Premium
Optimal selection of biological tissue using the energy dissipated in the first loading cycle
Author(s) -
Rojo Francisco J.,
Páez José M. García,
JorgeHerrero Eduardo,
Atienza José M.,
Millán Isabel,
Rocha Aurora,
de Córdova Alfonso Hoyos Fernández,
Guinea Gustavo V.
Publication year - 2010
Publication title -
journal of biomedical materials research part b: applied biomaterials
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.665
H-Index - 108
eISSN - 1552-4981
pISSN - 1552-4973
DOI - 10.1002/jbm.b.31732
Subject(s) - hysteresis , energy (signal processing) , dissipation , selection (genetic algorithm) , materials science , mathematics , structural engineering , statistics , computer science , engineering , physics , thermodynamics , artificial intelligence , quantum mechanics
Calf pericardium, similar to that used in the manufacturing of prosthetic valve cusps, was fatigue tested. After six batches of 100 cycles of 1 MPa of loading pressure, half of the samples broke. The mean energy dissipated in the first cycle by the surviving samples was 0.16 J, which is lower than the 0.28 J dissipated by the specimens that broke ( p = 0.005). The hysteresis of the first cycle was characteristic and different from the following ones and correlated superbly with fatigue resistance. Setting a threshold value for the energy of the first cycle of 0.20 J, the performance index (the percentage of true predictions) was almost 80%, and the area under the ROC curve was 0.823 (maximum value is 1). When including the mean thickness in the selection parameters, as an indirect measure of the specimen mass, the performance index grew over 95%, meaning that the error of the predictions was less than 5%. Combining both parameters in one, a high performance index is maintained at 87.5% and the area under the ROC curve increases to 0.917. This non‐destructive method should help optical methods in the process of selecting the most appropriate and homogenous biological material. © 2010 Wiley Periodicals, Inc. J Biomed Mater Res Part B: Appl Biomater, 2010.