The Use of Digital Imaging in Human Identification and Crime Scene Analysis

[hermantheshocker]

2001 - Ventura County, California. The decomposed body of an elderly male was found floating in the ocean outside a harbor. No personal identification was in the clothing, and the body was transported to the Ventura County medical examiner's office for examination.

The condition of the facial tissues prevented visual identification as a reliable means of identity determination (Fig.1). Dental records (written and X-rays) were obtained by law enforcement and used to compare with dental findings obtained at autopsy. The few teeth remaining in the jaws were consistent with the dental records and the identification was determined by the medical examiner. The cause of death was drowning and the manner of death was accidental. No foul play was suspected.


Fig.1 This adult male was missing for 2 weeks and was eventually recovered on an ocean beach after drowning. The condition of the facial tissue obviously prevents visual identification as a reliable means of identity determination.

The autopsy disclosed advanced decomposition, with bloating and multiple areas of postmortem marine animal depredation. Postmortem loss of tissue from the right wrist revealed a stainless steel orthopedic fixation device on his radius. The dentition was severely carious(decayed), with many teeth missing and no evidence of dental restorations.

Postmortem decomposition of the tissues around teeth will eventually allow the single-rooted teeth (front teeth and some premolars) to loosen (Fig.2). This makes teeth susceptible to being lost at the crime scene or during transport. It is very helpful to bag the head of the deceased, as well as use a total body bag in this case since the circumstances indicate the chance of tooth loss.


Fig.2 This autopsy photograph shows the potential for tooth loss after death.

Fair-quality fingerprints were obtained, but no matching prints were found in fingerprint databases. None of the local law enforcement agencies had records of missing persons matching the general characteristics of the decedent during the prior month.

The medical examiner released the general information about the decedent to the local press, along with a description of his clothing and the orthopedic device. An adult daughter of the decedent called the medical examiner after reading the news release and provided a description of her father, who had been missing for two weeks.

The physical description was consistent, and the daughter recognized some of the clothing. She also recalled that her father had broken his wrist several years prior. Calls to several hospitals led to finding a radiograph of the decedent's right wrist (Fig.3) that was taken after placement of an orthopedic device for stabilizing a fracture in 1994, seven years prior to his death. Postmortem radiographs of the forearm produced images of the orthopedic device that were very similar to the antemortem films. This was believed sufficient to establish the identification. The original image objects analyzed are shown in Fig.4.


Fig.3 The concordance of shape and materials of the device is seen in both the postmortem (top, 2001) and antemortem (bottom, 1994) radiographs.

Digital imaging methods
Radiopaque labels were contained within each of the two radiographs for the purpose of orientation, identification, and size verification. Each of these reference labels consists of the letter “R” above a three-letter sequence.


Fig.4 Images consisted of the antemortem forearm radiograph (labeled 1994) and the postmortem forearm radiograph (labeled 2001).

The postmortem device was digitally colored black (Fig.5) and placed on top of the antemortem device (Fig.6). The initial superimposition revealed noticeable linear and angular differences between the antemortem and postmortem device images. This was caused by differing positions of the forearm on the film cassette between the two radiographic sessions (1994 and 2001). This is called Type II distortion. The general size and shape of the postmortem device shows a high degree of concordance with the antemortem device. There was good agreement when more specific features (individual screw threads) were compared.


Fig.5 The postmortem device is colored black to contrast with the antemortem x-ray.


Fig.6 The superimposition of the radiographs. Some differences were caused by differing positions of the forearm on the film cassette between the two radiographic sessions.

Photoshop can correct for the majority of photographic distortion, but there can still be minor angular differences remaining after the rectification process. This is especially true when comparing images with large amounts of angular discrepancies and/or three-dimensional curved surfaces. This case falls into both of these categories. In order to accurately compare the screw thread outlines, slight alignment adjustments must be made when different areas are analyzed. It is unrealistic to expect the entire image to exactly superimpose in this type of analysis due to subtle differences in the radiographic samples. Sectional analysis is advised in this circumstance.

One of the major advantages of digital image analysis is the ability to quantify concordant and/or dissimilar features. Although visual comparisons as described above can be extremely helpful, the addition of quantitative analyses can provide a more objective result. The task is to first find the area (in pixels) of the antemortem device image. The same is done for the rectified postmortem image. Following superimposition, the degree of commonality can be quantified by the percentage of pixels the two samples share. The table in Fig.7 shows the results of the comparison between the antemortem and postmortem orthopedic devices and quantifies the differing pixel values for each image (placing the blue image over the red image produces the color purple).


Fig.7 The antemortem device template (blue) is superimposed on the postmortem (red) device template and aligned. The opacity of the antemortem template layer is reduced to allow visualization of the postmortem layer below. The common pixels are then counted (purple).

The use of digital comparison methods helps the investigator control dimension variables during the comparison of physical objects. In this case, the identification was confirmed by other means, with this portion of the investigation being supportive of the final decision.

The usefulness of dental identifications is well documented in the professional forensic journals and the popular media. In death investigation, the cases of drowning or water immersion for prolonged lengths of time pose a problem in getting a timely identification. Bodies severely decomposed and swollen from water absorption will lack clothing holding identification papers. Tattoos and jewelry may or may not be present. DNA may be possible, but can take time to get results.

Missing person files must include the individual’s medical history and dental history. This information, if available, is a vital potential link between any recovered human remains and a possible identity. Investigators should be determined to follow leads towards any source of medical information.

-

This post is for educational purpose only and is nonprofit. Under Section 107 of the Copyright Act of 1976; Allowance is made for "Fair Use" for purposes such as criticism, comment, news reporting, teaching, scholarship, and research. No copyright infringement intended. All the images have been upscaled and sharpened/enhanced. I upgraded the original Photoshop UI in Fig.7 to 2021 version. Some of the text might have been shortened and simplified, and/or reorganized for online view. Original case report by Bowers C.M., Johansen R.J. - J Forensic Sci. 2002 / Forensic Dental Evidence 2010.

[William May]

VERY interesting!

[soyadude]

Oh god. He looked like a mutated molerat.

[ChristySD]

Thank you Herman!! Quality post, as usual. Keep em’ coming, please!!!!

[REVZZYBAN]

Two for hermanshocker. Lot's of time, effort, energy, info go into your posts. Recognized & acknowledged. Thanks,

[-JasonR-]

Very interesting. Thanks Herman.

[Bananamanize]

This is some quality stuff you’ve got here!

[ezeemonee]

That orthopedic device looks like someone installed it with garage tools. That had to be painful.

[rambbb]

science ftw

[Craphead]

Jesus Effin Christ, that is one gruesome corpse!