Who Are You?

Crime scenes are like puzzles and investigators often have to deal with pieces of the puzzles that aren’t fit to determine what had happened. Because solving a crime and convicting a perpetrator depend upon evidences, investigators take methodical approach in finding and handling evidences at crime scenes. They believe that even though covering a track isn’t easy but often the most practiced criminals leave behind traces that connect them to their crime scenes (Lyle 2004). These traces will be able to confirm the identity of the perpetrators and subsequently relate them to their presence at the crime scene and their motive of crimes as well.

Nevertheless, not every corpse that comes into the mortuary is conveniently carrying a driver’s license and identity card. Often police are confronted with identifying an unknown corpse. The time since death and many other factors may complicate the identification process, which usually involves many different forensic disciplines and techniques, all of which are coordinated by the medical examiner / pathologist (Lyle 2004). In reality, it may take weeks, months or even some corpses are not able to be identified which may complicate the investigator’s framework.

Hence, biometrics has appeared in forensic and it is the method of identifying or verifying the identity of an individual based on physiological and behavioural characteristics that have universality, distinctiveness, permanence, and extractability such as fingerprint, iris and face (Singh & Purkait 2009). A biometric system is essentially an identification system that work in two different modes i.e. verification and recognition. Verification refers to authenticating the claimed identity of a person while recognition refers to determining the identity of a person (Prabhakar & Jain 2002). Recognition system can differentiate various physiological or behavioural human features such as iris, face and fingerprint (Castonon et al. 2006) and will be further authenticated by verification with existing databases including DNA and fingerprint.

Identification Process

There are several identification processes in establishing the corpse’s identity including DNA analysis, fingerprints, dental records and medical records. The former two mentioned are also commonly applicable to relate perpetrators at the crime scenes.

Firstly, DNA sequence is unique to determine much of who you are and it’s a hot topic in forensic circles as it can pinpoint the perpetrator of a crime. DNA fingerprinting can relate the identity of every single individual including corpses and criminals except identical twins. With the development of so-called multilocus probes in 1985, the use of DNA analysis was introduced for questions of legitimacy e.g. immigration cases and for forensic applications such as rape or murder cases. Subsequently, single-locus probes has replaced multilocus probes and can be used in forensic work involving mixed samples because of excellent discimination capacity (high polymorphism) and providing high sensitivity for use even with small amounts of biological evidence left behind (Hochmeister 1995).

DNA fingerprints for criminal identification can be traced when the evidences left at the crime scene include blood specimen, hairs, skin cells or genetic materials in other form of body fluids (saliva, urine, spermatozoa). These identification details are not always found though can be pre-screened using presumtives tests at the crime scene, hence forensic investigators sometimes have to rely on the photography comprising of other biometrics, physical and trace evidences as well (Singh & Purkait 2009).    

Alternatively, mitochondrial DNA (mtDNA) typing is increasingly used in human identity testing when biological evidence is degraded, when quantity of the samples in question are limited or when nuclear typing is not the option of analysis. Biological sources of mtDNA include hairs, bones and teeth. In human body, mtDNA is inherited strictly from the mother and hence cannot discriminate between maternally related individuals but is beneficial for missing person cases. In sexual assault cases, Y-STR marker analysis are extremely valuable where samples contain multiple male contributors. These markers also eliminates the need to separate semen and vaginal epithelial cells prior to analysis because it eliminating female contribution from amplication profile (McClintock 2008).

Secondly, fingerprint identification is the field of forensic expertise relates to the inference of the identity through the examination of the friction skin ridges of the fingers or toes and their formed prints. The extreme variability of the fingerprints derives firstly from the knowledge of the morphogenesis of the papillary ridges pertaining to embryology and secondly from statistical researches pertaining to dactyloscopy. This variability is mainly used in identity verification and forensic investigation (Meuwly 2009). The use of fingerprints as a biometric is both the oldest mode of computer-aided, personal identification and the most prevalent in use today (Burge & Burger 2002). Fingerprint matching prior to automation involved the manual examination of the so-called Galton details which are the ridge endings, bifurcations, lakes, islands, and pores that collectively known as minutiae. Fingerprints are used to link suspects to crime scenes, identify deceased persons, and also to associate persons with questioned documents (Allen et al. 2005). There fingerprints can be lifted easily using dusting, ninhydrin, cyniacrylate super glue or direct photography. Fingerprints are unique for every single fingers and even identical twins possess different fingerprints (PDRM Forensic Laboratory 2009).

