesis). The LV VM ICC is 95, and 94 % (84 %), and the RV VM ICC is 83, and 83 % (54 %). This study proposes a simple adjustment procedure, allowing for the adaptation to distinct segmentation philosophies. $��$-net exhibits state of-the-art performance in terms of dice coefficient. Conclusions: Biventricular mass and function parameters can be determined reliably in high quality by applying a deep neural network for cardiac MRI segmentation, especially in the anatomically complex right ventricle. Adaption to individual segmentation styles by applying a simple adjustment procedure is viable, allowing for the processing of novel data without time-consuming additional training.

Poster has been presented at the 26th European Congress of Pathology ObjectiveHigh quality of biological specimen collection and a repeatable process are imperative for biobanking. We present a new biobanking concept that ensures the integrity of annotation data and the biological process quality. MethodWe use an automated cryopreservation system (Smartfreezer®), storing specimens in the gasiform nitrogen phase at -180 C°. Individual specimens can be stored and retrieved. This ensures that only selected specimens are removed from the nitrogen phase. To enhance data quality and allow easier integration of data from heterogeneous resources we will implement a data management, which will use an ontology coded in Web Ontology Language 2. For this purpose we will re-use and extend a pre-existing ontology developed by UAMS in cooperation with BBMRI Sweden. ResultsOnly selected tubes are subject to a change of environment, which minimize the risks for temperature deviation and ice crystal formation. Storage in a gasiform nitrogen phase ensures complete biological preservation. Semantic infrastructure will help integrating data easily with further information belonging to the donor. Ontology will also help managing access permissions for specimens. ConclusionNew technologies as automated cryopreservation and ontology based data management help to improve the process quality and assist in producing more reliable and accessible results.