Using micro-CT imaging for triaging evidence in a vehicle fire investigation. A case report

Abstract

This case report highlights how micro-CT imaging was used to assist in the investigation of a homicide that involved a burnt-out car, with several burnt evidence samples submitted for examination. It is detailed how micro-CT was used for a police investigation to ensure a more effective workflow for triaging evidence. High resolution 2D micro-CT imaging provided quick characterisation of the contents of samples, eliminating the need for full 3D scanning of most samples. This eliminated time-consuming manual excavation of samples, preventing exposure to toxic chemicals. Furthermore, having the 2D and 3D micro-CT images allowed for a digital archive of evidence.

Introduction

Fire scene investigations entail an array of environments ranging from heat exposure to fire or explosions or a combination thereof, as well as different settings such as buildings, vehicles, or open spaces. Often other considerations are present at the scene, such as electrical, chemical or toxicity hazards, as well as the need for recovery of evidence or even human remains (Johnson et al., 2021). Fires are therefore particularly complex scenes to investigate. While often the criminal consideration of fire incident is to understand the origin, the cause and possible intentionality involved, fire investigations may equally begin with an understanding of the intent, such as concealment of a crime. In these circumstances, the origin and cause of fire may be less important, but retrieval of evidence for understanding circumstances prior to the fire may be more relevant and the immediate concern. Evidence may be retrieved from the scene for further destructive excavation, testing and analysis where X-rays may be acquired using portable equipment to understand the nature and contents of burnt evidence, to assess the need for destructive analysis, as well as to plan the retrieval of evidence within the burnt evidence matter (Nejtková, 2019; RSE, 2023; Svare and Daeid, 2023; Utt, 2010).

After a review of available literature, there is limited information on how micro-CT has been used to assist in the investigation of fire investigations. Micro Computed Tomography (micro-CT) has been predominantly used for industrial applications as a non-destructive means of inspection of internal parts and characterisation of different materials, as well as metrology for external and internal geometry (De Chiffre et al., 2014). However, high-resolution micro-CT imaging is a technique that is increasingly being used within police investigations, storage of evidence in a digital format, and for court presentation (Alsop et al., 2022; Baier et al., 2018, 2021; Franchetti et al., 2022; Hainsworth, 2022; Rutty et al., 2013; Thali et al., 2003). Specifically, within forensic case work, micro-CT has been used for skeletal analysis for paediatric case work of potential child abuse (Arthurs et al., 2017; Baier et al., 2019, 2021a; Primeau et al., 2024) blunt force trauma (Baier et al., 2018, 2021b; Brown et al., 2011; Collings and Brown, 2020) and tool mark analysis (Alsop et al., 2021a, 2022; Appleby et al., 2015; Baier et al., 2017; Giraudo et al., 2020; Komo and Grassberger, 2018; Norman et al., 2018a, 2018b; Palletti et al., 2017; Pounder and Sim, 2011; Stanley et al., 2018; Thali et al., 2003). The technique has also been used to scan 3D printed weapons (Goia et al., 2024), analysis of ballistic cartridges (Alsop et al., 2021b), gunshot residue analysis (Brożek-Mucha et al., 2020; Cecchetto et al., 2011, 2012), gunshot wounds (Cecchetto et al., 2011; Fais et al., 2015), blood pattern analysis (Dicken et al., 2015, 2019), and entomology for post-mortem interval (Richards et al., 2012). Micro-CT allows for 3D visualisation providing an enhanced approach for analysis of contents, comparing, reconstructing, materialising, and sharing data. Micro-CT has previously been used in police investigations specifically with material of a burnt nature (Baier et al., 2017; Boschin et al., 2015; Ellingham and Sandholzer, 2019; McKinnon et al., 2021). Especially for complex cases, the 3D volume rendered images or 3D printed objects from micro-CT scanning have been shown to be particularly helpful for delivery of expert witness testimony during court presentations (Baier et al. 2021a, 2021b; Errickson et al., 2014). The principles of micro-CT scanning are equivalent to medical CT scanning, using X-ray emission for cross-sectional projections taken during a 180° or 360° rotation at multiple angles about an axis through an object. This allows to provide a 3D reconstruction using a computer algorithm, but also allows 2D X-rays to be acquired (Franchetti et al., 2022; Hainsworth, 2022). Micro-CT can, as with medical CT equipment, provide quick X-rays, but at a much higher resolution and with higher penetration capabilities compared to medical equipment, but full 3D scanning time can be extensive. Hence, the advantage of micro-CT is the high resolution achievable, down to micro- or nanometres, as well as higher penetration for dense samples, while the disadvantages are the limitations of field of view and sample size, along with the availability of such specialised equipment. However, using industrial micro-CT equipment, large samples can be scanned, or can be scanned in height or width sections and combined in post-processing of images. Further, this type of equipment is becoming more available in academic and industrial institutions.

