Conţinutul numărului revistei |
Articolul precedent |
Articolul urmator |
331 6 |
Ultima descărcare din IBN: 2024-05-13 17:33 |
Căutarea după subiecte similare conform CZU |
004:[612.13:616.94] (1) |
Știința și tehnologia calculatoarelor. Calculatoare. Procesarea datelor (4280) |
Fiziologie. Fiziologie umană și comparată (737) |
Boli transmisibile. Boli infecţionase şi contagioase, stări febrile (587) |
SM ISO690:2012 IAPĂSCURTĂ, Victor, FIODOROV, Ion. New approaches to missing biomedical data recovery for machine learning. In: Journal of Engineering Sciences, 2023, vol. 30, nr. 1, pp. 106-117. ISSN 2587-3474. DOI: https://doi.org/10.52326/jes.utm.2023.30(1).09 |
EXPORT metadate: Google Scholar Crossref CERIF DataCite Dublin Core |
Journal of Engineering Sciences | ||||||
Volumul 30, Numărul 1 / 2023 / ISSN 2587-3474 /ISSNe 2587-3482 | ||||||
|
||||||
DOI:https://doi.org/10.52326/jes.utm.2023.30(1).09 | ||||||
CZU: 004:[612.13:616.94] | ||||||
Pag. 106-117 | ||||||
|
||||||
Descarcă PDF | ||||||
Rezumat | ||||||
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. |
||||||
Cuvinte-cheie multimodal biomedical time series data, missing values, data recovery, sepsis, machine learning, serii temporale de date biomedicale multimodale, valori lipsă, recuperare date, sepsis, învăţare automată |
||||||
|
Cerif XML Export
<?xml version='1.0' encoding='utf-8'?> <CERIF xmlns='urn:xmlns:org:eurocris:cerif-1.5-1' xsi:schemaLocation='urn:xmlns:org:eurocris:cerif-1.5-1 http://www.eurocris.org/Uploads/Web%20pages/CERIF-1.5/CERIF_1.5_1.xsd' xmlns:xsi='http://www.w3.org/2001/XMLSchema-instance' release='1.5' date='2012-10-07' sourceDatabase='Output Profile'> <cfResPubl> <cfResPublId>ibn-ResPubl-178860</cfResPublId> <cfResPublDate>2023-04-16</cfResPublDate> <cfIssue>1</cfIssue> <cfStartPage>106</cfStartPage> <cfISSN>2587-3474</cfISSN> <cfURI>https://ibn.idsi.md/ro/vizualizare_articol/178860</cfURI> <cfTitle cfLangCode='EN' cfTrans='o'>New approaches to missing biomedical data recovery for machine learning</cfTitle> <cfKeyw cfLangCode='EN' cfTrans='o'>multimodal biomedical time series data; missing values; data recovery; sepsis; machine learning; serii temporale de date biomedicale multimodale; valori lipsă; recuperare date; sepsis; învăţare automată</cfKeyw> <cfAbstr cfLangCode='EN' cfTrans='o'><p>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.</p></cfAbstr> <cfAbstr cfLangCode='RO' cfTrans='o'><p>Datele lipsă sunt o problemă comună pentru seturile de date medicale, în special pentru cele mari. Această problemă este de o importanță majoră, deoarece poate influența analiza și utilizarea ulterioară a datelor, de exemplu, în scopuri de învățare automată. Există abordări diferite pentru a trata datele lipsă. Una obișnuită este ștergerea observațiilor care conțin astfel de date, însă ea nu este aplicabilă atunci când volumul datelor este limitat. O altă abordare frecvent utilizată este “Last Observation Carried Forward (LOCF)”. Dar majoritatea acestor metode nu sunt universale și pot necesita ajustări la setul de date la îndemână. Această lucrare descrie posibilitatea abordării acestei probleme în cazul seriilor temporale multimodale de date biomedicale provenite de la pacienții cu sepsis. Ea descrie și compară trei abordări adaptate setului de date care este analizat și utilizat în cele din urmă pentru construirea unui sistem de predicție a sepsisului bazat pe date clinice înregistrate în mod obișnuit într-o unitate de terapie intensivă.