Review

Authors: M A Sacco, P Tarzia, L Tarda...

BACKGROUND

In the evolving field of forensic medicine, artificial intelligence (AI) technologies may revolutionize traditional autopsy practices by enhancing the precision and efficiency of postmortem examinations.

METHODS

A review of the literature was carried out on the Pub-med and Scopus search engines by inserting the keywords "artificial intelligence" AND "forensic" AND ("autopsy" OR "crime scene management" OR "forensic odontology" OR "post mortem interval" OR "forensic anthropology" OR "forensic sciences"). The works that analyzed the applications of artificial intelligence in the forensic and autopsy field were analyzed.

CONCLUSION

The results showed the application of different forms of artificial intelligence such as machine learning, deep learning, robotics, artificial neural networks. Various applications are therefore possible in the autopsy field including forensic identification, analysis of radiological data through Virtopsy, estimation of the weapon used through analysis of firearm damage with ballistics, estimation of the Post-Mortem Interval (PMI), forensic toxicology. AI's potential to aid in the precise identification of causes of death, estimation of postmortem intervals. With forensic pathologists facing the constant challenge of making accurate diagnoses under pressure, AI applications can offer much-needed support by reducing subjective judgment and the inherent human error due to fatigue. Therefore, the integration of AI into autopsies, while promising in terms of efficiency and accuracy, demands a careful balance between technological advancement and ethical responsibility to ensure trust and integrity in forensic practices.

Topics: Humans; Artificial Intelligence; Autopsy; Forensic Medicine; Machine Learning; Neural Networks, Computer

PubMed: 39101424
DOI: 10.7417/CT.2024.5114

Authors: Jeffrey Sklar

This Guest Editorial highlights the importance of autopsies in biomedical discovery.

Topics: Autopsy; Genetic Testing; Genomics; Humans; Pathology, Clinical

PubMed: 31199923
DOI: 10.1016/j.ajpath.2019.05.006

Authors: Murat Nihat Arslan, Yalçın Büyük, Nihan Ziyade...

BACKGROUND/AIMS

The aim of this study is to share autopsy findings of COVID-19-positive cases and autopsy algorithms for safely handling of suspicious bodies during this pandemic.

METHODS

COVID-19-positive cases of Istanbul Morgue Department were retrospectively analyzed. Sampling indications for PCR tests in suspicious deaths, macroscopic and microscopic findings obtained in cases with positive PCR tests were evaluated.

RESULTS

In the morgue department, 345(25.8%) of overall 1336 autopsy cases were tested for COVID-19. PCR test was found positive in 26 cases. Limited autopsy procedure was performed in 7 cases, while the cause of death was determined by external examination in the remaining 19 cases. Male-to-female ratio was found 3.3:1 and mean age was 60.0 ± 13.6 among all PCR-positive cases. Cause of death was determined as viral pneumonia in fully autopsied cases. Most common findings were sticky gelatinous fluid in cavities and firm and swollen lungs, varying degrees of consolidation. In microscopy, diffuse alveolar epithelial damage, type-II pneumocyte hyperplasia, hyaline membrane formation, fibrinous exudate, and fibrinous plaques in the alveoli were the most common findings.

CONCLUSIONS

In COVID-19 autopsies, pulmonary findings were found to be prominent and the main pathology was pneumonia. Older age and findings of chronic diseases indicate that the cases were in the multirisk group in terms of COVID-19 mortality.

Topics: Aged; Autopsy; COVID-19; Female; Humans; Lung; Male; Middle Aged; Pneumonia, Viral; Retrospective Studies; SARS-CoV-2

PubMed: 33755916
DOI: 10.1007/s11845-021-02602-6

Review

Authors: Cristina Basso, James R Stone

Historically, autopsy contributed to our current knowledge of cardiovascular anatomy, physiology, and pathology. Major advances in the understanding of cardiovascular diseases, including atherosclerosis and coronary artery disease, congenital heart diseases, and cardiomyopathies, were possible through autopsy investigations and clinicopathological correlations. In this review, the importance of performing clinical autopsies in people dying from cardiovascular disease, even in the era of advanced cardiovascular imaging is addressed. Autopsies are most helpful in the setting of sudden unexpected deaths, particularly when advanced cardiovascular imaging has not been performed. In this setting, the autopsy is often the only chance to make the correct diagnosis. In previously symptomatic patients who had undergone advanced cardiovascular imaging, autopsies still play many roles. Post-mortem examinations are important for furthering the understanding of key issues related to the underlying diseases. Autopsy can help to increase the knowledge of the sensitivity and specificity of advanced cardiovascular imaging modalities. Autopsies are particularly important to gain insights into both the natural history of cardiovascular diseases as well as less common presentations and therapeutic complications. Finally, autopsies are a key tool to quickly understand the cardiac pathology of new disorders, as emphasized during the recent coronavirus disease 2019 pandemic.