This paper highlights how high-resolution micro-CT imaging was used to support the investigation of a homicide that involved a burnt-out car, through which multiple burnt evidence samples were recovered. The utility of micro-CT is described here, incorporated into a police investigation as a means for triaging evidence, thus focusing the investigation by clarifying outstanding weaponry and evidence that did not require further investigation, significantly improving the efficiency of time and financial resources.

Case background

At approximately 0130hrs on 20/6/21, a stolen VW Golf bearing a false registration, pulled up on the offside of a private hire taxi. Once in position, a shotgun was discharged from the nearside passenger window of the Golf through the rear offside boot window of the taxi. The passenger in the rear offside seat was fatally shot in the head and the other two passengers subsequently exited the taxi and fled the scene. The VW Golf then turned around and drove away. The gunshot victim was evacuated from the scene by ambulance and subsequently died in hospital from his wounds.

At approximately 1100hrs on 20/6/21, a burnt-out VW Golf was reported to the police in Wolverhampton (Figure 1). The VW Golf was believed to be the offending vehicle involved in the murder investigation. The vehicle was subsequently tracked from where it had been parked and entered by the suspects, to where the homicide event took place, and lastly to the location where it was found in a burnt state. This was done using CCTV footage, using unique identifying features of the car; registration number, colour, and identifiers to specific make and model. The vehicle had significant fire damage and was examined by Forensic Scene Investigators together with a Fire Investigator from West Midlands Fire Service. To assist the investigation, West Midlands Police requested Micro-CT examination of burnt material recovered from the scene to ascertain whether any significant items were present within them, which may assist the investigation. This included items such as shotgun cartridges, other ammunition, firearms, mobile phones, sim cards, lighters, and gloves.

Figure 1.

Crime scene photographs of the burnt-out VW Golf, where the samples were taken from.

Methods/Materials

Materials

14 samples of burnt material recovered from the VW Golf, were submitted in sealed plastic containers for micro-CT scanning to the Forensic Centre for Digital Scanning and 3D Printing at the Warwick Manufacturing Group (WMG), University of Warwick. These included samples taken from the driver and passenger doors and footwells, and from between the door and seat.

Micro-CT

The samples were scanned in their containers using either a Unitom XL1 (TESCAN, Brno, Czech Republic) or XT H 450RT (Nikon, Leuven, Belgium) scanner, dependent on sample density. All samples were initially radiographed, with two radiographs taken 90° apart and triaged for full micro-CT scanning. Each radiograph took approximately 5 minutes to produce including set up time.

Following analysis of the radiographs with the investigative team, a full CT scan was conducted on two samples; one recovered from the driver’s door using the Nikon XT H 450RT, and one from the driver’s footwell using the TESCAN Unitom XL1. A calibration workpiece was included in the two full micro-CT scans to ensure the measurement error was equal to the spatial resolution of the scan. For 3D CT scans, multiple radiographs are taken as the sample rotates 360° and these radiographs are then reconstructed to form a three-dimensional volume with the proprietary software. Each CT scan took approximately 36 and 80 minutes, respectively, to acquire, and an additional 1 hour for set up and reconstruction time. A diagram of the workflow is presented in Figure 2.

Figure 2.

Flow diagram showing the workflow for sample scanning and analysis.

The 3D micro-CT data was then exported to VGStudio MAX 2.2 (VolumeGraphics, Heidelberg, Germany), for visualisation, measurement, and analysis. Following the scanning process, the samples were returned into safe storage until collection by West Midlands Police. As all processes and procedures used to micro-CT scan the samples and produce the digital 3D models are non-invasive, the evidence was not physically handled at any point from receiving the samples, scanning them, to returning them to Police custody.

Results

Samples submitted for micro-CT were successfully captured through the non-invasive micro-CT method as outlined above, with each sample being captured with two high resolution radiographs taken with two rotations, 90° apart. Results from the micro-CT scans are summarised in Table 1 and illustrated in Figures 3–7.

Table 1.

Results from the micro-CT scanning.