</p></cfAbstr> <cfResPubl_Class> <cfClassId>eda2d9e9-34c5-11e1-b86c-0800200c9a66</cfClassId> <cfClassSchemeId>759af938-34ae-11e1-b86c-0800200c9a66</cfClassSchemeId> <cfStartDate>2023-04-16T24:00:00</cfStartDate> </cfResPubl_Class> <cfResPubl_Class> <cfClassId>e601872f-4b7e-4d88-929f-7df027b226c9</cfClassId> <cfClassSchemeId>40e90e2f-446d-460a-98e5-5dce57550c48</cfClassSchemeId> <cfStartDate>2023-04-16T24:00:00</cfStartDate> </cfResPubl_Class> <cfPers_ResPubl> <cfPersId>ibn-person-30090</cfPersId> <cfClassId>49815870-1cfe-11e1-8bc2-0800200c9a66</cfClassId> <cfClassSchemeId>b7135ad0-1d00-11e1-8bc2-0800200c9a66</cfClassSchemeId> <cfStartDate>2023-04-16T24:00:00</cfStartDate> </cfPers_ResPubl> <cfPers_ResPubl> <cfPersId>ibn-person-12933</cfPersId> <cfClassId>49815870-1cfe-11e1-8bc2-0800200c9a66</cfClassId> <cfClassSchemeId>b7135ad0-1d00-11e1-8bc2-0800200c9a66</cfClassSchemeId> <cfStartDate>2023-04-16T24:00:00</cfStartDate> </cfPers_ResPubl> <cfFedId> <cfFedIdId>ibn-doi-178860</cfFedIdId> <cfFedId>10.52326/jes.utm.2023.30(1).09</cfFedId> <cfStartDate>2023-04-16T24:00:00</cfStartDate> <cfFedId_Class> <cfClassId>31d222b4-11e0-434b-b5ae-088119c51189</cfClassId> <cfClassSchemeId>bccb3266-689d-4740-a039-c96594b4d916</cfClassSchemeId> </cfFedId_Class> <cfFedId_Srv> <cfSrvId>5123451</cfSrvId> <cfClassId>eda2b2e2-34c5-11e1-b86c-0800200c9a66</cfClassId> <cfClassSchemeId>5a270628-f593-4ff4-a44a-95660c76e182</cfClassSchemeId> </cfFedId_Srv> </cfFedId> </cfResPubl> <cfPers> <cfPersId>ibn-Pers-30090</cfPersId> <cfPersName_Pers> <cfPersNameId>ibn-PersName-30090-3</cfPersNameId> <cfClassId>55f90543-d631-42eb-8d47-d8d9266cbb26</cfClassId> <cfClassSchemeId>7375609d-cfa6-45ce-a803-75de69abe21f</cfClassSchemeId> <cfStartDate>2023-04-16T24:00:00</cfStartDate> <cfFamilyNames>Iapăscurtă</cfFamilyNames> <cfFirstNames>Victor</cfFirstNames> </cfPersName_Pers> </cfPers> <cfPers> <cfPersId>ibn-Pers-12933</cfPersId> <cfPersName_Pers> <cfPersNameId>ibn-PersName-12933-3</cfPersNameId> <cfClassId>55f90543-d631-42eb-8d47-d8d9266cbb26</cfClassId> <cfClassSchemeId>7375609d-cfa6-45ce-a803-75de69abe21f</cfClassSchemeId> <cfStartDate>2023-04-16T24:00:00</cfStartDate> <cfFamilyNames>Fiodorov</cfFamilyNames> <cfFirstNames>Ion</cfFirstNames> </cfPersName_Pers> </cfPers> <cfSrv> <cfSrvId>5123451</cfSrvId> <cfName cfLangCode='en' cfTrans='o'>CrossRef DOI prefix service</cfName> <cfDescr cfLangCode='en' cfTrans='o'>The service of issuing DOI prefixes to publishers</cfDescr> <cfKeyw cfLangCode='en' cfTrans='o'>persistent identifier; Digital Object Identifier</cfKeyw> </cfSrv> </CERIF>