Topics: Autopsy; COVID-19; Cardiovascular Diseases; Cardiovascular System; Cause of Death; Death, Sudden; Humans

PubMed: 35514073
DOI: 10.1093/eurheartj/ehac220

Review

Authors: Ram Narayan Das, Bappa Mandal, Mou Das...

Fetal and perinatal autopsies are useful to identify the accurate cause of death and in the process recognize disorders which may require counselling for future pregnancies. Abnormalities of the CNS are an important cause of fetal loss and perinatal deaths. Most of these are structural abnormalities of the CNS, however a smaller portion show changes pertaining to prematurity, infections and even congenital tumors. In this review we evaluate CNS abnormalities of the fetus and the newborn as detected in autopsy series. We also describe our experience in a tertiary care hospital with a specialized neonatology unit over the last 8 years and discuss some of the newer methods like virtual autopsy.

Topics: Autopsy; Congenital Abnormalities; Female; Fetal Death; Fetus; Humans; Infant, Newborn; Pregnancy

PubMed: 35562151
DOI: 10.4103/ijpm.ijpm_977_21

Authors: Leigh Anne Tang, Jessica Korona-Bailey, Dimitrios Zaras...

BACKGROUND

Fatal drug overdose surveillance informs prevention but is often delayed because of autopsy report processing and death certificate coding. Autopsy reports contain narrative text describing scene evidence and medical history (similar to preliminary death scene investigation reports) and may serve as early data sources for identifying fatal drug overdoses. To facilitate timely fatal overdose reporting, natural language processing was applied to narrative texts from autopsies.

OBJECTIVE

This study aimed to develop a natural language processing-based model that predicts the likelihood that an autopsy report narrative describes an accidental or undetermined fatal drug overdose.

METHODS

Autopsy reports of all manners of death (2019-2021) were obtained from the Tennessee Office of the State Chief Medical Examiner. The text was extracted from autopsy reports (PDFs) using optical character recognition. Three common narrative text sections were identified, concatenated, and preprocessed (bag-of-words) using term frequency-inverse document frequency scoring. Logistic regression, support vector machine (SVM), random forest, and gradient boosted tree classifiers were developed and validated. Models were trained and calibrated using autopsies from 2019 to 2020 and tested using those from 2021. Model discrimination was evaluated using the area under the receiver operating characteristic, precision, recall, F-score, and F-score (prioritizes recall over precision). Calibration was performed using logistic regression (Platt scaling) and evaluated using the Spiegelhalter z test. Shapley additive explanations values were generated for models compatible with this method. In a post hoc subgroup analysis of the random forest classifier, model discrimination was evaluated by forensic center, race, age, sex, and education level.

RESULTS

A total of 17,342 autopsies (n=5934, 34.22% cases) were used for model development and validation. The training set included 10,215 autopsies (n=3342, 32.72% cases), the calibration set included 538 autopsies (n=183, 34.01% cases), and the test set included 6589 autopsies (n=2409, 36.56% cases). The vocabulary set contained 4002 terms. All models showed excellent performance (area under the receiver operating characteristic ≥0.95, precision ≥0.94, recall ≥0.92, F-score ≥0.94, and F-score ≥0.92). The SVM and random forest classifiers achieved the highest F-scores (0.948 and 0.947, respectively). The logistic regression and random forest were calibrated (P=.95 and P=.85, respectively), whereas the SVM and gradient boosted tree classifiers were miscalibrated (P=.03 and P<.001, respectively). "Fentanyl" and "accident" had the highest Shapley additive explanations values. Post hoc subgroup analyses revealed lower F-scores for autopsies from forensic centers D and E. Lower F-score were observed for the American Indian, Asian, ≤14 years, and ≥65 years subgroups, but larger sample sizes are needed to validate these findings.

CONCLUSIONS

The random forest classifier may be suitable for identifying potential accidental and undetermined fatal overdose autopsies. Further validation studies should be conducted to ensure early detection of accidental and undetermined fatal drug overdoses across all subgroups.

