Post-mortem X-ray computed tomography (PMCT) identification using ante-mortem CT-scan of the sphenoid sinus

Evaluating artificial intelligence for comparative radiography

INTRODUCTION

Comparative radiography is a forensic identification and shortlisting technique based on the comparison of skeletal structures in ante-mortem and post-mortem images. The images (e.g., 2D radiographs or 3D computed tomographies) are manually superimposed and visually compared by a forensic practitioner. It requires a significant amount of time per comparison, limiting its utility in large comparison scenarios.

METHODS

We propose and validate a novel framework for automating the shortlisting of candidates using artificial intelligence. It is composed of (1) a segmentation method to delimit skeletal structures' silhouettes in radiographs, (2) a superposition method to generate the best simulated "radiographs" from 3D images according to the segmented radiographs, and (3) a decision-making method for shortlisting all candidates ranked according to a similarity metric.

MATERIAL

The dataset is composed of 180 computed tomographies and 180 radiographs where the frontal sinuses are visible. Frontal sinuses are the skeletal structure analyzed due to their high individualization capability.

RESULTS

Firstly, we validate two deep learning-based techniques for segmenting the frontal sinuses in radiographs, obtaining high-quality results. Secondly, we study the framework's shortlisting capability using both automatic segmentations and superimpositions. The obtained superimpositions, based only on the superimposition metric, allowed us to filter out 40% of the possible candidates in a completely automatic manner. Thirdly, we perform a reliability study by comparing 180 radiographs against 180 computed tomographies using manual segmentations. The results allowed us to filter out 73% of the possible candidates. Furthermore, the results are robust to inter- and intra-expert-related errors.

Forensic Identification from Three-Dimensional Sphenoid Sinus Images Using the Iterative Closest Point Algorithm

Forensic identification of human remains is crucial for legal, humanitarian, and civil reasons. Wide heterogeneity in sphenoid sinus morphology can be used for personal identification. This study aimed to propose a new protocol for personal identification based on three-dimensional (3D) reconstruction of sphenoid sinus CT images using Iterative Closest Point (ICP) algorithm. Seven hundred thirty-two patients which consisted of 348 females and 384 males were retrospectively included. The study sample includes 732 previous images as a source point set and 743 later ones as a scene target set. The sphenoid sinus computed tomography (CT) images were processed on a workstation (Dolphin imaging) to obtain 3D images and stored as a file format of Stereo lithography (.STL). Then, a Python library vtkplotter was used to transform the STL format to PLY format, which was adapted to Point Cloud Library (PCL). The ICP algorithm was used for point clouds matching. The metric Rank-N recognition rate was used for evaluation. The scene target set of 743 individuals was compared with the source point set of 732 individual models and achieved Rank-1 accuracy of 96.24%, Rank-2 accuracy of 99.73%, and Rank-3 accuracy of 100%. Our results indicated that the 3D point cloud registration of sphenoid sinuses was useful for assessing personal identification in forensic contexts.

Human identification performed with skull's sphenoid sinus based on deep learning

Human identification plays a significant role in the investigations of disasters and criminal cases. Human identification could be achieved quickly and efficiently via 3D sphenoid sinus models by customized convolutional neural networks. In this retrospective study, a deep learning neural network was proposed to achieve human identification of 1475 noncontrast thin-slice CT scans. A total of 732 patients were retrieved and studied (82% for model training and 18% for testing). By establishing an individual recognition framework, the anonymous sphenoid sinus model was matched and cross-tested, and the performance of the framework also was evaluated on the test set using the recognition rate, ROC curve and identification speed. Finally, manual matching was performed based on the framework results in the test set. Out of a total of 732 subjects (mean age 46.45 years ± 14.92 (SD); 349 women), 600 subjects were trained, and 132 subjects were tested. The present automatic human identification has achieved Rank 1 and Rank 5 accuracy values of 93.94% and 99.24%, respectively, in the test set. In addition, all the identifications were completed within 55 s, which manifested the inference speed of the test set. We used the comparison results of the MVSS-Net to exclude sphenoid sinus models with low similarity and carried out traditional visual comparisons of the CT anatomical aspects of the sphenoid sinus of 132 individuals with an accuracy of 100%. The customized deep learning framework achieves reliable and fast human identification based on a 3D sphenoid sinus and can assist forensic radiologists in human identification accuracy.

