Scholarlink

<p>This study was conducted using an adopted design-based research (DBR) approach to investigate the impact of artificial intelligence (AI)-supported object recognition technology on middle school students’ learning of the “Systems in Our Body” unit. A dataset capable of identifying human organs was developed using the YOLOv8 algorithm and the Roboflow platform, and this dataset was integrated into the instructional process. The research consisted of four main stages: initial design, pilot implementation, revision, and main implementation. During the pilot implementation, both technical and pedagogical issues were identified, leading to the redesign of the lesson scenario. The main implementation was carried out with 5th grade students. A quasi-experimental pretest–posttest control group design was used to measure the effectiveness of the intervention. In addition, qualitative data were collected through a focus group interview with the experimental group students. Quantitative findings revealed that the AI-supported dataset significantly improved students’ academic achievement. Qualitative findings indicated positive changes in students’ cognitive and affective development. The study suggests that AI-assisted instructional tools can enhance science learning processes and increase classroom engagement. These findings suggest that AI-supported image recognition can foster both conceptual understanding and motivation in science education.&#xa0;&#xa0;</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

From Traditional to AI-Enhanced Science Learning: Real-time Organ Recognition with YOLOv8

  • 期刊论文

  • 发表日期：

    2026年6月27日

查看全文
  • Gülsüm Meço,
  • Aslı Görgülü

摘要

This study was conducted using an adopted design-based research (DBR) approach to investigate the impact of artificial intelligence (AI)-supported object recognition technology on middle school students’ learning of the “Systems in Our Body” unit. A dataset capable of identifying human organs was developed using the YOLOv8 algorithm and the Roboflow platform, and this dataset was integrated into the instructional process. The research consisted of four main stages: initial design, pilot implementation, revision, and main implementation. During the pilot implementation, both technical and pedagogical issues were identified, leading to the redesign of the lesson scenario. The main implementation was carried out with 5th grade students. A quasi-experimental pretest–posttest control group design was used to measure the effectiveness of the intervention. In addition, qualitative data were collected through a focus group interview with the experimental group students. Quantitative findings revealed that the AI-supported dataset significantly improved students’ academic achievement. Qualitative findings indicated positive changes in students’ cognitive and affective development. The study suggests that AI-assisted instructional tools can enhance science learning processes and increase classroom engagement. These findings suggest that AI-supported image recognition can foster both conceptual understanding and motivation in science education.