********** New Tool In Forensic Anthropology*********
Geometric morphometrics is the study of shape, shape variation and covariance with extrinsic factors. Basically, this is a statistical tool that can be used to reconstruct an object or a structure in a real or 3-D space.
This technique uses certain specific points called landmarks to reconstruct the geometrical shape of the specimen. Selected landmarks are superimposed on a reference model using statistical techniques like General Procrustes Analysis and the shape of specimen is analyzed after removing the size differences.
The technique is software oriented, all that has to be done is to put the centroid size values of chosen landmarks in the software and the most precise geometric shape will be the output.
Diagnose sexuelle probabiliste (DSP) is a quantitative technique that can be used for gender determination on the basis of certain measurements. It is based on the measurement of certain distances in the two coxal bones of the pelvic girdle.
Involves following steps:
1-Selection of landmarks and collection of landmark data: Landmarks should be such chosen that they can be easily located in all specimens under study and share a common loci.Second,it should be selected to properly capture shape to be observed.
Landmark data is nothing but the centroid size i.e. square root of the mean of squared inter landmark distances. Centroid size for any given specimen will change if different landmark configurations are used to define its geometric shape.
TYPES OF LANDMARKS
TYPE-1: Most fruitful set of landmarks.Example-Dacyron(the sutural point of intersection of maxillary, lacrimal and frontal bones).
TYPE-2: Curvature maxima associated with local structures such as bony processes for muscle attachments.Example – ectochonchion (most lateral point on orbital margin).
TYPE-3: Least meaningful landmarks.Example – euryon (either of the endpoints marking the greatest transverse diameter of skull.
CONCTRUCTED POINTS: Geometric combinations of other existing landmarks or lines erected at specific angles to construct a new landmarks.Example – mental foramen in mandible.
FUZZY LANDMARKS: Represents position of a biological structure that is precisely delineated but occupies an area that is larger than a single point in space. Example – frontal eminences.
Semilandmarks or sliding landmarks are the landmarks used for capturing the curvature, hence, enabling to study and determine morphology of curved surface.Once superimposed, they can be analyzed just as standard landmarks.Semi landmarks can only be located between two standard landmarks on a curved surface[.They are not placed on curved surface but on tangent to curve or surface.
2-Measurement of errors: The landmarks data will always be associated with some errors which cannot be completely eliminated but can be minimized by careful landmark selection, Euclidian distance matrix analysis.
3-Standarisation of data (Superimposition): Landmarks and Semilandmarks coordinates can be recorded on each specimen,but,size,orientation and position may vary from each of those specimens.Due to this new variables are added that distract from analysis of shape.
This is fixed by superimposition with Generalized Procrustes Analysis(GPA).
It removes variation of size,orientation and position by superimposing landmarks in a common coordinate system.Landmarks for all specimens are optimally translated,scaled and rotated based on least squared estimation.
Translation and rotation minimizes the squared and summed differences between landmarks of each specimen.Then landmarks are individually scaled to same centroid size.Now,landmark configuration for all specimens are brought into common coordinate system so that the only variable left is of shape.
4-Statistical analysis: A number of statistical techniques have been used for statistical analysis of data.These techniques basically reduce number of variables and represent best set of covariance between sets available, theyinclude:
1)Principal component analysis 2)Partial least square regression 3)Multivariate regression
Partial least squares(PLS): PLS is similar in aspects to Principal Component Analysis that it reduces the number of variables being observed so that patterns are easily observed in data, but uses a linear regression model.It looks at two or more sets of variables measured on same species and extracts the linear combination that best represents the two.It is largely used for sexual dimorphism.
Principal component analysis is used when large set of landmarks create correlated relationships that might be difficult to differentiate without reducing them in order to look at the overall variability in the data. In other words PCA preserves the shape variables that were scaled, rotated and translated during GPA.
DIAGNOSE SEXUELLE PROBABILISTE
Diagnose sexuelle probabiliste (DSP) is a method devised to estimate the sexual dimorphism in skeleton and also its magnitude. It was first developed by Murail et al.The method is based on the measurement of certain distances in the two coxal bones of the pelvic girdle.10 distances are measured and their magnitude shows considerable amount of variation in both the sexes.These distances are almost similar in the two coxal bones of a single specimen or there is a slight variation which is negligible, thus, if one distance cannot be measured in left coxal bone, its value in the right can be put in place.
These distances in the decreasing order of interest are:
1-PUM(acetabulo-symphyseal pubic length): Minimum distance from the superior and medial part of pubic symphysis to the nearest point on the acetabular rim at the level of lunate surface.
2-SPU(cotylo-pubic breadth): Pubic distance between the most lateral acetabular point and the medial aspect of the pubis.
3-DCOX(maximum pelvis height): Maximum height of the coxal bone measured from the inferior border to the most superior portion of the iliac crest.