Pragmatic approaches to reducing radiation dose in brain computed tomography scan using scan parameter modification

Background:

High radiation dose of patients has become a concern in the computed tomography (CT) examinations. The aim of this study is to guide the radiology technician in modifying or optimizing the underlying parameters of the CT scan to reduce the patient radiation dose and produce an acceptable image quality for diagnosis.

Methods:

The body mass measurement device phantom was repeatedly scanned by changing the scan parameters. To analyze the image quality, software-based and observer-based evaluations were employed. To study the effect of scan parameters such as slice thickness and reconstruction filter on image quality and radiation dose, the structural equation modeling was used.

Results:

By changing the reconstruction filter from standard to soft and slice thickness from 2.5 mm to 5 mm, low-contrast resolution did not change significantly. In addition, by increasing the slice thickness and changing the reconstruction filter, the spatial resolution at different radiation conditions did not significantly differ from the standard irradiation conditions (P > 0.05).

Conclusion:

In this study, it was shown that in the brain CT scan imaging, the radiation dose was reduced by 30%–50% by increasing the slice thickness or changing the reconstruction filter. It is necessary to adjust the CT scan protocols according to clinical requirements or the special conditions of some patients while maintaining acceptable image quality.

Computer-aided superimposition of the frontal sinus via 3D reconstruction for comparative forensic identification

The anatomical uniqueness of the frontal sinus morphology has been widely used for comparative forensic identification using various techniques, mostly including 2D X-rays or one fixed slice of an axial computed tomography (CT) scan image. However, computer-aided 3D automatic graphical comparison techniques can provide accurate comparisons between two 3D models that allow users to comply with even the strictest deviation standards, avoiding error-prone identification of frontal sinuses with similar morphologies. The study proposes the use of a computer-aided comparative paradigm based on the 3D-3D frontal sinus model superimposition process and further assesses the anatomical uniqueness of frontal sinuses using a large Chinese Han sample. Three hundred thirty-six patients older than 20 years with two multi-slice CT scans were collected. Frontal sinus 3D models were semi-automatically segmented through Dolphin Imaging software. Automatic pairwise comparisons of 336 matched pairs from the same person and 340 mismatched pairs from different individuals with an analysis of average root mean square (RMS) point-to-point distance were performed using Geomagic Studio Qualify software. RMS ranged between 0.005 and 1.032 (mean RMS 0.390 ± 0.25 mm) in the group of matches and between 1.107 and 19.363 (mean RMS 4.49 ± 2.69 mm) in the group of mismatches. On average, the RMS value was over ten-fold greater in mismatches than in matches. Statistically significant differences in RMS between the group of matches and mismatches were assessed using the Mann-Whitney U test (p < 0.05). This study supports the value of the frontal sinus with a 3D computer-aided superimposition method for human identification with large samples when DNA, fingerprints, and dental materials are not accessible.

Automatic forensic identification using 3D sphenoid sinus segmentation and deep characterization

Recent clinical research studies in forensic identification have highlighted the interest in sphenoid sinus anatomical characterization. Their pneumatization, well known as extremely variable in degrees and directions, could contribute to the radiologic identification, especially if dental records, fingerPrints, or DNA samples are not available. In this paper, we present a new approach for automatic person identification based on sphenoid sinus features extracted from computed tomography (CT) images of the skull. First, we present a new approach for fully automatic 3D reconstruction of the sphenoid hemisinuses which combines the fuzzy c-means method and mathematical morphology operations to detect and segment the object of interest. Second, deep shape features are extracted from both hemisinuses using a dilated residual version of a stacked convolutional auto-encoder. The obtained binary segmentation masks are thus hierarchically mapped into a compact and low-dimensional space preserving their semantic similarity. We finally employ the ℓ₂ distance to recognize the sphenoid sinus and therefore identify the person. This novel sphenoid sinus recognition method obtained 100% of identification accuracy when applied on a dataset composed of 85 CT scans stemming from 72 individuals. Automatic Forensic Identification using 3D Sphenoid Sinus Segmentation and Deep Characterization from Dilated Residual Auto-Encoders.