Sample	Location	Description	Figure	Image resolution
1	Drivers’ door	Circuitry, cabling, metal clips and springs were apparent, as well as a large dense object. A full CT scan was required to determine the nature of the large denser object. Follow up 3D CT scan revealed that the unknown component was likely a window motor	Figures 3A and 4A	75 μm (120 μm for CT)
2	Driver’s footwell	A small piece of circuitry along with wires, clips and likely burnt plastic were apparent	Figure 3B	80 μm
3	Driver’s footwell	A screw and metal clips were present along with small density particles	Figure 3C	90 μm
4	Driver’s footwell	Screws, metal clips and likely broken glass were present along with small dense particles. A follow up 3D CT scan revealed high density spheres with diameters ranging from 0.25 to 1.53 mm for the 13 measured	Figures 3D and 4A	65 μm (80 μm for CT)
5	Driver’s footwell	A single metal clip and a small number of dense particles were identified	Figure 5A	40 μm
6	Driver’s footwell	A screw and metal clip were identified alongside further dense debris	Figure 5B	40 μm
7	Driver’s footwell	Small dense particles identified	Figure 5C	35 μm
8	Driver’s footwell	Lightbulb and fitting were identified	Figure 5D	70 μm
9	Driver’s footwell	Screws, metal clips, springs and some dense particles were identified	Figure 6A	90 μm
10	Near offside door	A single large dense component was identified	Figure 6B	50 μm
11	Passenger’s footwell	Scale bar and radiograph resolution shown at the bottom. Clips, wiring and likely broken glass was identified	Figure 6C	50 μm
12	Passenger’s footwell	Broken glass identified	Figure 6D	50 μm
13	Gap between the front passenger door and seat	Metal clips identified	Figure 7A	60 μm
14	Gap between the front passenger door and seat	A single metal clip and unknown object identified	Figure 7B	40 μm

Figure 3.

(A) Sample 1, contained a larger object (arrow) that required a full micro-CT scan, see Figure 4(A). (B) Sample 2. (C) Sample 3. (D) Sample 4, contained small dense particles (arrows) that required a full micro-CT scan, see Figure 4(B)).

Figure 4.

(A) Micro-CT scan of sample 1 with low density material digitally filtered (pre-filtered image in top left). Zoomed in regions of the circuitry (top right), an unknown component (middle right) and the dense object (bottom). 2D cross-sections of what was identified as a window motor are shown at the bottom of the image (see inset images for cross-section positions). (B) Micro-CT scan of sample 4. Low density material has been digitally filtered. 13 high density spheres were identified with diameters ranging from 0.25 to 1.53 mm.

Figure 5.

(A) Sample 5. (B) Sample 6. (C) Sample 7. (D) Sample 8.

Figure 6.

(A) Sample 9. (B) Sample 10. (C) Sample 11. (D) Sample 12.

Figure 7.

(A) Sample 13 from the gap between the front passenger door and seat. (B) Sample 14 from the gap between the front passenger door and seat.

Sample 1 (Figure 3(A)), showed to have remnants of circuitry, cabling, metal clips and springs, A full micro-CT scan was required to identify the large dense object. From the full micro-CT scan (Figure 4(A)), the object was identified as a window motor. Sample 2 (Figure 3(B)), contained a small piece of circuitry along with wires, clips and burnt plastic. Sample 3 (Figure 3(C)), contained a screw and metal clips along with small dense particles. Sample 4 (Figure 3(D)), contained screws, metal clips and likely broken glass along with small dense particles. A follow up CT scan was performed to further examine the small dense particles. The full micro-CT scan (Figure 4(B)) revealed the small high-density spheres had different diameters ranging from 0.25 to 1.53 mm for the 13 spheres measured.

Sample 5 (Figure 5(A)), contained a single metal clip and a small number of dense particles. Sample 6 (Figure 5(B)), contained a screw and metal clips alongside further dense debris. Sample 7 (Figure 5(C)), contained small dense particles. Sample 8 (Figure 5(D)), contained a lightbulb and fitting.

Sample 9 (Figure 6(A)) contained screws, metal clips, springs and some dense particles. Sample 10 (Figure 6(B)) contained a large dense component. Sample 11 (Figure 6(C)) contained clips, wiring and likely broken glass. Sample 12 (Figure 6(D)) contained broken glass.

Sample 13 (Figure 7(A)) contained metal clips. B) Sample 14 (Figure 7(B)) contained a single metal clip and an unknown object (potentially broken glass or plastic).

The resulting micro-CT scans illustrated the content of the submitted samples in greyscale depending on density of the material. This allowed the contents to be separated and viewed within the containers. Several samples proved to contain metal components of no immediate relevance to the investigations. Two samples (Sample 1 and Sample 4) that had further micro-CT examination conducted was able to show what the samples contained. Sample 1 contained a circuit board, an unknown component that was later considered to be the button for the central locking system and parts of the window motor (Figure 4(A)). Sample 4 contained several smaller high-density spheres, of which 13 were measured (numbered on Figure 4(B)). The 13 measured spheres had a mean diameter of 0.6 mm with a range of 0.255-1.53 mm. Initially these spheres were considered shotgun pellets; however, due to their varying sizes and measured diameters, they were subsequently disregarded and identified as steel ball bearings.

