Interpretable clinical time-series modeling with intelligent feature selection for early prediction of antimicrobial multidrug resistance
Electronic health records provide rich, heterogeneous data about the evolution of the patients’ health status. However, such data need to be processed carefully, with the aim of extracting meaningful information for clinical decision support. In this paper, we leverage interpretable (deep) learning...
Ausführliche Beschreibung
Gespeichert in:
| Autor*in: |
Martínez-Agüero, Sergio [verfasserIn] |
|---|
| Format: |
E-Artikel |
|---|---|
| Sprache: |
Englisch |
| Erschienen: |
2022transfer abstract |
|---|
| Schlagwörter: |
|---|
| Umfang: |
16 |
|---|
| Übergeordnetes Werk: |
Enthalten in: Surgeon-patient matching based on pairwise comparisons information for elective surgery - Jiang, Yan-Ping ELSEVIER, 2020, Amsterdam [u.a.] |
|---|---|
| Übergeordnetes Werk: |
volume:133 ; year:2022 ; pages:68-83 ; extent:16 |
| Links: |
|---|
| DOI / URN: |
10.1016/j.future.2022.02.021 |
|---|
| Katalog-ID: |
ELV057490953 |
|---|
| LEADER | 01000caa a22002652 4500 | ||
|---|---|---|---|
| 001 | ELV057490953 | ||
| 003 | DE-627 | ||
| 005 | 20230626045218.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 220808s2022 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1016/j.future.2022.02.021 |2 doi | |
| 028 | 5 | 2 | |a /cbs_pica/cbs_olc/import_discovery/elsevier/einzuspielen/GBV00000000001931.pica |
| 035 | |a (DE-627)ELV057490953 | ||
| 035 | |a (ELSEVIER)S0167-739X(22)00064-4 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 082 | 0 | 4 | |a 004 |q VZ |
| 084 | |a 85.35 |2 bkl | ||
| 084 | |a 54.80 |2 bkl | ||
| 100 | 1 | |a Martínez-Agüero, Sergio |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a Interpretable clinical time-series modeling with intelligent feature selection for early prediction of antimicrobial multidrug resistance |
| 264 | 1 | |c 2022transfer abstract | |
| 300 | |a 16 | ||
| 336 | |a nicht spezifiziert |b zzz |2 rdacontent | ||
| 337 | |a nicht spezifiziert |b z |2 rdamedia | ||
| 338 | |a nicht spezifiziert |b zu |2 rdacarrier | ||
| 520 | |a Electronic health records provide rich, heterogeneous data about the evolution of the patients’ health status. However, such data need to be processed carefully, with the aim of extracting meaningful information for clinical decision support. In this paper, we leverage interpretable (deep) learning and signal processing tools to deal with multivariate time-series data collected from the Intensive Care Unit (ICU) of the University Hospital of Fuenlabrada (Madrid, Spain). The presence of antimicrobial multidrug-resistant (AMR) bacteria is one of the greatest threats to the health system in general and to the ICUs in particular due to the critical health status of the patients therein. Thus, early identification of bacteria at the ICU and early prediction of their antibiotic resistance are key for the patients’ prognosis. While intelligent data-based processing and learning schemes can contribute to this early prediction, their acceptance and deployment in the ICUs require the automatic schemes to be not only accurate but also understandable by clinicians. Accordingly, we have designed trustworthy intelligent models for the early prediction of AMR based on the combination of meaningful feature selection with interpretable recurrent neural networks. These models were created using irregularly sampled clinical measurements, both considering the health status of the patient and the global ICU environment. We explored several strategies to cope with strongly imbalance data, since only a few ICU patients are infected by AMR bacteria. It is worth noting that our approach exhibits a good balance between performance and interpretability, especially when considering the difficulty of the classification task at hand. A multitude of factors are involved in the emergence of AMR (several of them not fully understood), and the records only contain a subset of them. In addition, the limited number of patients, the imbalance between classes, and the irregularity of the data render the problem harder to solve. Our models are also enriched with SHAP post-hoc interpretability and validated by clinicians who considered model understandability and trustworthiness of paramount concern for pragmatic purposes. Moreover, we use linguistic fuzzy systems to provide clinicians with explanations in natural language. Such explanations are automatically generated from a pool of interpretable rules that describe the interaction among the most relevant features identified by SHAP. Notice tha... | ||
