Potential use of deep learning techniques for postmortem imaging

Postmortem Temporal Changes in Liver and Spleen Stiffness: Evaluation with Shear Wave Elastography in a Rat Model

Background/Objectives: Postmortem changes in tissue stiffness and organ morphology are critical for forensic medicine and pathology. Shear wave elastography (SWE) has emerged as a non-invasive tool to assess tissue stiffness, yet its potential for postmortem interval estimation remains underexplored. While previous studies have demonstrated early postmortem alterations in tissue elasticity, the temporal progression of these changes in different organs is not fully understood. This study aims to investigate the temporal changes in liver and spleen stiffness during the postmortem period using SWE and to evaluate the predictive potential of elastographic parameters for postmortem interval estimation. Methods: Twelve male Sprague-Dawley rats were sacrificed via cervical dislocation following deep anesthesia. Postmortem liver and spleen measurements, including longitudinal and short diameters and SWE values (kPa), were recorded at 0, 2, 4, 6, 9, 12, 18, 24, and 36 h. All elastographic measurements were obtained using a 5 mm circular region of interest (ROI) for the liver and a 3 mm ROI for the spleen. Changes over time were analyzed using repeated measures ANOVA, with post hoc Bonferroni corrections applied where necessary. Additionally, Receiver Operating Characteristic (ROC) curve analysis and binary logistic regression analysis were performed to assess the predictive accuracy of SWE parameters in estimating postmortem time. Results: Postmortem liver and spleen stiffness exhibited a significant declining trend over time (p < 0.001, η2 = 0.749 and η2 = 0.810, respectively). Liver and spleen dimensions initially increased, reaching peak values around 6 h, followed by a gradual reduction. ROC analysis demonstrated that spleen SWE (AUC = 0.917) and liver SWE (AUC = 0.845) were the strongest predictors of early postmortem time. Binary logistic regression further confirmed that liver and spleen SWE were statistically significant predictors of postmortem time (p = 0.006 and p = 0.020, respectively). Conclusions: This study provides evidence that postmortem liver and spleen stiffness decline progressively over time, while organ dimensions exhibit a biphasic pattern. Elastographic parameters, particularly SWE values, demonstrated strong predictive accuracy in estimating early postmortem intervals. These findings suggest that SWE may serve as a valuable imaging modality for forensic applications, providing objective insights into postmortem biomechanical changes and time-of-death estimation. Further research should explore the applicability of SWE in different tissue types and under varying environmental conditions.

Postmortem changes in porcine eyes on computed tomography images

Porcine eyes were examined using postmortem computed tomography (PMCT) under controlled postmortem time and temperature conditions to assess the mechanisms and timing of changes in ocular structure. Eight porcine heads were halved, and PMCT scans were conducted from postmortem interval (PMI) days 0 to 13. CT images were obtained to evaluate the vitreous volumes, vitreous CT values, axial lengths of the eyes, lens dislocation, and intraocular gas. The vitreous volume decreased over time, with the highest median rate of 17.7 % at PMI 1, followed by 12.0 %, 11.7 %, and 11.3 % at PMIs 6, 7, and 8, respectively. There was a significant decrease in the axial eye length from PMIs 0 to 1, while the transverse diameter remained unchanged. Lens dislocation was observed in all cases at PMI 9. Receiver operating characteristic analysis using the PMI as the predictive value for the presence of lens dislocation revealed a cutoff value of PMI 6, with an area under the curve of 0.98. Intraocular gas was observed in four cases. In two cases with intraocular gas, intravascular gas appeared to be continuous with the intraocular gas via the ciliary body. Lens dislocation occurred 6 days postmortem in porcine eyes at moderate temperatures. Intraocular gas was also observed 6 days postmortem, which may have been caused by the influx of intravascular gas into the eye via the ciliary body. These structural changes in the porcine model, may help in estimating the time of death.

