国家紧急医疗服务信息系统公共发布研究数据集
该数据集包含2020年美国各地紧急医疗服务机构提交的院前患者护理报告数据，数据模态为临床表格和文本记录，涵盖院外心脏骤停等事件的详细信息，如患者特征、响应时间与治疗措施，主要用于医疗健康研究领域，特别是通过无监督机器学习和因果分析来揭示城乡差异并优化紧急医疗服务策略。
基本信息
资源简介
该数据集是一个院前患者护理报告的便利样本，源自美国各地紧急医疗服务机构提交的数据。数据集基于NEMSIS 3.4.0版本，由MaineHealth组织管理并由Northeastern大学的Roux研究所维护。它包含2020年的记录，涵盖院外心脏骤停等紧急医疗事件的详细信息，如患者特征、急救响应时间、治疗措施和临床结果。主要应用于医疗健康研究领域，特别是用于分析城乡差异、评估治疗策略以及通过无监督机器学习和因果分析优化紧急医疗服务。
下载信息
注册下载
Tips: 该数据集需要在对应的数据源网站注册通过后，才能进行数据下载，注册有对应要求，或者需要收费。
暂未开放公开下载
Tips: 该数据集属于公开下载，应该可以免费公开下载。
免登录有偿下载
Tips: 该数据集 Qianfanghub 可以协助提供有偿下载服务，注意，服务不针对数据相关产权，只是技术服务费。
提供高速下载与技术交付服务（收技术服务费，非数据销售）
暂未开放千方医数集，医疗数据集部分，是为社区服务的公开医疗数据集搜索引擎，并不存储或者下载原始的任何数据。 如果您有其他医疗数据需求，可以和客服联系，或者下工单。我们有强大的三甲医疗机构帮助您提供个性化的医疗数据定制、采集、标注服务。
使用方式
数据集获取
git clone https://github.com/csheung/clustering-masters-project.git
curl -L -o repo.zip https://github.com/csheung/clustering-masters-project/archive/refs/heads/main.zip
unzip repo.zip
源站 README 摘录（使用方式）
Master’'s Project (Fall 2023)
Uncovering Urban-Rural Disparities in Out-of-Hospital Cardiac Arrest (OHCA) Treatment Strategies through Unsupervised Machine Learning and Causal Analysis on the NEMSIS Dataset
This study looks to utilize clustering in understanding OHCA dataset by identifying potential clusters. It is hoped to gain better understanding toward the OHCA treatment and the urban-rural disparities, and stimulate discussions into causal relationships and possible interventions to improve rural prehospital cardiac arrest care.
This report conducted a retrospective analysis of a nationwide sample of OHCA incidents. Data were extracted from the 2020 National EMS Information System (NEMSIS) Version 3.4.0 Public-Release Research Dataset, administered by the MaineHealth organization and maintained by the Roux Institute at Northeastern University. This dataset is a convenience sample of data from prehospital patient care reports filed by EMS agencies in the United States (US).
The 2020 dataset includes EMS incidents that occurred January 1 to December 31, 2020, submitted from 12,319 EMS agencies located in 50 states and territories. Each record in this dataset corresponds to one EMS unit dispatched to a reported incident. This dataset is de-identified and publicly accessible.
Part 1. Data Pre-processing
The first step is to extract necessary feature data from respective dataframes such as epinephrine used, medications performed, medical protocols or procedures applied.
a. Extract keys of events with epinephrine used
- Open the file of “1_medical_df.ipynb”
b. Run the code blocks through the Jupyter Notebook - Read the csv file “FACTPCRMEDICATION_CA.csv” and turn it into a DataFrame
- Drop unnecessary data based on the “eMedications_03” column
- Extract the event keys with the medical code in [317361, 328316] where 317361 equals each 0.1 MG/ML used and 328316 equals each 1 MG/ML used according to the RxNORM code.
- <img width=“711” alt=“Screenshot 2023-12-21 at 7 25 14 PM” src=“https://github.com/csheung/clustering-masters-project/assets/99443055/d9a881ca-0f1a-44fa-a922-212d34985601”>
- Calculate the Frequency (column “EpinephrineFrequency”) and the Total Amount (column “EpinephrineTotalAmount”) of Epinephrine Used based on the dosage identified with the medical code.
- Keep the DataFrame with the necesaary columns of “EpinephrineUsed”, "Epinephrine
精度瓶颈？数据缺失？
当前公开数据无法满足您的算法精度？千方提供针对 心脏骤停 的高质量、多模态真实临床数据定制解决方案。
