Featured image of post MIMIC-iv 重症医学数据库部署与使用

MIMIC-iv 重症医学数据库部署与使用

部署与使用最新的MIMIC-iv重症监护数据库

简介

医疗数据库不仅是我们进行医疗大数据分析的主要来源，同时也是科学研究的重要基础，可以用来改善患者的治疗和诊断，帮助改善医疗服务的质量。它也可以支持临床研究和政策制定，帮助改善健康管理模式，提高治疗效果，减少治疗成本。

MIMIC（Medical Information Mart for Intensive Care）是一个由_麻省理工大学_联合_贝斯以色列女执事医疗中心_等共同建立的重症监护数据库，它是目前急诊和危重症领域最大、可以公开访问的临床数据库。MIMIC最新版本MIMIC-IV发布于2020年，其中最新的数据发布于2023年2月6日。

MIMIC-iv提供了一个多维度的数据集，可以支持重症医学领域的研究，它包含来自美国麻省总医院的多种类型的重症患者的数据。

尽管MIMIC是公开数据集，但是该数据集的下载需要通过考核与人工审核。

bookmark

内容

截至本文发布时的最新版本（v2.2）时，MIMIC IV总共公布了HOSP、ICU、ED、CXR、NOTE五大模块的数据。

（目前还有waveforms待发布

部署

本文将以Windows为例，将MIMIC-iv导入PostgreSQL数据库中。

下载安装PostgreSQL

bookmark

下载数据

略，解压到F:\mimiciv

其中.sql.gz无需解压

Clone导入代码

1
2
3
4
5

git clone https://github.com/MIT-LCP/mimic-code.git
cd mimic-code

# 也可以使用QGit 镜像
# git clone https://git.qmcmc.cn/mirror/mimic-code

HOSP+ICU模块

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12

DROP DATABASE IF EXISTS mimiciv;
CREATE DATABASE mimiciv OWNER postgres;

\c mimiciv;
\encoding 'UTF8'
\i F:/mimic-code/mimic-iv/buildmimic/postgres/create.sql;   -- 创建表
\set ON_ERROR_STOP 1
\set mimic_data_dir 'F:/mimiciv'  --定义数据目录
\i F:/mimic-code/mimic-iv/buildmimic/postgres/load_7z.sql;   -- 导入数据
\i F:/mimic-code/mimic-iv/buildmimic/postgres/index.sql;   -- 导入索引
\i F:/mimic-code/mimic-iv/buildmimic/postgres/constraint.sql;   -- 导入约束
\i F:/mimic-code/mimic-iv/buildmimic/postgres/validate.sql;   -- 验证导入

ED模块

基本同理

1
2
3
4
5
6
7
8

\c mimiciv;
\encoding 'UTF8'
\i F:/mimic-code/mimic-iv-ed/buildmimic/postgres/create.sql;   -- 创建表
\set ON_ERROR_STOP 1
\set mimic_data_dir 'F:/mimiciv/ed'  --定义数据目录
\i F:/mimic-code/mimic-iv-ed/buildmimic/postgres/load_7z.sql;   -- 导入数据
\i F:/mimic-code/mimic-iv-ed/buildmimic/postgres/index.sql;   -- 导入索引
\i F:/mimic-code/mimic-iv-ed/buildmimic/postgres/constraint.sql;   -- 导入约束

Note模块

1
2
3
4
5
6

\c mimiciv;
\encoding 'UTF8'
\i F:/mimic-code/mimic-iv-note/buildmimic/postgres/create.sql;   -- 创建表
\set ON_ERROR_STOP 1
\set mimic_data_dir 'F:/mimiciv/note'  --定义数据目录
\i F:/mimic-code/mimic-iv-note/buildmimic/postgres/load_7z.sql;   -- 导入数据

CXR模块

官方代码库中似乎没有导入SQL语句，我们可以手动创建

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91

-- create.sql
-------------------------------------------
-- Create the tables and MIMIC-IV schema --
-------------------------------------------

----------------------
-- Creating schemas --
----------------------

\echo '######################'
\echo 'Creating schemas.....'

DROP SCHEMA IF EXISTS mimiciv_cxr CASCADE;
CREATE SCHEMA mimiciv_cxr;

---------------------
-- Creating tables --
---------------------

\echo 'Creating mimiciv_cxr tables.....'

