NEW APPROACHES TO MISSING BIOMEDICAL DATA RECOVERY FOR MACHINE LEARNING
Missing data is a common problem for medical data sets, especially large ones. This issue is of major importance since it can influence the analysis and further use of the data, e.g., for machine learning purposes. There are various methods for recovering missing data.One such method is to remove ob...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Technical University of Moldova
2023-03-01
|
| Series: | Journal of Engineering Science (Chişinău) |
| Subjects: | |
| Online Access: | https://press.utm.md/index.php/jes/article/view/2023-30-1-09/09-pdf |
| _version_ | 1797628690667929600 |
|---|---|
| author | IAPĂSCURTĂ, Victor FIODOROV, Ion |
| author_facet | IAPĂSCURTĂ, Victor FIODOROV, Ion |
| author_sort | IAPĂSCURTĂ, Victor |
| collection | DOAJ |
| description | Missing data is a common problem for medical data sets, especially large ones. This issue is of major importance since it can influence the analysis and further use of the data, e.g., for machine learning purposes. There are various methods for recovering missing data.One such method is to remove observations with missing values, but this is not very usefulgiven the limited amount of data available. Another commonly used approach is the LastObservation Carried Forward (LOCF). But most such methods are not universal and may needadjustments to the data set at hand. This article describes the possibility of solving thisproblem in the case of multimodal time series of biomedical data coming from patients withsepsis. It describes and compares three approaches tailored to a sepsis dataset, which isanalyzed and finally used to build a sepsis prediction system based on clinical data routinelyrecorded in an intensive care unit. |
| first_indexed | 2024-03-11T10:42:29Z |
| format | Article |
| id | doaj.art-0b37912cbdf84489a76aec83d3519330 |
| institution | Directory Open Access Journal |
| issn | 2587-3474 2587-3482 |
| language | English |
| last_indexed | 2024-03-11T10:42:29Z |
| publishDate | 2023-03-01 |
| publisher | Technical University of Moldova |
| record_format | Article |
| series | Journal of Engineering Science (Chişinău) |
| spelling | doaj.art-0b37912cbdf84489a76aec83d35193302023-11-14T08:04:49ZengTechnical University of MoldovaJournal of Engineering Science (Chişinău)2587-34742587-34822023-03-01XXX1106117https://doi.org/10.52326/jes.utm.2023.30(1).09NEW APPROACHES TO MISSING BIOMEDICAL DATA RECOVERY FOR MACHINE LEARNINGIAPĂSCURTĂ, Victor0https://orcid.org/0000-0002-4540-7045FIODOROV, Ion1https://orcid.org/0000-0003-0938-3442Technical University of Moldova, 168, Stefan cel Mare si Sfant Blvd., Chisinau, MD – 2004, Republic of Moldova; N. Testemitanu State University of Medicine and Pharmacy, 165, Stefan cel Mare si Sfant Blvd., Chisinau, MD – 2004, Republic of MoldovaTechnical University of Moldova, 168, Stefan cel Mare si Sfant Blvd., Chisinau, MD – 2004, Republic of MoldovaMissing data is a common problem for medical data sets, especially large ones. This issue is of major importance since it can influence the analysis and further use of the data, e.g., for machine learning purposes. There are various methods for recovering missing data.One such method is to remove observations with missing values, but this is not very usefulgiven the limited amount of data available. Another commonly used approach is the LastObservation Carried Forward (LOCF). But most such methods are not universal and may needadjustments to the data set at hand. This article describes the possibility of solving thisproblem in the case of multimodal time series of biomedical data coming from patients withsepsis. It describes and compares three approaches tailored to a sepsis dataset, which isanalyzed and finally used to build a sepsis prediction system based on clinical data routinelyrecorded in an intensive care unit.https://press.utm.md/index.php/jes/article/view/2023-30-1-09/09-pdfmultimodal biomedical time series datamissing valuesdata recoverysepsismachine learning |
| spellingShingle | IAPĂSCURTĂ, Victor FIODOROV, Ion NEW APPROACHES TO MISSING BIOMEDICAL DATA RECOVERY FOR MACHINE LEARNING Journal of Engineering Science (Chişinău) multimodal biomedical time series data missing values data recovery sepsis machine learning |
| title | NEW APPROACHES TO MISSING BIOMEDICAL DATA RECOVERY FOR MACHINE LEARNING |
| title_full | NEW APPROACHES TO MISSING BIOMEDICAL DATA RECOVERY FOR MACHINE LEARNING |
| title_fullStr | NEW APPROACHES TO MISSING BIOMEDICAL DATA RECOVERY FOR MACHINE LEARNING |
| title_full_unstemmed | NEW APPROACHES TO MISSING BIOMEDICAL DATA RECOVERY FOR MACHINE LEARNING |
| title_short | NEW APPROACHES TO MISSING BIOMEDICAL DATA RECOVERY FOR MACHINE LEARNING |
| title_sort | new approaches to missing biomedical data recovery for machine learning |
| topic | multimodal biomedical time series data missing values data recovery sepsis machine learning |
| url | https://press.utm.md/index.php/jes/article/view/2023-30-1-09/09-pdf |
| work_keys_str_mv | AT iapascurtavictor newapproachestomissingbiomedicaldatarecoveryformachinelearning AT fiodorovion newapproachestomissingbiomedicaldatarecoveryformachinelearning |