Biology Reference
In-Depth Information
incisal edges. A maximum error of
0.4 mm per point would still leave each landmark on
the incisal surface in all cases. Of course there might also be a lateral component, a sliding
of points along the top margin of the tooth, that contributes to this RMS scatter. The RMS
scatter summed up over all 60 points (all six anterior teeth) in the mandibular dentition,
was 3.1 mm and 4.28 mm in the maxillary dentition. Thus, the net error was 2% of total
size. The substantial difference between the two dentitions is due to the increased size of
the maxillary structure. The error as a percentage of size was similar.
An initial determination indicated five matches based on Procrustes and seven matches
based on Procrustes-SP superimposition, but seven of the specimens in these matches
were the repeated scans of the same individuals. That inclusion of repeated measurements
provided an accidental, but meaningful, test of the effectiveness of the analytic method.
Interestingly, the scans of one particular individual were taken a year apart, and were still
within twice the range of the repeated measurements RMS scatter. Thus, the differences
due to events occurring over the intervening time, in addition to retaking the cast and
rescanning that cast, were still within the range of measurement error.
After removing the repeatedly measured individuals, the matching rates based on
Procrustes and Procrustes-SP superimpositions were determined. Match rates were much
higher for maxillary dentition than for mandibular, and for each jaw, they were higher for
Procrustes superimposed data than for Procrustes-SP. Considerably lower rates were
obtained when matches were sought between maxillary and mandibular sets (
Table 14.1
).
Based on the Procrustes-SP procedure, there were two matched pairs of different speci-
mens for a total of four individuals that were not unique (in only 1099). It is also
notable that the two pairs that matched using the Procrustes-SP were not the same indivi-
duals who matched using the Procrustes method.
The dependence of match rate on measurement error is summarized in
Table 14.2
.In
actual forensic cases, the information about the incisal edges will probably not be recorded
with a resolution as high as that obtained from the three-dimensional laboratory data, so
an alteration of the match rate is expected as measurement error increases. The level of
error in repeated measurements was known to be underestimated because error in the
casting and scanning operations was not included. Clearly, the rates of matches in this
large population increases rapidly as measurement resolution or repeatability decreases,
as represented by the increased RMS errors used to calculate this table.
6
TABLE 14.1
Match Rates in the Maxillary and Mandibular Dentitions, and in Both Combined, Based on
Procrustes and Procrustes-SP Superimpositions
Procrustes
Procrustes-SP
Number of Matches Number of Individuals Number of Matches Number of Individuals
Maxillary
1691
487
763
396
Mandibular
129
131
75
83
Both
1
2
2
4
The matching criterion used was twice the RMS scatter of repeatedly measured specimens.
