Citation

dos Santos LS, Vieira Muterlle P. J. Forensic Sci. 2018; 63(6): 1846-1853.

Affiliation

Materials Laboratory, Brasilia University, Darcy Ribeiro Campus, Bloco SG9, Asa Norte Brasília, CEP: 70910-900, DF, Brazil.

Copyright

(Copyright © 2018, American Society for Testing and Materials, Publisher John Wiley and Sons)

DOI

10.1111/1556-4029.13771

PMID

29495115

Abstract

Factors influencing effectiveness of automated comparisons, test-fired bullets, and cartridge cases from 0.38 Special revolvers were logged into the Evofinder^®Ballistic ID System. Tests were performed as follows: First test correlated test-fires of the same type, second test compared different types of ammunition components, third test replicated the second test in a larger database, and fourth test replicated the third test with students having no previous firearm identification experience. System effectiveness with projectiles in the first test was 0.89. With cartridge cases, effectiveness was 0.79 with combined results, but analysis of separate results by breech face and firing pin revealed low effectiveness by breech face (0.40). In the second, third, and fourth tests, effectiveness with projectiles were 0.61, 0.51, and 0.44. In addition, these tests had effectiveness with cartridge cases equivalent to 0.55, 0.43, and 0.44.

RESULTS are useful to establish routine protocols, system improvements, or comparative assessment of other electronic systems.

© 2018 American Academy of Forensic Sciences.

Language: en

Keywords

automated comparison; ballistic comparison; correlation quality; effectiveness criterion; firearms identification; forensic science