Scholarlink

Background <p>The increasing implementation of artificial intelligence (AI) in forensic medicine, similar to other medical fields, is foreseeable. Against this backdrop, the question arises regarding the challenges associated with the use of AI in forensic medicine, a&#xa0;topic that has not yet been addressed.</p> Objective <p>This article examines key issues and the resulting legal and ethical challenges in the implementation of AI into forensic practice. Additionally, potential solution approaches are presented.</p> Material and methods <p>The publication is based on a&#xa0;systematic literature analysis of AI applications in forensic medicine and related fields, such as medical diagnostics and digital forensics. It also considers legal frameworks, particularly the European AI Act, as well as perspectives from AI ethics.</p> Results and discussion <p>For AI applications in forensic medicine six key challenges were identified: reliability, transparency and explainability (the black box problem), accountability for errors, data quantity and quality, data bias and fairness, and the acceptance of AI-generated diagnoses. Various approaches are proposed to address the identified challenges. These include the use of “explainable AI” (in short: XAI) in conjunction with AI experts, the development of standard protocols and multicenter collaborations to ensure data quality and system reliability. Furthermore, enhanced interdisciplinary collaboration between computer scientists, forensic pathologists, legal experts and ethicists is advocated to promote the acceptance and integration of AI systems into practice.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

„KI“ in der Rechtsmedizin – von der Forschung in die Praxis: Welche Herausforderungen ergeben sich?

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年3月7日

查看全文
  • M. Rüther,
  • S. B. Eickhoff,
  • L. König,
  • Stefanie Ritz,
  • B. Schäffer

摘要

Background

The increasing implementation of artificial intelligence (AI) in forensic medicine, similar to other medical fields, is foreseeable. Against this backdrop, the question arises regarding the challenges associated with the use of AI in forensic medicine, a topic that has not yet been addressed.

Objective

This article examines key issues and the resulting legal and ethical challenges in the implementation of AI into forensic practice. Additionally, potential solution approaches are presented.

Material and methods

The publication is based on a systematic literature analysis of AI applications in forensic medicine and related fields, such as medical diagnostics and digital forensics. It also considers legal frameworks, particularly the European AI Act, as well as perspectives from AI ethics.

Results and discussion

For AI applications in forensic medicine six key challenges were identified: reliability, transparency and explainability (the black box problem), accountability for errors, data quantity and quality, data bias and fairness, and the acceptance of AI-generated diagnoses. Various approaches are proposed to address the identified challenges. These include the use of “explainable AI” (in short: XAI) in conjunction with AI experts, the development of standard protocols and multicenter collaborations to ensure data quality and system reliability. Furthermore, enhanced interdisciplinary collaboration between computer scientists, forensic pathologists, legal experts and ethicists is advocated to promote the acceptance and integration of AI systems into practice.