脑瘫儿童家庭AI模型问答有效性比较数据集（ChatGPT, Gemini, DeepSeek）

数据集获取（Figshare）

• 源站标题：Table 2_The effectiveness of artificial intelligence models in addressing the concerns of families of children with cerebral palsy: a comparative analysis of ChatGPT, Gemini, and DeepSeek.docx
• DOI：10.3389/fped.2025.1694035.s002
• 发布日期：2025-12-08
• 文件数：1 个
• 文件总体积：42.85 KB
• 源站页面：https://frontiersin.figshare.com/articles/dataset/Table_2_The_effectiveness_of_artificial_intelligence_models_in_addressing_the_concerns_of_families_of_children_with_cerebral_palsy_a_comparative_analysis_of_ChatGPT_Gemini_and_DeepSeek_docx/30817787

命令行下载

curl -L -o "Table 2_The effectiveness of artificial intelligence models in addressing the concerns of families of children with cerebral palsy: a comparative anal.docx" "https://ndownloader.figshare.com/files/60176747"

文件清单

• Table 2_The effectiveness of artificial intelligence models in addressing the concerns of families of children with cerebral palsy: a comparative anal.docx（42.85 KB）

文件总体积 42.85 KB（源站 API 实测）。

数据加载示例（图像类）

from PIL import Image
import glob, os

files = (glob.glob(os.path.join(path, "**", "*.png"), recursive=True)
       + glob.glob(os.path.join(path, "**", "*.jpg"), recursive=True)
       + glob.glob(os.path.join(path, "**", "*.tif"), recursive=True))
print("图像文件数:", len(files))
img = Image.open(files[0]); print("尺寸/模式:", img.size, img.mode)

# torchvision Dataset 方式:
# from torchvision import datasets
# ds = datasets.ImageFolder(path)  # 要求 子目录=类别

目录组织与标注格式以源站说明和下载后实际文件为准。

数据说明

BackgroundCerebral palsy (CP) is a non-progressive but permanent motor disorder that significantly affects children and their families. With the rise of artificial intelligence (AI)-based information systems, families increasingly use these tools to address their concerns.
ObjectiveThis study compared the clinical relevance of responses from ChatGPT, Gemini, and DeepSeek to frequently asked CP-related questions by families.
MethodsTen key questions, compiled from reputable medical websites, were posed separately to each AI model. Responses were rated by two independent experts using a 4-point Likert scale, with a third reviewer resolving discrepancies.
ResultsGemini achieved the highest mean score (3.2), followed by ChatGPT (2.9) and DeepSeek (2.9). All models provided strong general CP information but underperformed in complex areas such as surgical indications. Gemini gave more structured and comprehensive responses, while ChatGPT and DeepSeek occasionally lacked detail or clarity.
ConclusionWhile AI language models can offer useful CP-related information, their reliability in complex clinical decision-making remains limited. Expert oversight is essential, and future systems should integrate multimodal capabilities for improved family guidance and engagement.

许可

CC BY 4.0

引用

学术使用请引用 DOI 10.3389/fped.2025.1694035.s002（source: https://frontiersin.figshare.com/articles/dataset/Table_2_The_effectiveness_of_artificial_intelligence_models_in_addressing_the_concerns_of_families_of_children_with_cerebral_palsy_a_comparative_analysis_of_ChatGPT_Gemini_and_DeepSeek_docx/30817787）。