Diagnostic Challenges in Uncommon Firearm Injury Cases: A Multidisciplinary Approach

Background: Firearm wounds tend to have a precise pattern. Despite this, real-world case presentations can present uncertain elements, sometimes deviating from what is considered standard, and present uncommon features that are difficult for forensic pathologists and ballistic experts to explain. Methods: A retrospective analysis of autopsy reports from the Institute of Legal Medicine, University of Catania, covering 2019-2023, included 348 judicial inspections and 378 autopsies performed as part of the institute's overall activities. Among these, seventeen cases of firearm deaths were identified, with three atypical cases selected for detailed analysis. An interdisciplinary approach involving forensic pathology, radiology, and ballistics was used. Results: The selected cases included: (1) A 56-year-old female with a thoracic gunshot wound involving three 7.65 caliber bullets, displaying complex trajectories and retained bullets; (2) A 48-year-old male with two cranial gunshot injuries, where initial evaluation suggested homicide staged as a suicide, later confirmed to be a single self-inflicted shot; and (3) A 51-year-old male was found in a car with two gunshot wounds to the head, involving complex forensic evaluation to distinguish between entrance and exit wounds and determine trajectory. The findings showed significant deviations from standard patterns, underscoring the critical role of radiological imaging and ballistic analysis in understanding wound morphology and projectile trajectories. Conclusions: This case series highlights the necessity for standardized yet adaptable protocols and cooperation among forensic specialists. A flexible approach allows forensic investigations to be tailored to the specific circumstances of each case, ensuring that essential examinations are conducted while unnecessary procedures are avoided. Comprehensive data collection from autopsies, gross organ examinations, and, when needed, radiological and histological analysis is essential to accurately diagnose injuries, trace bullet trajectories, retrieve retained projectiles, and determine the fatal wound, particularly in complex cases or those involving multiple shooters.

Automated detection of fatal cerebral haemorrhage in postmortem CT data

During the last years, the detection of different causes of death based on postmortem imaging findings became more and more relevant. Especially postmortem computed tomography (PMCT) as a non-invasive, relatively cheap, and fast technique is progressively used as an important imaging tool for supporting autopsies. Additionally, previous works showed that deep learning applications yielded robust results for in vivo medical imaging interpretation. In this work, we propose a pipeline to identify fatal cerebral haemorrhage on three-dimensional PMCT data. We retrospectively selected 81 PMCT cases from the database of our institute, whereby 36 cases suffered from a fatal cerebral haemorrhage as confirmed by autopsy. The remaining 45 cases were considered as neurologically healthy. Based on these datasets, six machine learning classifiers (k-nearest neighbour, Gaussian naive Bayes, logistic regression, decision tree, linear discriminant analysis, and support vector machine) were executed and two deep learning models, namely a convolutional neural network (CNN) and a densely connected convolutional network (DenseNet), were trained. For all algorithms, 80% of the data was randomly selected for training and 20% for validation purposes and a five-fold cross-validation was executed. The best-performing classification algorithm for fatal cerebral haemorrhage was the artificial neural network CNN, which resulted in an accuracy of 0.94 for all folds. In the future, artificial neural network algorithms may be applied by forensic pathologists as a helpful computer-assisted diagnostics tool supporting PMCT-based evaluation of cause of death.

Deep learning for acute rib fracture detection in CT data: a systematic review and meta-analysis

OBJECTIVES

To review studies on deep learning (DL) models for classification, detection, and segmentation of rib fractures in CT data, to determine their risk of bias (ROB), and to analyse the performance of acute rib fracture detection models.

METHODS

Research articles written in English were retrieved from PubMed, Embase, and Web of Science in April 2023. A study was only included if a DL model was used to classify, detect, or segment rib fractures, and only if the model was trained with CT data from humans. For the ROB assessment, the Quality Assessment of Diagnostic Accuracy Studies tool was used. The performance of acute rib fracture detection models was meta-analysed with forest plots.

RESULTS

A total of 27 studies were selected. About 75% of the studies have ROB by not reporting the patient selection criteria, including control patients or using 5-mm slice thickness CT scans. The sensitivity, precision, and F1-score of the subgroup of low ROB studies were 89.60% (95%CI, 86.31%-92.90%), 84.89% (95%CI, 81.59%-88.18%), and 86.66% (95%CI, 84.62%-88.71%), respectively. The ROB subgroup differences test for the F1-score led to a p-value below 0.1.

