Myocardial infarction complications

Prediction of myocardial infarction complications

Characteristics
Multivariate
Subject Area
Health and Medicine
Associated Tasks
Classification
Attribute Type
--
# Instances
1.7k
# Attributes
111

Info

Problems of real-life complexity are needed to test and compare various data mining and pattern recognition methods. The proposed database can be used to solve two practically important problems: predicting complications of Myocardial Infarction (MI) based on information about the patient (i) at the time of admission and (ii) on the third day of the hospital period. Another important group of tasks is phenotyping of disease (cluster analysis), dynamic phenotyping (filament extraction and identification of disease trajectories) and visualisation (disease mapping).
MI is one of the most challenging problems of modern medicine. Acute myocardial infarction is associated with high mortality in the first year after it. The incidence of MI remains high in all countries. This is especially true for the urban population of highly developed countries, which is exposed to chronic stress factors, irregular and not always balanced nutrition. In the United States, for example, more than a million people suffer from MI every year, and 200-300 thousand of them die from acute MI before arriving at the hospital.
The course of the disease in patients with MI is different. MI can occur without complications or with complications that do not worsen the long-term prognosis. At the same time, about half of patients in the acute and subacute periods have complications that lead to worsening of the disease and even death. Even an experienced specialist can not always foresee the development of these complications. In this regard, predicting complications of myocardial infarction in order to timely carry out the necessary preventive measures is an important task.

Problems to solve
In general columns 2-112 can be used as input data for prediction. Possible complications (outputs) are listed in columns 113-124.
There are four possible time moments for complication prediction: on base of the information known at
1. the time of admission to hospital: all input columns (2-112) except 93, 94, 95, 100, 101, 102, 103, 104, 105 can be used for prediction;
2. the end of the first day (24 hours after admission to the hospital): all input columns (2-112) except 94, 95, 101, 102, 104, 105 can be used for prediction;
3. the end of the second day (48 hours after admission to the hospital) all input columns (2-112) except 95, 102, 105 can be used for prediction;
4. the end of the third day (72 hours after admission to the hospital) all input columns (2-112) can be used for prediction.

You can find detailed description of database, descriptive statistics and csv version of database in DOI: 10.25392/leicester.data.12045261.v3

Introductory Paper

Trajectories, bifurcations, and pseudo-time in large clinical datasets: applications to myocardial infarction and diabetes data

By S. E. Golovenkin, Jonathan Bac, A. Chervov, E. M. Mirkes, Y. Orlova, E. Barillot, A. Gorban, A. Zinovyev. 2020

Published in GigaScience
More

Provided by
University of California, Irvine

Creators
  • S.E. Golovenkin
  • V.A. Shulman
  • D.A. Rossiev
  • P.A. Shesternya
  • S.Yu. Nikulina
  • Yu.V. Orlova
  • V.F. Voino-Yasenetsky
DOI

10.24432/C53P5M

Login to Download

New to AIM-AHEAD Connect?
Create an account!