DROP TABLE IF EXISTS mimiciv_cxr.chexpert;
CREATE TABLE mimiciv_cxr.chexpert 
(
  "subject_id" int8,
  "study_id" int8,
  "Atelectasis" varchar(255),
  "Cardiomegaly" varchar(255),
  "Consolidation" varchar(255),
  "Edema" varchar(255),
  "Enlarged Cardiomediastinum" varchar(255),
  "Fracture" varchar(255),
  "Lung Lesion" varchar(255),
  "Lung Opacity" varchar(255),
  "No Finding" varchar(255),
  "Pleural Effusion" varchar(255),
  "Pleural Other" varchar(255),
  "Pneumonia" varchar(255),
  "Pneumothorax" varchar(255),
  "Support Devices" varchar(255)
);


DROP TABLE IF EXISTS mimiciv_cxr.metadata;
CREATE TABLE mimiciv_cxr.metadata 
(
  "dicom_id" varchar(255),
  "subject_id" int8,
  "study_id" int8,
  "PerformedProcedureStepDescription" varchar(255),
  "ViewPosition" varchar(255),
  "Rows" varchar(255),
  "Columns" varchar(255),
  "StudyDate" varchar(255),
  "StudyTime" varchar(255),
  "ProcedureCodeSequence_CodeMeaning" varchar(255),
  "ViewCodeSequence_CodeMeaning" varchar(255),
  "PatientOrientationCodeSequence_CodeMeaning" varchar(255)
);


DROP TABLE IF EXISTS mimiciv_cxr.negbio;
CREATE TABLE mimiciv_cxr.negbio 
(
  "subject_id" int8,
  "study_id" int8,
  "Atelectasis" varchar(255),
  "Cardiomegaly" varchar(255),
  "Consolidation" varchar(255),
  "Edema" varchar(255),
  "Enlarged Cardiomediastinum" varchar(255),
  "Fracture" varchar(255),
  "Lung Lesion" varchar(255),
  "Lung Opacity" varchar(255),
  "No Finding" varchar(255),
  "Pleural Effusion" varchar(255),
  "Pleural Other" varchar(255),
  "Pneumonia" varchar(255),
  "Pneumothorax" varchar(255),
  "Support Devices" varchar(255)
);


DROP TABLE IF EXISTS mimiciv_cxr.split;
CREATE TABLE mimiciv_cxr.split 
(
  "dicom_id" varchar(255),
  "study_id" int8,
  "subject_id" int8,
  "split" varchar(255)
);

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36

--load.sql
-----------------------------------------
-- Load data into the MIMIC-IV schemas --
-----------------------------------------

-- To run from a terminal:
--  psql "dbname=<DBNAME> user=<USER>" -v mimic_data_dir=<PATH TO DATA DIR> -f load_gz.sql


-- hosp schema
\cd :mimic_data_dir


-- making sure that all tables are emtpy and correct encoding is defined -utf8- 
SET CLIENT_ENCODING TO 'utf8';

\echo '######################'

\echo 'Delete mimiciv_cxr table.....'

DELETE FROM mimiciv_cxr.split;
DELETE FROM mimiciv_cxr.negbio;
DELETE FROM mimiciv_cxr.metadata;
DELETE FROM mimiciv_cxr.chexpert;

\echo '######################'
\echo 'Copying mimiciv_cxr table.....'
\echo 'Copying split.....'


\COPY mimiciv_cxr.split FROM 'split.csv' DELIMITER ',' CSV HEADER NULL '';
\COPY mimiciv_cxr.negbio FROM 'negbio.csv' DELIMITER ',' CSV HEADER NULL '';
\COPY mimiciv_cxr.metadata FROM 'metadata.csv' DELIMITER ',' CSV HEADER NULL '';
\COPY mimiciv_cxr.chexpert FROM 'chexpert.csv' DELIMITER ',' CSV HEADER NULL '';

\echo 'mimiciv_cxr successfully generated.'

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46

--index.sql
--------------------------------------
--------------------------------------
-- Indexes for all MIMIC-IV modules --
--------------------------------------
--------------------------------------

---------
-- cxr --
---------

SET search_path TO mimiciv_cxr;

\echo 'Create mimiciv cxr index.....'

DROP INDEX IF EXISTS chexpert_idx01;
CREATE INDEX chexpert_idx01
  ON chexpert (subject_id);

DROP INDEX IF EXISTS chexpert_idx02;
CREATE INDEX chexpert_idx02
  ON chexpert (study_id);
  
DROP INDEX IF EXISTS metadata_idx01;
CREATE INDEX metadata_idx01
  ON metadata (subject_id);

DROP INDEX IF EXISTS metadata_idx02;
CREATE INDEX metadata_idx02
  ON metadata (study_id);
  
DROP INDEX IF EXISTS negbio_idx01;
CREATE INDEX negbio_idx01
  ON negbio (subject_id);

DROP INDEX IF EXISTS negbio_idx02;
CREATE INDEX negbio_idx02
  ON negbio (study_id);
  
DROP INDEX IF EXISTS split_idx01;
CREATE INDEX split_idx01
  ON split (subject_id);

DROP INDEX IF EXISTS split_idx02;
CREATE INDEX split_idx02
  ON split (study_id);

然后在控制台

1
2
3
4
5
6
7

\c mimiciv;
\encoding 'UTF8'
\i F:/mimic/cxr/create.sql;   -- 创建表
\set ON_ERROR_STOP 1
\set mimic_data_dir 'F:/mimic/cxr'  --定义数据目录
\i F:/mimic/cxr/load.sql;  -- 导入数据
\i F:/mimic/cxr/index.sql;

导入结果

使用数据

生物医学
署名-相同方式共享 4.0 国际 (CC BY-SA 4.0)
最后更新于 2023-02-12 17:29 CST