Machine Learning-Based Pressure Ulcer Prediction in Modular
Critical Care Data

J 2022

Machine Learning-Based Pressure Ulcer Prediction in Modular Critical Care Data

ŠÍN, Petr, Alica HOKYNKOVÁ, Marie NOVÁKOVÁ, Andrea POKORNÁ, Rostislav KRČ et. al.

Basic information

Original name

Machine Learning-Based Pressure Ulcer Prediction in Modular Critical Care Data

Authors

ŠÍN, Petr (203 Czech Republic, belonging to the institution), Alica HOKYNKOVÁ (703 Slovakia, guarantor, belonging to the institution), Marie NOVÁKOVÁ (203 Czech Republic, belonging to the institution), Andrea POKORNÁ (203 Czech Republic, belonging to the institution), Rostislav KRČ (203 Czech Republic) and Jan PODROUŽEK (203 Czech Republic)

Edition

Diagnostics, Basel, MDPI, 2022, 2075-4418

Other information

Language

English

Type of outcome

Článek v odborném periodiku

Field of Study

30218 General and internal medicine

Country of publisher

Switzerland

Confidentiality degree

není předmětem státního či obchodního tajemství

References:

URL

Impact factor

Impact factor: 3.600

RIV identification code

RIV/00216224:14110/22:00129663

Organization unit

Faculty of Medicine

DOI

http://dx.doi.org/10.3390/diagnostics12040850

UT WoS

000785475800001

Keywords in English

pressure ulcer; pressure injury; machine learning; MIMIC database; MIMIC-IV; open data; artificial neural network; random forest

Abstract

V originále

Increasingly available open medical and health datasets encourage data-driven research with a promise of improving patient care through knowledge discovery and algorithm development. Among efficient approaches to such high-dimensional problems are a number of machine learning methods, which are applied in this paper to pressure ulcer prediction in modular critical care data. An inherent property of many health-related datasets is a high number of irregularly sampled time-variant and scarcely populated features, often exceeding the number of observations. Although machine learning methods are known to work well under such circumstances, many choices regarding model and data processing exist. In particular, this paper address both theoretical and practical aspects related to the application of six classification models to pressure ulcers, while utilizing one of the largest available Medical Information Mart for Intensive Care (MIMIC-IV) databases. Random forest, with an accuracy of 96%, is the best-performing approach among the considered machine learning algorithms.

Links

NU21-09-00541, research and development project

Name: Role oxidativního stresu při hojení dekubitů u pacientů s míšní lézí

Investor: Ministry of Health of the CR, Subprogram 1 - standard

Citovat

ŠÍN, Petr, Alica HOKYNKOVÁ, Marie NOVÁKOVÁ, Andrea POKORNÁ, Rostislav KRČ and Jan PODROUŽEK. Machine Learning-Based Pressure Ulcer Prediction in Modular Critical Care Data. Diagnostics. Basel: MDPI, 2022, vol. 12, No 4, p. 1-13. ISSN 2075-4418. Available from: https://dx.doi.org/10.3390/diagnostics12040850.

@article{1845958,
   author = {Šín, Petr and Hokynková, Alica and Nováková, Marie and Pokorná, Andrea and Krč, Rostislav and Podroužek, Jan},
   article_location = {Basel},
   article_number = {4},
   doi = {http://dx.doi.org/10.3390/diagnostics12040850},
   keywords = {pressure ulcer; pressure injury; machine learning; MIMIC database; MIMIC-IV; open data; artificial neural network; random forest},
   language = {eng},
   issn = {2075-4418},
   journal = {Diagnostics},
   title = {Machine Learning-Based Pressure Ulcer Prediction in Modular Critical Care Data},
   url = {https://doi.org/10.3390/diagnostics12040850},
   volume = {12},
   year = {2022}
}

TY  - JOUR
ID  - 1845958
AU  - Šín, Petr - Hokynková, Alica - Nováková, Marie - Pokorná, Andrea - Krč, Rostislav - Podroužek, Jan
PY  - 2022
TI  - Machine Learning-Based Pressure Ulcer Prediction in Modular Critical Care Data
JF  - Diagnostics
VL  - 12
IS  - 4
SP  - 1-13
EP  - 1-13
PB  - MDPI
SN  - 20754418
KW  - pressure ulcer
KW  - pressure injury
KW  - machine learning
KW  - MIMIC database
KW  - MIMIC-IV
KW  - open data
KW  - artificial neural network
KW  - random forest
UR  - https://doi.org/10.3390/diagnostics12040850
N2  - Increasingly available open medical and health datasets encourage data-driven research with a promise of improving patient care through knowledge discovery and algorithm development. Among efficient approaches to such high-dimensional problems are a number of machine learning methods, which are applied in this paper to pressure ulcer prediction in modular critical care data. An inherent property of many health-related datasets is a high number of irregularly sampled time-variant and scarcely populated features, often exceeding the number of observations. Although machine learning methods are known to work well under such circumstances, many choices regarding model and data processing exist. In particular, this paper address both theoretical and practical aspects related to the application of six classification models to pressure ulcers, while utilizing one of the largest available Medical Information Mart for Intensive Care (MIMIC-IV) databases. Random forest, with an accuracy of 96%, is the best-performing approach among the considered machine learning algorithms.
ER  -

ŠÍN, Petr, Alica HOKYNKOVÁ, Marie NOVÁKOVÁ, Andrea POKORNÁ, Rostislav KRČ and Jan PODROUŽEK. Machine Learning-Based Pressure Ulcer Prediction in Modular Critical Care Data. \textit{Diagnostics}. Basel: MDPI, 2022, vol.~12, No~4, p.~1-13. ISSN~2075-4418. Available from: https://dx.doi.org/10.3390/diagnostics12040850.

Detailed Information on Publication Record