Renato Queiroz Nogueira Lira, Luana Geovana Motta de Sousa, Maisa Luana Memoria Pinho, Renan Cesar Pinto da Silva Andrade de Lima, Pedro Garcia Freitas, Bruno Scholles Soares Dias, Andreia Cristina Breda de Souza, André Ferreira Leite
{"title":"基于深度学习的人体枪伤分类。","authors":"Renato Queiroz Nogueira Lira, Luana Geovana Motta de Sousa, Maisa Luana Memoria Pinho, Renan Cesar Pinto da Silva Andrade de Lima, Pedro Garcia Freitas, Bruno Scholles Soares Dias, Andreia Cristina Breda de Souza, André Ferreira Leite","doi":"10.1007/s00414-024-03355-4","DOIUrl":null,"url":null,"abstract":"<p><p>In this paper, we present a forensic perspective on classifying gunshot wound patterns using Deep Learning (DL). Although DL has revolutionized various medical specialties, such as automating tasks like medical image classification, its applications in forensic contexts have been limited despite the inherently visual nature of the field. This study investigates the application of DL techniques (59 architectures) to classify gunshot wounds in a forensic context, focusing on distinguishing between entry and exit wounds and determining the Medical-Legal Shooting Distance (MLSD), which classifies wounds as contact, close range, or distant, based on digital images from real crime scene cases. A comprehensive database was constructed with 2,551 images, including 1,883 entries and 668 exit wounds. The ResNet152 architecture demonstrated superior performance in both entry and exit wound classification and MLSD categorization. For the first task, achieved accuracy of 86.90% and an AUC of 82.09%. For MLSD, the ResNet152 showed an accuracy of 92.48% and AUC up to 94.36%, though sample imbalance affected the metrics. Our findings underscore the challenges of standardizing wound images due to varying capture conditions but reflect the practical realities of forensic work. This research highlights the significant potential of DL in enhancing forensic pathology practices, advocating for Artificial Intelligence (AI) as a supportive tool to complement human expertise in forensic investigations.</p>","PeriodicalId":14071,"journal":{"name":"International Journal of Legal Medicine","volume":" ","pages":""},"PeriodicalIF":2.2000,"publicationDate":"2024-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Deep learning-based human gunshot wounds classification.\",\"authors\":\"Renato Queiroz Nogueira Lira, Luana Geovana Motta de Sousa, Maisa Luana Memoria Pinho, Renan Cesar Pinto da Silva Andrade de Lima, Pedro Garcia Freitas, Bruno Scholles Soares Dias, Andreia Cristina Breda de Souza, André Ferreira Leite\",\"doi\":\"10.1007/s00414-024-03355-4\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p>In this paper, we present a forensic perspective on classifying gunshot wound patterns using Deep Learning (DL). Although DL has revolutionized various medical specialties, such as automating tasks like medical image classification, its applications in forensic contexts have been limited despite the inherently visual nature of the field. This study investigates the application of DL techniques (59 architectures) to classify gunshot wounds in a forensic context, focusing on distinguishing between entry and exit wounds and determining the Medical-Legal Shooting Distance (MLSD), which classifies wounds as contact, close range, or distant, based on digital images from real crime scene cases. A comprehensive database was constructed with 2,551 images, including 1,883 entries and 668 exit wounds. The ResNet152 architecture demonstrated superior performance in both entry and exit wound classification and MLSD categorization. For the first task, achieved accuracy of 86.90% and an AUC of 82.09%. For MLSD, the ResNet152 showed an accuracy of 92.48% and AUC up to 94.36%, though sample imbalance affected the metrics. Our findings underscore the challenges of standardizing wound images due to varying capture conditions but reflect the practical realities of forensic work. This research highlights the significant potential of DL in enhancing forensic pathology practices, advocating for Artificial Intelligence (AI) as a supportive tool to complement human expertise in forensic investigations.</p>\",\"PeriodicalId\":14071,\"journal\":{\"name\":\"International Journal of Legal Medicine\",\"volume\":\" \",\"pages\":\"\"},\"PeriodicalIF\":2.2000,\"publicationDate\":\"2024-11-06\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"International Journal of Legal Medicine\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://doi.org/10.1007/s00414-024-03355-4\",\"RegionNum\":3,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"MEDICINE, LEGAL\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal of Legal Medicine","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1007/s00414-024-03355-4","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"MEDICINE, LEGAL","Score":null,"Total":0}
Deep learning-based human gunshot wounds classification.
In this paper, we present a forensic perspective on classifying gunshot wound patterns using Deep Learning (DL). Although DL has revolutionized various medical specialties, such as automating tasks like medical image classification, its applications in forensic contexts have been limited despite the inherently visual nature of the field. This study investigates the application of DL techniques (59 architectures) to classify gunshot wounds in a forensic context, focusing on distinguishing between entry and exit wounds and determining the Medical-Legal Shooting Distance (MLSD), which classifies wounds as contact, close range, or distant, based on digital images from real crime scene cases. A comprehensive database was constructed with 2,551 images, including 1,883 entries and 668 exit wounds. The ResNet152 architecture demonstrated superior performance in both entry and exit wound classification and MLSD categorization. For the first task, achieved accuracy of 86.90% and an AUC of 82.09%. For MLSD, the ResNet152 showed an accuracy of 92.48% and AUC up to 94.36%, though sample imbalance affected the metrics. Our findings underscore the challenges of standardizing wound images due to varying capture conditions but reflect the practical realities of forensic work. This research highlights the significant potential of DL in enhancing forensic pathology practices, advocating for Artificial Intelligence (AI) as a supportive tool to complement human expertise in forensic investigations.
期刊介绍:
The International Journal of Legal Medicine aims to improve the scientific resources used in the elucidation of crime and related forensic applications at a high level of evidential proof. The journal offers review articles tracing development in specific areas, with up-to-date analysis; original articles discussing significant recent research results; case reports describing interesting and exceptional examples; population data; letters to the editors; and technical notes, which appear in a section originally created for rapid publication of data in the dynamic field of DNA analysis.