Automated contour detection in spine radiographs and computed tomography reconstructions for forensic comparative identification

This study was conducted to test an automated method to identify unknown individuals. It relies on a previous radiographic file and uses an edge-based comparison of lumbar CT/PMCT reconstructions and radiographs. The living group was composed of 15 clinical lumbar spine CT scans and 15 paired radiographs belonging to the same patients. The deceased group consisted of 5 lumbar spine PMCT scans and 5 paired antemortem radiographs of deceased individuals plus the 15 unpaired radiographs belonging to the living. An automated method using image filtering (anisotropic diffusion) and edge detection (Canny filter) provided image contours. Cross comparisons of all the exams in each group were performed using similarity measurements under the affine registration hypothesis. The Dice coefficient and Hausdorff distance values were significantly linked (p < 0.001 and p = 0.001 respectively) to the matched examinations in the living group (p < 0.001; pseudo-R2 = 0.70). 12 of the 15 examinations were correctly paired, 2 were wrongly paired and 3 were not paired when they must have been. In the deceased group, the Hausdorff distance was significantly linked (p = 0.018) to the matched examinations (p < 0.001; pseudo-R2 = 0.62; Dice coefficient p = 0.138). The paired examinations were all correctly found, but one was wrongly paired. The negative predictive value was above 98% for both groups. We highlighted the feasibility of comparative radiological identification using automated edge detection in cross-modality (CT/PMCT scan and radiographs) examinations. This method could be of significant help to a radiologist or coroner in identifying unknown cadavers.

Three-dimensional analysis of sphenoid sinus uniqueness for assessing personal identification: a novel method based on 3D-3D superimposition

Sphenoid sinuses are considered the most variable structures of human body: therefore, they may be used for personal identification, through the application of 3D segmentation procedures. This study aims at proposing a new protocol for personal identification based on 3D-3D superimposition of sphenoid sinuses segmented from head CT scans. Adult subjects (equally divided among males and females) who underwent two head CT scans were extracted from a hospital database. Sphenoid sinuses were segmented through ITK-SNAP software and the corresponding 3D models were automatically superimposed to obtain 40 matches (when they belonged to the same person) and 260 mismatches (when they were extracted from different individuals). The RMS (root mean square) point-to-point distance was then calculated for all the superimpositions: differences according to sex and group (matches and mismatches) were assessed through the Mann-Whitney test (p < 0.05). On average, the RMS value was almost ten times smaller in matches (0.22 ± 0.11 mm) than in mismatches (2.16 ± 0.57 mm) with a statistically significant difference according to group (p < 0.05), but not to sex (p > 0.05). The study proposed a new method for assessing personal identification from segmented 3D models of sphenoid sinuses, useful in the forensic contexts where other methods might not be implementable or successful.

Post-mortem computed tomography as part of dental identification - a proposed guideline

PURPOSE

This paper presents a proposed guideline for the use of post-mortem computed tomography (PMCT) during forensic dental identification. Currently, whole-body PMCT is widely used prior to autopsies for the diagnosis of fractures, organ changes, hemorrhages, and for the localization of foreign bodies, but it may also facilitate the odontological identification process in single cases and in cases involving multiple fatalities. Several studies have described the use of PMCT in forensic odontological work, but we have not found any comprehensive set of guidelines on how to perform a forensic odontological examination using PMCT. The aim was to develop guidelines for creating post-mortem dental charts during forensic odontological identification examinations using the standard functions of PMCT.

METHODS

A proposed guideline was developed from 15 selected cases examined at the Section of Forensic Pathology, Department of Forensic Medicine at the University of Copenhagen in Denmark from October 2011 to May 2012. Using the functionalities and three-dimensional (3D) reconstructions of OsiriX DICOM-viewer software (Pixmeo Sarl, Bernex, Geneva, Switzerland) we adjusted the contrast and brightness settings and developed a proposed guideline for creating PMCT-based dental charts. A four-step guideline was produced.

CONCLUSION

In our casework, we are currently using the guidelines proposed herein. The use of PMCT has allowed us to target our clinical examinations, greatly improving their efficiency. Furthermore, PMCT allows the storage of data for later documentation and research. Further research is needed to validate the proposed guideline.