CONCLUSION

ROB in studies mostly stems from an inappropriate patient and data selection. The studies with low ROB have better F1-score in acute rib fracture detection using DL models.

ADVANCES IN KNOWLEDGE

This systematic review will be a reference to the taxonomy of the current status of rib fracture detection with DL models, and upcoming studies will benefit from our data extraction, our ROB assessment, and our meta-analysis.

Raman spectroscopy for postmortem interval estimation of human skeletal remains: A scoping review

Estimating postmortem intervals (PMI) is crucial in forensic investigations, providing insights into criminal cases and determining the time of death. PMI estimation relies on expert experience and a combination of thanatological data and environmental factors but is prone to errors. The lack of reliable methods for assessing PMI in bones and soft tissues necessitates a better understanding of bone decomposition. Several research groups have shown promise in PMI estimation in skeletal remains but lack valid data for forensic cases. Current methods are costly, time-consuming, and unreliable for PMIs over 5 years. Raman spectroscopy (RS) can potentially estimate PMI by studying chemical modifications in bones and teeth correlated with burial time. This review summarizes RS applications, highlighting its potential as an innovative, nondestructive, and fast technique for PMI estimation in forensic medicine.

Overcoming Fear, Uncertainty, and Doubt: Artificial Intelligence (AI) and the Value of Trust

This is an editorial based on personal experience dealing with the fear, uncertainty, and doubt regarding artificial intelligence (AI) and radiology (my field of specialization). In the end, the most important tools to engage with these are education, research, and policy or regulation with the ultimate goal of forging trust, not just in the AI but also in the people that help make this technology possible.

A comprehensive review on modelling the adsorption process for heavy metal removal from waste water using artificial neural network technique

Water is the most necessary and significant element for all life on earth. Unfortunately, the quality of the water resources is constantly declining as a result of population development, industry, and civilization progress. Due to their extreme toxicity, heavy metals removal from water has drawn researchers' attention. A lot of scientific applications use artificial neural networks (ANNs) because of their excellent ability to map nonlinear relationships. ANNs shown excellent modelling capabilities for the water treatment remediation. The adsorption process uses a variety of variables, making the interaction between them nonlinear. Selecting the best technique can produce excellent results; the adsorption approach for removing heavy metals is highly effective. Different studies show that the ANNs modelling approach can accurately forecast the adsorbed heavy metals and other contaminants in order to remove them.

Deep Neural Networks-Based Age Estimation of Cadavers Using CT Imaging of Vertebrae

Although age estimation upon death is important in the identification of unknown cadavers for forensic scientists, to the best of our knowledge, no study has examined the utility of deep neural network (DNN) models for age estimation among cadavers. We performed a postmortem computed tomography (CT) examination of 1000 and 500 male and female cadavers, respectively. These CT slices were converted into 3-dimensional images, and only the thoracolumbar region was extracted. Eighty percent of them were categorized as training datasets and the others as test datasets for both sexes. We fine-tuned the ResNet152 models using the training datasets. We conducted 4-fold cross-validation, and the mean absolute error (MAE) of the test datasets was calculated using the ensemble learning of four ResNet152 models. Consequently, the MAE of the male and female models was 7.25 and 7.16, respectively. Our study shows that DNN models can be useful tools in the field of forensic medicine.

Applications of artificial intelligence in forensic sciences: Current potential benefits, limitations and perspectives

In recent years, new studies based on artificial intelligence (AI) have been conducted in the forensic field, posing new challenges and demonstrating the advantages and disadvantages of using AI methodologies to solve forensic well-known problems. Specifically, AI technology has tried to overcome the human subjective bias limitations of the traditional approach of the forensic sciences, which include sex prediction and age estimation from morphometric measurements in forensic anthropology or evaluating the third molar stage of development in forensic odontology. Likewise, AI has been studied as an assisting tool in forensic pathology for a quick and easy identification of the taxonomy of diatoms. The present systematic review follows the PRISMA 2020 statements and aims to explore an emerging topic that has been poorly analyzed in the forensic literature. Benefits, limitations, and forensic implications concerning AI are therefore highlighted, by providing an extensive critical review of its current applications on forensic sciences as well as its future directions. Results are divided into 5 subsections which included forensic anthropology, forensic odontology, forensic pathology, forensic genetics, and other forensic branches. The discussion offers a useful instrument to investigate the potential benefits of AI in the forensic fields as well as to point out the existing open questions and issues concerning its application on real-life scenarios. Procedural notes and technical aspects are also provided to the readers.

