Kerstin Kreutz

Accuracy of metric sex analysis of skeletal remains using Fordisc® based on a recent skull collection

It has been generally accepted in skeletal sex determination that the use of metric methods is limited due to the population dependence of the multivariate algorithms. The aim of the study was to verify the applicability of software-based sex estimations outside the reference population group for which discriminant equations have been developed. We examined 98 skulls from recent forensic cases of known age, sex, and Caucasian ancestry from cranium collections in Frankfurt and Mainz (Germany) to determine the accuracy of sex determination using the statistical software solution Fordisc® which derives its database and functions from the US American Forensic Database. In a comparison between metric analysis using Fordisc® and morphological determination of sex, average accuracy for both sexes was 86 vs 94%, respectively, and males were identified more accurately than females. The ratio of the true test result rate to the false test result rate was not statistically different for the two methodological approaches at a significance level of 0.05 but was statistically different at a level of 0.10 (p = 0.06). Possible explanations for this difference comprise different ancestry, age distribution, and socio-economic status compared to the Fordisc® reference sample. It is likely that a discriminant function analysis on the basis of more similar European reference samples will lead to more valid and reliable sexing results. The use of Fordisc® as a single method for the estimation of sex of recent skeletal remains in Europe cannot be recommended without additional morphological assessment and without a built-in software update based on modern European reference samples.

Dating skeletal remains with luminol-chemiluminescence. Validity, intra- and interobserver error.

UNLABELLED Amongst numerous other morphological, chemical, physical, and histological dating methods, the assessment of the chemiluminescence of weathered organic traces on bone finds by means of the luminol reaction is common practice. Opinion differs on the diagnostic value of this procedure for differentiating between historical and recent skeletonized human remains. DESIGN Chemiluminescence (CL) was tested in randomized, blind experiments on pulverized bone samples with known postmortem intervals from four different periods (1st-3rd century, 1878-1902, 1965-1972, 1995-2003). In addition to determining the sensitivity and specificity of the CL-analysis, likelihood ratios and predictive values were calculated. The degree of interobserver and intraobserver agreement was determined using Kappa statistics. RESULTS Based on our sample, we found false positive results in approximately 7.5% of all examined historical cases and false negative results in approximately 15% of the recent samples. Altogether, a correct classification of 88.75% (overall classification) was found (sensitivity=0.85, specificity=0.925; likelihood ratio=11.3; odds ratio=69.8). Taking a-priori probabilities into consideration (prevalence for recent=26%), a positive predictive value (PPV) of only 79.9% could be calculated, while the negative predictive value (NPV) lay at 94.6%. The results from subsequent Kappa statistics further illustrated that the quality of the results was, on the whole, observer independent (kappa=0.93; 95% CI=0.84-1.0). Intraobserver bias calculation yielded a value of kappa=0.975 (95% CI=0.92-1.0). CONCLUSION The analysis of CL on dry bones in determining the postmortem interval is not valid as an only method. In other respects, the luminol test is reproducible, repeatable and cost-effective and could help in assessing the postmortem interval (PMI). The results of this study document the potential danger of forensically relevant cases being erroneously sorted out if further examinations do not take place. We would like to emphasize that still only a combination of different methods including the analysis of environmental conditions and the interpretation of the anthropological profile can allow an acceptable estimate of the PMI from skeletonized remains, even if the methodological spectrum is being expanded by different absolute dating methods using radionuclide techniques.

Superimposition technique for skull identification with Afloat® software

The craniofacial superimposition technique is still an important way of identifying skulls when there is no reference sample for a forensic DNA analysis, when DNA typing from remaining tissue samples has failed, or when antemortem dental records are not available. Through the widespread use of digital photography, the probability of portrait photographs of the missing person being available for comparison is reasonably high. We present a superimposition technique that is inexpensive, since it uses a free software tool compatible with the soft-and hardware components already being used in many facilities. Into the bargain, this technique yields high-quality results.

The ideal subject distance for passport pictures

In an age of global combat against terrorism, the recognition and identification of people on document images is of increasing significance. Experiments and calculations have shown that the camera-to-subject distance - not the focal length of the lens - can have a significant effect on facial proportions. Modern passport pictures should be able to function as a reference image for automatic and manual picture comparisons. This requires a defined subject distance. It is completely unclear which subject distance, in the taking of passport photographs, is ideal for the recognition of the actual person. We show here that the camera-to-subject distance that is perceived as ideal is dependent on the face being photographed, even if the distance of 2m was most frequently preferred. So far the problem of the ideal camera-to-subject distance for faces has only been approached through technical calculations. We have, for the first time, answered this question experimentally with a double-blind experiment. Even if there is apparently no ideal camera-to-subject distance valid for every face, 2m can be proposed as ideal for the taking of passport pictures. The first step would actually be the determination of a camera-to-subject distance for the taking of passport pictures within the standards. From an anthropological point of view it would be interesting to find out which facial features allow the preference of a shorter camera-to-subject distance and which allow the preference of a longer camera-to-subject distance.

Geometric facial comparisons in speed-check photographs

In many cases, it is not possible to call the motorists to account for their considerable excess in speeding, because they deny being the driver on the speed-check photograph. An anthropological comparison of facial features using a photo-to-photo comparison can be very difficult depending on the quality of the photographs. One difficulty of that analysis method is that the comparison photographs of the presumed driver are taken with a different camera or camera lens and from a different angle than for the speed-check photo. To take a comparison photograph with exactly the same camera setup is almost impossible. Therefore, only an imprecise comparison of the individual facial features is possible. The geometry and position of each facial feature, for example the distances between the eyes or the positions of the ears, etc., cannot be taken into consideration. We applied a new method using 3D laser scanning, optical surface digitalization, and photogrammetric calculation of the speed-check photo, which enables a geometric comparison. Thus, the influence of the focal length and the distortion of the objective lens are eliminated and the precise position and the viewing direction of the speed-check camera are calculated. Even in cases of low-quality images or when the face of the driver is partly hidden, good results are delivered using this method. This new method, Geometric Comparison, is evaluated and validated in a prepared study which is described in this article.

Macroscopical Findings on Soil-Embedded Skeletal Remains Allowing the Exclusion of a Forensically Relevant Lay Time

So far, a reliable determination of the postmortem interval (PMI) of human bones or skeletal remains is possible neither by morphological examinations nor by extensive technical investigations. Concerning forensic osteological practice, in most cases, the question of the length of the PMI can be restricted to the exclusion of a forensically relevant lay time (in the present context corresponding to a PMI of >50 years). We reviewed 21 original publications from the literature and surveyed the results on the basis of our own experience with macroscopical findings of soil-embedded bones with known lay times. A total of 1259 bones and the respective findings were evaluated. Eleven macromorphological findings are presented that are of relevance when present and, in addition, five findings that are relevant through their absence for the determination of a PMI of more than 50 years. In none of the reviewed publications were these criteria described in soil-embedded bones with a lay time of less than 50 years. These elaborated findings are a basis for the macromorphological exclusion of a forensically relevant lay time of soil-embedded skeletal remains in a given case.