| BackgroundIndividuals identify is an important content of forensic medicine research and is an important aim in practical forensic staff. In the practice of forensic identification depending on status of the samples and prenatal personal information, we can analyses the DNA of soft tissues or analysis the architecture of hard tissues. However, in large-scale catastrophic events、dismembered body cases、criminal cases of burned corpse、highly decomposed and so on, the soft tissues have already decomposed, thus it is hard to use fingerprints、iris、DNA for personal identification. While the hard tissues can be preserved relatively fine under the same conditions. So all the radiological information of human bones and teeth are the most valuable indexes for forensic identification.The maxillofacial region have more specifical skeletal landmarks, and were the best preserved part even influenced by kinds of lethal reasons and postmortem interferences. Thus it maybe the most potential and most valuable index for identification. With the development of medical image, the method of identification for forensic radiology is also developing. The methods are more and more scientific reasonable and practical. In1976and1991, Schwartz and Happonen carried out study on ordinary panoramic x-ray for forensic identification. They held that the main advantage of the panoramic x-ray is to show all the teeth, jaw and adjacent structures in one film. These studies on the panoramic x-ray made pioneering contributions to forensic identification. Digital panoramic tomography has many advantages, such as allegro checking high efficiency powerful post-processing. Meanwhile it is developing clearly and is easy to a store and transfer. In recent years, using digital panoramic tomography for identification became the research focus of the current forensic individual identification. Now the literature of study on digital panoramic tomography for identification are rare, only Gao Dong and Xu Baofu have done these reach. We can draw a conclusion that research on forensic identification for digital Orthopantomography has many issues worthy to explore.For many years, through the joint efforts of forensic dentistry scientists all over the world, we have reached a consensus of dental identification, according to Physiological and anatomical characteristics of the teeth and its adjacent structures, pathological changes and the teeth of various treatments such as individual identification to do identification work. However, with growing awareness of people’s oral health and advancing of dental treatment, dental problems are gradually reducing, age of onset is also delaying. So the method solely relying on dental disease and treatment characteristics can not adapt to the current situation. Meanwhile, odontologists all over the world are facing the same problem that dental identification methods are not uniform and the standard is also inconsistent.ObjectiveIn summary, by accurate measurement and appropriate processing on digital panoramic tomography of Chinese Han race population in Shenzhen, the main purposes of this study can be concluded as:1. To observe the coefficient of discrimination power (DP) of the data from each observed index.2. To census the distributing probabilities of the observed indexes of ShenZhen.3. To establish a database of original data of the measured values and distributing probabilities of the observed indexes for Chinese Han race population in Shenzhen.4. To provide a kind of theory of logic and methods to support for practical forensic identification.5. To lay the foundation for establishing personal recognition database of missing people and automated computer recognition system.Method1Selecting samplesSelect200samples preserved in database of department of stomatology in The Second People’s Hospital of Shenzhen. Requisition:(1) Age over18;(2) Chinese Han race in Shenzhen; Integrity of lower teeth (except for the Third Molar);(3) Dental and maxillofacial boneswithout obvious deformity;non fractures,paradentosis, systemic m etabolic diseases and endocrine diseases (4) Shooting under similar conditions, images are clear and sharp.2. Measuring the observing index200digital panoramic tomographies have been accurately measured by e-film Workstation2.12.Each mandibular teeth select6indexes to measure, including3length indexes and3angle indexes, and10bony indexes are selected to measure.3. Analyzing the results of measurementDo statistical description by SSPS16.0. And using Microsoft Office Excel2003to do sectional assignments. In order to exclude the influence of extreme values, definite (97.5%quantile-2.5%quantile)/5as sector. Respectively assigned1,2,3,4,5.And calculate the DP values.4. Comparing the influence of different staff Repeated measure and record of all200the cases by an oral postgraduate. Using t-test or rank sum test to examine the influence of different staff.5. Defining the threshold of matching probability According to chronological order, divided5person, s homologous dental orthopantomograms (10samples) into two groups. The Previous compose Group A; the latter ones with25non-homologous samples compose Group B. Accounting matching probability between Group A and Group B.Result1. Most observing indexes are in Gauss distribution.2. The frequency of each observed index in the group are different. The maximal one is96.5%, while in some sector the frequency is0.3. The DP of each observed index varies from0.58365to0.79125. The index ofΩ137、Ω143、Ω244、Ω145、Ω146each has a lower DP,0.06755,0.17195,0.49445,0.4761,0.1958.The total probability of discrimination power (TDP) of all the89observed index can reach to over0.999999.4.The statistical test results of measuring by different staff is P>0.01, which do not have statistically significance. We still cannot believe that different staff will impact the result significantly.5. The maximal matching probabilities of the section-numbers between two homologous dental orthopantomograms is77.53%, the minimal is69.66%. Whilethe maximal matching probabilities of the section-numbers between two non-homologous dental orthopantomograms is42.70%, much lower than69.66%. ConclusionThrough accurately measured and appropriate processing on digital panoramic tomography of Chinese Han race population in Shenzhen, we can draw conclusions:1. It is better to select lengthy indexes which are easy to measure and not to influence by other objective factors.2. The discrimination power (DP) of each observing index are different. The maximal one is0.79125.3. Established89indexes in digital panoramic tomography for forensic identification.4. The observing indexes of Ω137、Ω143、Ω144、Ω145、Ω146are not suitable for forensic identification.5. Measured by different staff wound not significantly affect the measurement results.6. Established convenient and effective method of forensic odontology individual identification on digital panoramic tomography of Chinese Han race population in Shenzhen. |
PDF Full Text Request