Topics: Humans; Autopsy; Natural Language Processing; Drug Overdose; Algorithms; Random Forest

PubMed: 37204824
DOI: 10.2196/45246

Review

Authors: K G Raviraj, S S Shobhana

Many articles on COVID19 deaths have been published since the pandemic has occurred. On reviewing the articles published until June 2021, the findings were very heterogeneous. Adding to the existing knowledge, there were also some unique observations made in the pathogenesis of COVID19. This review was done to determine the findings obtained and inferences drawn from various studies published globally among patients who died due to COVID19. PRISMA guidelines were used to conduct this systematic review. A search of databases like PubMed, ScienceDirect and Epistemonikos was done. The articles focusing on postmortem sample studies involving full autopsies, minimally invasive autopsies and tissue biopsy studies were screened and searched. The studies included were all the case reports, case series, narrative reviews and systematic reviews obtained in full text and in the English language containing study information, and samples obtained postmortem. The information obtained was tabulated using Microsoft excel sheets. The duplicates were removed at the beginning of the tabulation. Zotero referencing software was used for article sorting and citation and bibliography. Two authors independently reviewed the articles throughout the process to prevent bias. Adding to the heterogeneity of COVID19, the concept of lethality in preexisting disease conditions, the occurrence of secondary bacterial and fungal infections, and other pathogenetic mechanisms uniquely encountered are to be considered in treating the patients. Also, the presence of SARS-CoV-2 postmortem is established and should be considered a hazard.

Topics: Humans; Autopsy; COVID-19; SARS-CoV-2; Pandemics; Biopsy

PubMed: 35817946
DOI: 10.1007/s12024-022-00494-1

Review

Authors: Stephan A Bolliger, Michael J Thali

In order to create a three-dimensional (3D) documentation of findings which can be reassessed if necessary by other experts, the research project 'Virtopsy®' was launched in the late 1990s. This project combined autopsy results with forensic imaging in the form of computed tomography, magnetic resonance tomography and 3D surface scanning. The success of this project eventually succeeded in convincing the courts in Switzerland to accept these novel methods as evidence. As opposition towards autopsies has grown over the last decades, Virtopsy also strives to find and elaborate additional methods which can answer the main forensic questions without autopsy. These methods comprise post-mortem angiography for illustration of the vascular bed and image-guided tissue and fluid sampling for histological, toxicological and microbiological examinations. Based on the promising results, post-mortem imaging, especially with 3D surface scanning, has meanwhile also been applied to living victims of assault, who have suffered patterned injuries due to bites, blows with objects, etc. In our opinion, forensic imaging is an objective method which offers the possibility for a reassessment of the findings by other experts, even after burial or cremation of the corpse, or healing of the injuries in living victims, thus leading to a greater security in court.

Topics: Autopsy; Forensic Medicine; Humans; Imaging, Three-Dimensional; Magnetic Resonance Imaging; Tomography, X-Ray Computed

PubMed: 26101279
DOI: 10.1098/rstb.2014.0253

Authors: Camila Penso, Andréa L Corso, Cláudia R Hentges...

OBJECTIVES

To evaluate neonatal autopsy rates at a tertiary hospital in southern Brazil ascertain the level of agreement between premortem and postmortem diagnosis.

METHODS

The authors reviewed all neonatal autopsies performed over a 10-year period and described the percentage of neonates who died and underwent autopsy. The authors tested for agreement between autopsy findings and the cause of death as defined by the neonatologist. Agreement between clinical diagnosis and autopsy findings was classified using the modified Goldman criteria. Additional findings at autopsy were grouped by organ system. Linear regression and multiple comparisons were used for statistical analyses.

RESULTS

During the study period, 382 neonates died at the Neonatal Intensive Care Unit (NICU). Consent to perform an autopsy was obtained for 73 (19.1%). The complete agreement between autopsy findings and the neonatologist's premortem diagnosis was found in 48 patients (65.8%). Additional findings were obtained at autopsy in 25 cases (34.2%). In 5 cases (6.9%), the autopsy findings contributed to subsequent genetic counseling. Seven autopsies (9.6%) revealed a diagnosis that would have changed patient management if established premortem. The autopsy rate increased by an average of 1.87% each year.

CONCLUSION

Despite a high level of agreement between clinical diagnosis and pathological findings, autopsies provided relevant data regarding the cause of death, providing additional clinical information to neonatologists and allowing genetic counseling of family members.

Topics: Autopsy; Brazil; Cause of Death; Humans; Infant, Newborn; Intensive Care Units, Neonatal; Linear Models; Retrospective Studies

PubMed: 35189081
DOI: 10.1016/j.jped.2022.01.002

Review

Authors: Mattia Riefolo, Francesca Ambrosi, Alessandra De Palma...

A brief overview on the management of autopsies during the SARS-CoV-19 epidemic is proposed. In particular, the point is made of the Italian laws on the subject, the characteristics required for the autopsy room and the sampling suggested for the histological examination.

Topics: Humans; Autopsy; Severe acute respiratory syndrome-related coronavirus

PubMed: 38054900
DOI: 10.32074/1591-951X-921