Thirdly, teeth can be as distinctive as fingerprints, provide significant corroborative evidence, and remain even after the rest of the body has disintegrated through environmental influence. The science dealing with establishing identity of a person by teeth for assistance in legal and criminal problems is popularly known as Forensic Odontology (Harini et al. 2010). Guidelines established by the American Board of Forensic Odontology (ABFO) cover identifying human remains from dental evidence, identifying accidental fatalities such as airline crashes, and bite analysis in criminal case. Comparison of ante-mortem (AM) and post-mortem (PM) dental records can be used to identify victims in mass fatalities even when DNA or fingerprints are unavailable while bite marks found on a body at a crime scene used to identify suspects (Asa 2009). Stages of dental growth can be used to estimate and narrow down the age of deceased. Moreover, dental biometrics utilizes dental radiographs for human identification with information about the condition of teeth, their roots and jaw placement (Goudelis et al. 2008).

Last but not the least, Chen and Jain (2005) proposed an automatic method for matching dental radiographs that has two main stages: feature extraction and matching.

An area-based metric is used for matching and compute the similarity between the AM and PM images to retrieve the identities from a database. These concept of comparison between PM and AM records could also be applicable in any other medical records of the deceased or unknown bodies. In fact, medical records include X-ray or any radiography records, skeletal evidence of diseases (bone cancer, tuberculosis or rickets), surgical appliances such as artificial hips and pacemaker, even other injuries, scars or abnormalities of body parts (Lyle 2004).

Advantages of the Techniques

Out of the four techniques discussed in the previous section, medical and dental records only enable the investigators to narrow down the suspects and identity of the corpse but it would not be 100% pinpoint to the exact individual’s identity based on the biological and physicochemical evidences at scene or on dead body.

Nevertheless, National Academy of Sciences (NAS) US commission report stated that “Apart from DNA, there is not a single forensic discipline that has been proven with a high degree of certainty to be able to match a piece of evidence to a suspect” after examining the state of forensic in US law enforcement. FBI are using the Combined DNA Index System (CODIS) with the combination of Profiler plus and Cofiler plus systems focusing on total of 13+1 STR loci (Refer Illustration 1).

Ilustration 1

These 13 STR markers provide a random match probability of 1 in 100 million. CODIS contained over 1.5 million STR profiles and linked all 50 states in the US with capability to search criminal DNA profiles in similar fashion as the FBI fingerprint database (Butler 2005).

Based on the lastest development in Malaysia, Department of Chemistry currently use the Identifiler Allelic Ladders System focusing on 15+1 STR loci for DNA typing in medicolegal cases by matching with the existing DNA databank (Refer Illustration 2). The compilation of population databases are based on guidelines by US National Research Council (NRC) in which also adopts recommendations of Scientific Working Groups on DNA Analysis Methods (SWGDAM). These have been useful in the application of forensic DNA profiling including kinship determination and serious crime cases like rape, murder, kidnap and robbery. Statistical evaluation showed that the probability of coincidental match of DNA profiles among Malaysian is ranging from 1 in billion to sextillion from case-to-case basis and the accuracy is close to 99.9% (Lim 2009).

Ilustration 2

Collaboration between FBI and National Institute Standards and Technologies (NIST) has formed Automated Fingerprint Identification Systems (AFIS), a computerized databases of fingerprints by latent fingerprint experts in US which is better method for storing, retrieving and matching fingerprints by scanning thousands of unknown ten-prints within seconds (Dror & Mnookin 2010). Nowadays, MAFIS also frequently used in Malaysia for the fingerprint matching and the databases are enforced under Section 10(1)(C) I & II of the Criminal Registration Act No.7/1969 (PDRM Forensic Laboratory 2009). The probability of two person bearing the same fingerprint is 1 in 64 billion throughout world population with closely 99.9% accuracy too (Ninthin et al. 2009). The uniqueness of fingerprint makes its matching are 100% undisputable (Leo 2005).

References