Discussion

Within modern policing there is a constant demand on critical resources that are required to effectively carry out investigations. This situation is made more challenging with the emergence of new techniques and methods that require training, research knowledge and access to specialist equipment. Collaborating with external institutions can help overcome this problem by providing access to essential infrastructure that can be used to support police investigations.

One such centre is the Forensic Centre for Digital scanning and 3D printing at the Warwick Manufacturing Group (WMG), University of Warwick. This facility works in partnership with West Midlands Police who act as the official sponsor of the streamlined service delivery of micro-CT scanning for national and international police forces. The centre has been involved in more than 360 police investigations and delivered micro-CT scanning for 35 police forces. These cases span from child abuse, vulnerable adult abuse, sharp and blunt force trauma, dismemberment cases, strangulations, weapons and 3D printed weapons, and possible tampering of medical equipment. A wide range of micro-CT scanners are available comprising five different systems, each with their own specification and capabilities, allowing almost any type of material to be scanned and a wide range of sample sizes with a quick turnaround time.

This capability has been used to support the homicide investigation as detailed within this paper, allowing a fast triage of a significant number of evidence samples collected from a burnt-out car. This provided better resource management and increased responsiveness in a complex investigation. The process of implementing non-invasive high-resolution micro-CT X-rays and scanning into the investigation ensured a better use of both time and financial resources, as well as providing a permanent record of evidence in a digital format. Thereby limiting the need for extensive evidence storage, as well as the potential for using the imagery for court presentations.

A major concern for this investigation was the recovery and identification of the weapon used for the homicide or possible additional weapons carried by the assailants. Hence there was a need for an immediate clarification of where the investigation should focus. Micro-CT scanning found no further evidence of ammunition or weapons within the evidence samples. This allowed the investigation to focus on the recovery of the missing weapon, in addition to the wider homicide investigation. Further, it allowed awareness of the outstanding weapon posing a continuing risk to both the investigators when searching premises as well as for further usage for illegal purposes. The car was very quickly traced using CCTV from the location of the homicide to a location where it was deposited, prior to being moved to the location where it was found in a burnt condition. Additionally, had evidence been identified from micro-CT imaging, of clothing (zippers, buttons) or mobile phones, this could have indicated that the car had been burnt for the purpose of destroying evidence such as DNA or fingerprints, rather than a potentially incidental fire.

Had micro-CT not been available as an investigative tool, portable X-ray equipment could have been used, although this would unlikely have had the penetration needed for denser objects for optimal image quality and would not have provided high-resolution images. Further, X-ray equipment producing only 2D images would not have the ability to provide 3D reconstruction or the ability to filter non-relevant material or provide measurements. Alternatively, the burnt samples could have been sent for an internal or external manual excavation and subsequent cataloguing and analysis, which would have been laborious and time consuming, likely delaying the investigation by weeks. In addition, there would have been possible health and safety issues concerning toxic exposure when manually handling the samples. Therefore, combining the fire investigation with the usage of micro-CT scanning, allowed for a quick triage of samples in high detail, enabling the investigation to disregard the burnt samples as important to the investigation, thereby saving time and financial resources. Further, having the micro-CT images allowed digital storage of the evidence, thereby enabling the burnt samples to be destroyed confidently knowing they were not relevant, eliminating the necessity for the physical storage of evidence. Lastly, having the micro-CT imagery would have allowed these to have been used for court presentation, if any relevance had been found.

There are some limitations to the usage of micro-CT images. Had all samples required a full 3D scan, this would have lengthened the imaging time as the 3D scan time for the two samples required between 1 and 2 hours. The subsequent reconstructing of the images required additional time, as well as time for analysis of the 3D volumetric data. Hence, providing only high resolution 2D radiographs of most of the samples was quicker, and required minimal data processing. Further, there are issues with the size of data produced from micro-CT images. As an example, the 3D reconstructed data for one sample to be imported to the visualisation software, had a data size of 27 GB. Lastly, specialised software and therefore user knowledge is required for the visualisation of the 3D data, therefore making sharing and viewing data outside of the scanning centre difficult.

Conclusion

This paper has presented the novel usage of micro-CT imaging for a police investigation of a homicide involving a burnt-out car, illustrating some of the possibilities that the imaging modality of micro-CT scanning can provide for a police enquiry. This paper has highlighted the advantages of providing a fast triaging of burnt samples in high-resolution visualisation of smaller objects, as well as having the penetration power for high-resolution visualising of larger denser objects. Using micro-CT imaging allowed the investigation to quickly advance the focus of the further investigation for the outstanding weapon. Additionally, imaging the samples, avoided the need for manually excavating the samples with potential exposure to toxic chemicals. Lastly, the high-resolution images provided digital storage of evidence negating the need for physical storage and could have been used for court presentation had any relevant material been present. Overall, micro-CT can be used for better resource management and time administration for complex police investigations.