Features

Attribute Name Role Type Demographic Description Units Missing Values
ID ID Integer Record ID (ID): Unique identifier. Cannot be related to participant. It can be used for reference only. no
AGE Feature Integer Age Age of patient. no
SEX Feature Binary Sex 0: female, 1: male no
INF_ANAM Feature Categorical Quantity of myocardial infarctions in the anamnesis. 0: zero 1: one 2: two 3: three and more yes
STENOK_AN Feature Categorical Exertional angina pectoris in the anamnesis. 0: never 1: during the last year 2: one year ago 3: two years ago 4: three years ago 5: 4-5 years ago yes
FK_STENOK Feature Categorical Functional class (FC) of angina pectoris in the last year. 0: there is no angina pectoris 1: I FC 2: II FC 3: III FC 4: IV FC yes
IBS_POST Feature Categorical Coronary heart disease (CHD) in recent weeks, days before admission to hospital 0: none 1: exertional angina pectoris 2: unstable angina pectoris yes
IBS_NASL Feature Binary Heredity on CHD 0: isn't burdened 1: burdened yes
GB Feature Categorical Presence of an essential hypertension 0: there is no essential hypertension 1: Stage 1 2: Stage 2 3: Stage 3 yes
SIM_GIPERT Feature Binary Symptomatic hypertension yes
DLIT_AG Feature Categorical there was no arterial hypertension 1: one year 2: two years 3: three years 4: four years 5: five years 6: 6-10 years 7: more than 10 years yes
ZSN_A Feature Categorical Presence of chronic Heart failure (HF) in the anamnesis: Partially ordered attribute: there are two lines of severities: 0<1<2<4, 0<1<3<4. State 4 means simultaneous states 2 and 3 0: there is no chronic heart failure 1: I stage 2: II stage (heart failure due to right ventricular systolic dysfunction) 3: II stage (heart failure due to left ventricular systolic dysfunction) 4: IIB stage (heart failure due to left and right ventricular systolic dysfunction) yes
nr_11 Feature Binary Observing of arrhythmia in the anamnesis yes
nr_01 Feature Binary Premature atrial contractions in the anamnesis yes
nr_02 Feature Binary Premature ventricular contractions in the anamnesis yes
nr_03 Feature Binary Paroxysms of atrial fibrillation in the anamnesis yes
nr_04 Feature Binary A persistent form of atrial fibrillation in the anamnesis yes
nr_07 Feature Binary Ventricular fibrillation in the anamnesis yes
nr_08 Feature Binary Ventricular paroxysmal tachycardia in the anamnesis yes
np_01 Feature Binary First-degree AV block in the anamnesis yes
np_04 Feature Binary Third-degree AV block in the anamnesis yes
np_05 Feature Binary LBBB (anterior branch) in the anamnesis yes
np_07 Feature Binary Incomplete LBBB in the anamnesis yes
np_08 Feature Binary Complete LBBB in the anamnesis yes
np_09 Feature Binary Incomplete RBBB in the anamnesis yes
np_10 Feature Binary Complete RBBB in the anamnesis yes
endocr_01 Feature Binary Diabetes mellitus in the anamnesis yes
endocr_02 Feature Binary Obesity in the anamnesis yes
endocr_03 Feature Binary Thyrotoxicosis in the anamnesis yes
zab_leg_01 Feature Binary Chronic bronchitis in the anamnesis yes
zab_leg_02 Feature Binary Obstructive chronic bronchitis in the anamnesis yes
zab_leg_03 Feature Binary Bronchial asthma in the anamnesis yes
zab_leg_04 Feature Binary Chronic pneumonia in the anamnesis yes
zab_leg_06 Feature Binary Pulmonary tuberculosis in the anamnesis yes
S_AD_KBRIG Feature Integer Systolic blood pressure according to Emergency Cardiology Team mmHg yes
D_AD_KBRIG Feature Integer Diastolic blood pressure according to Emergency Cardiology Team mmHg yes
S_AD_ORIT Feature Integer Systolic blood pressure according to intensive care unit mmHg yes
D_AD_ORIT Feature Integer Diastolic blood pressure according to intensive care unit mmHg yes
O_L_POST Feature Binary Pulmonary edema at the time of admission to intensive care unit yes
K_SH_POST Feature Binary Cardiogenic shock at the time of admission to intensive care unit yes
MP_TP_POST Feature Binary Paroxysms of atrial fibrillation at the time of admission to intensive care unit, (or at a pre-hospital stage) yes
SVT_POST Feature Binary Paroxysms of supraventricular tachycardia at the time of admission to intensive care unit, (or at a pre-hospital stage) yes