Application of postmortem imaging modalities in cases of sudden death due to cardiovascular diseases-current achievements and limitations from a pathology perspective : Endorsed by the Association for European Cardiovascular Pathology and by the International Society of Forensic Radiology and Imaging

Postmortem imaging (PMI) is increasingly used in postmortem practice and is considered a potential alternative to a conventional autopsy, particularly in case of sudden cardiac deaths (SCD). In 2017, the Association for European Cardiovascular Pathology (AECVP) published guidelines on how to perform an autopsy in such cases, which is still considered the gold standard, but the diagnostic value of PMI herein was not analyzed in detail. At present, significant progress has been made in the PMI diagnosis of acute ischemic heart disease, the most important cause of SCD, while the introduction of postmortem CT angiography (PMCTA) has improved the visualization of several parameters of coronary artery pathology that can support a diagnosis of SCD. Postmortem magnetic resonance (PMMR) allows the detection of acute myocardial injury-related edema. However, PMI has limitations when compared to clinical imaging, which severely impacts the postmortem diagnosis of myocardial injuries (ischemic versus non-ischemic), the age-dating of coronary occlusion (acute versus old), other potentially SCD-related cardiac lesions (e.g., the distinctive morphologies of cardiomyopathies), aortic diseases underlying dissection or rupture, or pulmonary embolism. In these instances, PMI cannot replace a histopathological examination for a final diagnosis. Emerging minimally invasive techniques at PMI such as image-guided biopsies of the myocardium or the aorta, provide promising results that warrant further investigations. The rapid developments in the field of postmortem imaging imply that the diagnosis of sudden death due to cardiovascular diseases will soon require detailed knowledge of both postmortem radiology and of pathology.

Application of artificial intelligence and machine learning technology for the prediction of postmortem interval: A systematic review of preclinical and clinical studies

BACKGROUND /PURPOSE

Establishing an accurate postmortem interval (PMI) is exceptionally crucial in forensic investigation. Artificial intelligence (AI) and Machine learning (ML) models are widely employed in forensic practice. ML is a part of AI, both terms are highly associated and sometimes used interchangeably. This systematic review aims to evaluate the application and performance of AI technology for the prediction of PMI.

METHODS

Systematic literature search across different electronic databases using PubMed/Google Scholar/EMBASE/Scopus/CINAHL/Web of Science/Cochrane library was conducted from inception to 3 December 2021 for preclinical and clinical studies reported ML models for PMI estimation.

RESULTS

We identified 18 studies (12 preclinical and 06 clinical) that met the inclusion criteria in the qualitative analysis. Most of the studies employed supervised learning (N = 15), and others employed unsupervised learning (N = 3). Due to the heterogeneity of the samples, quantitative analysis was not performed.

CONCLUSION

In this systematic review, we discussed the performance of AI-based automated systems in PMI estimation. ML models have demonstrated accuracy and precision and the ability to overcome human errors and bias. However, the research is limited, conducted in primarily small, selected human populations. In addition, we suggest further research in larger population-based studies is needed to fully understand the extent of integrated ML models.

Personal identification with artificial intelligence under COVID-19 crisis: a scoping review

BACKGROUND

Artificial intelligence is useful for building objective and rapid personal identification systems. It is important to research and develop personal identification methods as social and institutional infrastructure. A critical consideration during the coronavirus disease 2019 pandemic is that there is no contact between the subjects and personal identification systems. The aim of this study was to organize the recent 5-year development of contactless personal identification methods that use artificial intelligence.