4-IIMT(depth of greater sciatic notch): Distance from the postero-inferior iliac spine to the anterior border of greater sciatic notch.
5-ISMM (post-acetabular ischium length): Distance from the most anterior and inferior point of ischial tuberosity to the furthest point of the acetabular border.
6-SCOX(iliac breadth): Distance between the antero-superior iliac spine and postero-superior iliac spine.
7-SS (spino-sciatic length): Minimum distance between antero-inferior iliac spine and the deepest point in greater sciatic notch.
8-SA (spino-auricular length): Distance between the antero-inferior iliac spine and the auricular point (intersection of arcuate line with auricular surface).
9-SIS(cotylo-sciatic breadth): Distance between the lateral border of acetabulum and the midpoint of anterior portion of the greater sciatic notch.
10-VEAC(vertical acetabular diameter): Maximum vertical diameter of the acetabulum, measured on the acetabular rim, as the prolongation of longitudinal axis of the ischium.
Picture Showing various anthropometric measurements of coxal bones used in DSP in the decreasing order of their importance (PUM-VEAC).
At least 4 of the first 8 variables are required. When first 4 variables are not available, the last 2 may be added up.The measured values are put in the software and it returns the answer as Male, Female or Undefined.
Talking about the reconstruction of missing left condylar process of the mandible belonging to a famous Italian humanist, Pico della Mirandola using geometric morphometrics. Two methods of virtual reconstruction of the missing part were utilized:
1)reflection right hemimandible and replacement of the missing left condylar process extracted from reflection.
2)the molding of model A towards the preserved portion of its left counterpart.
The experimental results of reconstructing the missing left condylar process by means of reflecting the light was not successful.The aforesaid test demonstrated prediction of missing left condylar process with the help of warping the reflected complete hemimandible onto the remains of left hemi mandible provides the best results.The technique utilized here was geometric morphometrics.
Geometric morphometrics has enabled reconstruction of distorted or fragmentary hominid fossil for the purpose of their comparative analysis that would be impossible on incomplete structures.This technique is also important for forensic investigations for personnel identification with the help of craniofacial reconstruction and skull-photo superimposition.Creation of 3-D digital models with the help of geometric morphometrics creates the fundamental of the new field of virtual amthropology.
With the help this technique, reconstruction problems related to deformed and missing data could be solved increasing the reliability and reproducibility of results.
Limitations: Measurement errors may be introduced in different phases of geometrics morphometrics.These errors might be introduced during collection, preservation and storage of specimens.The commonly used fixation of fish in formalin followed by long-term storage in ethanol produces significant differences in shape and size.
It is also possible that the extent of measurement error induced due to preservation varies from species to species.Differences in the way specimen is positioned in front of data acquisition devices like CT scanner,Laser scanner introduce another source of error.For instance, it was found that the shape of dry human pelvic bones after reassembly was different relative to the shape of complete fresh pelvis.
Another limitation is regarding the choice of landmarks,the landmarks chosen should be able to capture the geometric shape of the specimen and should occupy homologous loci in different specimens of same species.
Diagnose sexuelle probabiliste on the other hand is a reliable technique of establishing sex using the two pelvic bones. This technique was developed in 2005 by Murail et al. This technique is entirely based on the measurements of ten anthropological distances in the coxal bones.Sliding and spreading calipers are used to measure the distances.These measurements are found to differ significantly in males and females rendering the hormonal changes in females as manifestations for preparations of parturition.
Limitations: The only limitation with DSP lies in the choice of variables which depends upon the part of coxal bone available for measurement.The first four variables PUM,SPU,DCOX and IIMT is the best combination variables and the last four variables SS,SA,SIS and VEAC offer the worst combination[.hus,the accuracy of the result is directly dependent on the variable one chooses for establishing the gender from a given specimen of coxal bones.
Main drawback of the DSP method is number of indeterminate bones depending upon available variables.This drawback is due to the choice of the authors to use a threshold value of 0.95 instead of the usual 0.50 for discriminate analysis.
This choice determines the high accuracy rate but also a percentage of indeterminate bones that depends upon the available variables and also the degree of sexual dimorphism in a given population.
We can easily say that Both the emerging and still improving techniques are a boon to forensic anthropology owing to their benefits and very low frequency of errors.
Research Papers to Read on the topic
- DSP: A tool for probabilistic sex diagnosis using worldwide variability in hip-bone measurements, 2005
- Sex determination using the Probabilistic Sex Diagnosis (DSP: Diagnose Sexuelle Probabiliste) tool in a virtual environment, Forensic Science International, Jan 2014
- Validation and reliability of the sex estimationof the human os coxae using freely available DSP2 softwarefor bioarchaeology and forensic anthropology, American Journal for Physical Anthropology, June 2017
- DSP: A probabilistic approach to sex estimation free from population specificity using innominate measurements, Chapter In book: Sex Estimation of the Human Skeleton, pp.243-269