GT_POST Feature Binary Paroxysms of ventricular tachycardia at the time of admission to intensive care unit, (or at a pre-hospital stage) yes
FIB_G_POST Feature Binary Ventricular fibrillation at the time of admission to intensive care unit, (or at a pre-hospital stage) yes
ant_im Feature Categorical Presence of an anterior myocardial infarction (left ventricular) (ECG changes in leads V1: V4 ) 0: there is no infarct in this location 1: QRS has no changes 2: QRS is like QR-complex 3: QRS is like Qr-complex 4: QRS is like QS-complex yes
lat_im Feature Categorical Presence of a lateral myocardial infarction (left ventricular) (ECG changes in leads V5: V6 , I, AVL) 0: there is no infarct in this location 1: QRS has no changes 2: QRS is like QR-complex 3: QRS is like Qr-complex 4: QRS is like QS-complex yes
inf_im Feature Categorical Presence of an inferior myocardial infarction (left ventricular) (ECG changes in leads III, AVF, II). 0: there is no infarct in this location 1: QRS has no changes 2: QRS is like QR-complex 3: QRS is like Qr-complex 4: QRS is like QS-complex yes
post_im Feature Categorical Presence of a posterior myocardial infarction (left ventricular) (ECG changes in V7: V9, reciprocity changes in leads V1 – V3) 0: there is no infarct in this location 1: QRS has no changes 2: QRS is like QR-complex 3: QRS is like Qr-complex 4: QRS is like QS-complex yes
IM_PG_P Feature Binary Presence of a right ventricular myocardial infarction yes
ritm_ecg_p_01 Feature Binary ECG rhythm at the time of admission to hospital: sinus (with a heart rate 60-90) yes
ritm_ecg_p_02 Feature Binary ECG rhythm at the time of admission to hospital: atrial fibrillation yes
ritm_ecg_p_04 Feature Binary ECG rhythm at the time of admission to hospital: atrial yes
ritm_ecg_p_06 Feature Binary ECG rhythm at the time of admission to hospital: idioventricular yes
ritm_ecg_p_07 Feature Binary ECG rhythm at the time of admission to hospital: sinus with a heart rate above 90 (tachycardia) yes
ritm_ecg_p_08 Feature Binary ECG rhythm at the time of admission to hospital: sinus with a heart rate below 60 (bradycardia) yes
n_r_ecg_p_01 Feature Binary Premature atrial contractions on ECG at the time of admission to hospital yes
n_r_ecg_p_02 Feature Binary Frequent premature atrial contractions on ECG at the time of admission to hospital yes
n_r_ecg_p_03 Feature Binary Premature ventricular contractions on ECG at the time of admission to hospital yes
n_r_ecg_p_04 Feature Binary Frequent premature ventricular contractions on ECG at the time of admission to hospital yes
n_r_ecg_p_05 Feature Binary Paroxysms of atrial fibrillation on ECG at the time of admission to hospital yes
n_r_ecg_p_06 Feature Binary Persistent form of atrial fibrillation on ECG at the time of admission to hospital yes
n_r_ecg_p_08 Feature Binary Paroxysms of supraventricular tachycardia on ECG at the time of admission to hospital yes
n_r_ecg_p_09 Feature Binary Paroxysms of ventricular tachycardia on ECG at the time of admission to hospital yes
n_r_ecg_p_10 Feature Binary Ventricular fibrillation on ECG at the time of admission to hospital yes
n_p_ecg_p_01 Feature Binary Sinoatrial block on ECG at the time of admission to hospital yes
n_p_ecg_p_03 Feature Binary First-degree AV block on ECG at the time of admission to hospital yes
n_p_ecg_p_04 Feature Binary Type 1 Second-degree AV block (Mobitz I/Wenckebach) on ECG at the time of admission to hospital yes
n_p_ecg_p_05 Feature Binary Type 2 Second-degree AV block (Mobitz II/Hay) on ECG at the time of admission to hospital yes
n_p_ecg_p_06 Feature Binary Third-degree AV block on ECG at the time of admission to hospital yes
n_p_ecg_p_07 Feature Binary LBBB (anterior branch) on ECG at the time of admission to hospital yes
n_p_ecg_p_08 Feature Binary LBBB (posterior branch) on ECG at the time of admission to hospital yes
n_p_ecg_p_09 Feature Binary Incomplete LBBB on ECG at the time of admission to hospital yes
n_p_ecg_p_10 Feature Binary Complete LBBB on ECG at the time of admission to hospital yes