METHODS

This study used a scoping review approach to map the progression of contactless personal identification systems using artificial intelligence over the past 5 years. An electronic systematic literature search was conducted using the PubMed, Web of Science, Cochrane Library, CINAHL, and IEEE Xplore databases. Studies published between January 2016 and December 2020 were included in the study.

RESULTS

By performing an electronic literature search, 83 articles were extracted. Based on the PRISMA flow diagram, 8 eligible articles were included in this study. These eligible articles were divided based on the analysis targets as follows: (1) face and/or body, (2) eye, and (3) forearm and/or hand. Artificial intelligence, including convolutional neural networks, contributed to the progress of research on contactless personal identification methods.

CONCLUSIONS

This study clarified that contactless personal identification methods using artificial intelligence have progressed and that they have used information obtained from the face and/or body, eyes, and forearm and/or hand.

RiFNet: Automated rib fracture detection in postmortem computed tomography

Imaging techniques are widely used for medical diagnostics. In some cases, a lack of medical practitioners who can manually analyze the images can lead to a bottleneck. Consequently, we developed a custom-made convolutional neural network (RiFNet = Rib Fracture Network) that can detect rib fractures in postmortem computed tomography. In a retrospective cohort study, we retrieved PMCT data from 195 postmortem cases with rib fractures from July 2017 to April 2018 from our database. The computed tomography data were prepared using a plugin in the commercial imaging software Syngo.via whereby the rib cage was unfolded on a single-in-plane image reformation. Out of the 195 cases, a total of 585 images were extracted and divided into two groups labeled "with" and "without" fractures. These two groups were subsequently divided into training, validation, and test datasets to assess the performance of RiFNet. In addition, we explored the possibility of applying transfer learning techniques on our dataset by choosing two independent noncommercial off-the-shelf convolutional neural network architectures (ResNet50 V2 and Inception V3) and compared the performances of those two with RiFNet. When using pre-trained convolutional neural networks, we achieved an F₁ score of 0.64 with Inception V3 and an F₁ score of 0.61 with ResNet50 V2. We obtained an average F₁ score of 0.91 ± 0.04 with RiFNet. RiFNet is efficient in detecting rib fractures on postmortem computed tomography. Transfer learning techniques are not necessarily well adapted to make classifications in postmortem computed tomography.

Classifying Microscopic Acute and Old Myocardial Infarction Using Convolutional Neural Networks

ABSTRACT

Convolutional neural network (CNN) has advanced in recent years and translated from research into medical practice, most notably in clinical radiology and histopathology. Research on CNNs in forensic/postmortem pathology is almost exclusive to postmortem computed tomography despite the wealth of research into CNNs in surgical/anatomical histopathology. This study was carried out to investigate whether CNNs are able to identify and age myocardial infarction (a common example of forensic/postmortem histopathology) from histology slides. As a proof of concept, this study compared 4 CNNs commonly used in surgical/anatomical histopathology to identify normal myocardium from myocardial infarction. A total of 150 images of the myocardium (50 images each for normal myocardium, acute myocardial infarction, and old myocardial infarction) were used to train and test each CNN. One of the CNNs used (InceptionResNet v2) was able to show a greater than 95% accuracy in classifying normal myocardium from acute and old myocardial infarction. The result of this study is promising and demonstrates that CNN technology has potential applications as a screening and computer-assisted diagnostics tool in forensic/postmortem histopathology.

An algorithm for automatically generating gas, bone and foreign body visualizations from postmortem computed tomography data

Post mortem computed tomography (PMCT) can aid in localizing foreign bodies, bone fractures, and gas accumulations. The visualization of these findings play an important role in the communication of radiological findings. In this article, we present an algorithm for automated visualization of gas distributions on PMCT image data of the thorax and abdomen. The algorithm uses a combination of region growing segmentation and layering of different visualization methods to automatically generate overview images that depict radiopaque foreign bodies, bones and gas distributions in one image. The presented method was tested on 955 PMCT scans of the thorax and abdomen. The algorithm managed to generate useful images for all cases, visualizing foreign bodies as well as gas distribution. The most interesting cases are presented in this article. While this type of visualization cannot replace a real radiological analysis of the image data, it can provide a quick overview for briefings and image reports.