Use of pattern recognition to classify beef cardiovascular tissues on the basis of their trace metal compositions

✍ Scribed by John Webb; Katharine A. Kirk; William Niedermeier; James H. Griggs; Malcolm E. Turner; Thomas N. James

• Publisher: Elsevier Science
• Year: 1976
• Weight: 347 KB
• Volume: 5
• Category: Article
• ISSN: 0006-3061
• DOI: 10.1016/s0006-3061(00)82024-4

✦ Synopsis

The pattern recognition procedure of discriminant analysis has been used to characterize the trace metal profiles created by the concentrations of 8 trace metals in 15 anatomic sites of beef heart tissue. Metals analyzed were copper. tin, lead, molybdenum, strontium, cesium, barium, and amminum. Anatomic sites sampIed included main pulmonary artery, aorta, mitral and tricuspid valves, left and tight coronary arteries, OS cordis, right atrium, left atrial appendage, crista supraventricularis, left bundle branch, free wall of the right and left ventricles, interventricular septum, and papillary muscle of the left ventricle. The striking features of the data were: (1) All specimens of the mitral valve, tricuspid valve, and OS cordis were ambiguously described by their trace metal profiles; (2) the four blood vessels constituted two groups of two tissues each (aorta, main pulmonary artery; left and right coronary arteries); (3) tissues derived from ordinary and specialized myocardium were quite different from blood vessels, heart valves and OS cordis. Using these profdes, 85% of the specimens analyzed were correctly classified by discriminant analysis with respect to their anatomic origin.