n_p_ecg_p_11 Feature Binary Incomplete RBBB on ECG at the time of admission to hospital yes
n_p_ecg_p_12 Feature Binary Complete RBBB on ECG at the time of admission to hospital yes
fibr_ter_01 Feature Binary Fibrinolytic therapy by Сеliasum 750k yes
fibr_ter_02 Feature Binary Fibrinolytic therapy by Сеliasum 1m IU yes
fibr_ter_03 Feature Binary Fibrinolytic therapy by Сеliasum 3m IU yes
fibr_ter_05 Feature Binary Fibrinolytic therapy by Streptase yes
fibr_ter_06 Feature Binary Fibrinolytic therapy by Сеliasum 500k yes
fibr_ter_07 Feature Binary Fibrinolytic therapy by Сеliasum 250k yes
fibr_ter_08 Feature Binary Fibrinolytic therapy by Сеliasum 1.5m IU yes
GIPO_K Feature Binary Hypokalemia ( < 4 mmol/L) yes
K_BLOOD Feature Continuous Serum potassium content mmol/L yes
GIPER_NA Feature Binary Increase of sodium in serum (more than 150 mmol/L) yes
NA_BLOOD Feature Continuous Serum sodium content mmol/L yes
ALT_BLOOD Feature Continuous Serum AlAT content (ALT_BLOOD) IU/L yes
AST_BLOOD Feature Continuous Serum AsAT content IU/L yes
KFK_BLOOD Feature Continuous Serum CPK content IU/L yes
L_BLOOD Feature Continuous White blood cell count billions per liter yes
ROE Feature Continuous ESR (Erythrocyte sedimentation rate) мм yes
TIME_B_S Feature Categorical Time elapsed from the beginning of the attack of CHD to the hospital 1: less than 2 hours 2: 2-4 hours 3: 4-6 hours 4: 6-8 hours 5: 8-12 hours 6: 12-24 hours 7: more than 1 days 8: more than 2 days 9: more than 3 days yes
R_AB_1_n Feature Categorical Relapse of the pain in the first hours of the hospital period 0: there is no relapse 1: only one 2: 2 times 3: 3 or more times yes
R_AB_2_n Feature Categorical Relapse of the pain in the second day of the hospital period 0: there is no relapse 1: only one 2: 2 times 3: 3 or more times yes
R_AB_3_n Feature Categorical Relapse of the pain in the third day of the hospital period 0: there is no relapse 1: only one 2: 2 times 3: 3 or more times yes
NA_KB Feature Binary Use of opioid drugs by the Emergency Cardiology Team yes
NOT_NA_KB Feature Binary Use of NSAIDs by the Emergency Cardiology Team yes
LID_KB Feature Binary Use of lidocaine by the Emergency Cardiology Team yes
NITR_S Feature Binary Use of liquid nitrates in the ICU yes
NA_R_1_n Feature Integer Use of opioid drugs in the ICU in the first hours of the hospital period yes
NA_R_2_n Feature Integer Use of opioid drugs in the ICU in the second day of the hospital period yes
NA_R_3_n Feature Integer Use of opioid drugs in the ICU in the third day of the hospital period yes
NOT_NA_1_n Feature Categorical Use of NSAIDs in the ICU in the first hours of the hospital period 0: no 1: once 2: twice 3: three times 4: four or more times yes
NOT_NA_2_n Feature Integer Use of NSAIDs in the ICU in the second day of the hospital period yes
NOT_NA_3_n Feature Integer Use of NSAIDs in the ICU in the third day of the hospital period yes
LID_S_n Feature Binary Use of lidocaine in the ICU yes
B_BLOK_S_n Feature Binary Use of beta-blockers in the ICU yes
ANT_CA_S_n Feature Binary Use of calcium channel blockers in the ICU yes
GEPAR_S_n Feature Binary Use of а anticoagulants (heparin) in the ICU yes
ASP_S_n Feature Binary Use of acetylsalicylic acid in the ICU yes
TIKL_S_n Feature Binary Use of Ticlid in the ICU yes
TRENT_S_n Feature Binary Use of Trental in the ICU yes
FIBR_PREDS Target Binary Atrial fibrillation no
PREDS_TAH Target Binary Supraventricular tachycardia no
JELUD_TAH Target Binary Ventricular tachycardia no
FIBR_JELUD Target Binary Ventricular fibrillation no
A_V_BLOK Target Binary Third-degree AV block no
OTEK_LANC Target Binary Pulmonary edema no
RAZRIV Target Binary Myocardial rupture no
DRESSLER Target Binary Dressler syndrome no
ZSN Target Binary Chronic heart failure no
REC_IM Target Binary Relapse of the myocardial infarction no
P_IM_STEN Target Binary Post-infarction angina no
LET_IS Target Categorical Lethal outcome (cause) 0: unknown (alive) 1: cardiogenic shock 2: pulmonary edema 3: myocardial rupture 4: progress of congestive heart failure 5: thromboembolism 6: asystole 7: ventricular fibrillation no