| 520 | |a Electronic health records provide rich, heterogeneous data about the evolution of the patients’ health status. However, such data need to be processed carefully, with the aim of extracting meaningful information for clinical decision support. In this paper, we leverage interpretable (deep) learning and signal processing tools to deal with multivariate time-series data collected from the Intensive Care Unit (ICU) of the University Hospital of Fuenlabrada (Madrid, Spain). The presence of antimicrobial multidrug-resistant (AMR) bacteria is one of the greatest threats to the health system in general and to the ICUs in particular due to the critical health status of the patients therein. Thus, early identification of bacteria at the ICU and early prediction of their antibiotic resistance are key for the patients’ prognosis. While intelligent data-based processing and learning schemes can contribute to this early prediction, their acceptance and deployment in the ICUs require the automatic schemes to be not only accurate but also understandable by clinicians. Accordingly, we have designed trustworthy intelligent models for the early prediction of AMR based on the combination of meaningful feature selection with interpretable recurrent neural networks. These models were created using irregularly sampled clinical measurements, both considering the health status of the patient and the global ICU environment. We explored several strategies to cope with strongly imbalance data, since only a few ICU patients are infected by AMR bacteria. It is worth noting that our approach exhibits a good balance between performance and interpretability, especially when considering the difficulty of the classification task at hand. A multitude of factors are involved in the emergence of AMR (several of them not fully understood), and the records only contain a subset of them. In addition, the limited number of patients, the imbalance between classes, and the irregularity of the data render the problem harder to solve. Our models are also enriched with SHAP post-hoc interpretability and validated by clinicians who considered model understandability and trustworthiness of paramount concern for pragmatic purposes. Moreover, we use linguistic fuzzy systems to provide clinicians with explanations in natural language. Such explanations are automatically generated from a pool of interpretable rules that describe the interaction among the most relevant features identified by SHAP. Notice tha... | ||
| 650 | 7 | |a 37M10 |2 Elsevier | |
| 650 | 7 | |a 68T01 |2 Elsevier | |
| 650 | 7 | |a 94D05 |2 Elsevier | |
| 650 | 7 | |a 68T07 |2 Elsevier | |
| 650 | 7 | |a 92C50 |2 Elsevier | |
| 700 | 1 | |a Soguero-Ruiz, Cristina |4 oth | |
| 700 | 1 | |a Alonso-Moral, Jose M. |4 oth | |
| 700 | 1 | |a Mora-Jiménez, Inmaculada |4 oth | |
| 700 | 1 | |a Álvarez-Rodríguez, Joaquín |4 oth | |
| 700 | 1 | |a Marques, Antonio G. |4 oth | |
| 773 | 0 | 8 | |i Enthalten in |n Elsevier Science |a Jiang, Yan-Ping ELSEVIER |t Surgeon-patient matching based on pairwise comparisons information for elective surgery |d 2020 |g Amsterdam [u.a.] |w (DE-627)ELV004280385 |
| 773 | 1 | 8 | |g volume:133 |g year:2022 |g pages:68-83 |g extent:16 |
| 856 | 4 | 0 | |u https://doi.org/10.1016/j.future.2022.02.021 |3 Volltext |
| 912 | |a GBV_USEFLAG_U | ||
| 912 | |a GBV_ELV | ||
| 912 | |a SYSFLAG_U | ||
| 936 | b | k | |a 85.35 |j Fertigung |q VZ |
| 936 | b | k | |a 54.80 |j Angewandte Informatik |q VZ |
| 951 | |a AR | ||
| 952 | |d 133 |j 2022 |h 68-83 |g 16 | ||
| author_variant |
s m a sma |
|---|---|
| matchkey_str |
martnezagerosergiosogueroruizcristinaalo:2022----:nepealciiatmsremdlnwtitlietetrslcinoeryrdcinf |
| hierarchy_sort_str |
2022transfer abstract |
| bklnumber |
85.35 54.80 |
| publishDate |
2022 |
| allfields |
10.1016/j.future.2022.02.021 doi /cbs_pica/cbs_olc/import_discovery/elsevier/einzuspielen/GBV00000000001931.pica (DE-627)ELV057490953 (ELSEVIER)S0167-739X(22)00064-4 DE-627 ger DE-627 rakwb eng 004 VZ 85.35 bkl 54.80 bkl Martínez-Agüero, Sergio verfasserin aut Interpretable clinical time-series modeling with intelligent feature selection for early prediction of antimicrobial multidrug resistance 2022transfer abstract 16 nicht spezifiziert zzz rdacontent nicht spezifiziert z rdamedia nicht spezifiziert zu rdacarrier Electronic health records provide rich, heterogeneous data about the evolution of the patients’ health status. However, such data need to be processed carefully, with the aim of extracting meaningful information for clinical decision support. In this paper, we leverage interpretable (deep) learning and signal processing tools to deal with multivariate time-series data collected from the Intensive Care Unit (ICU) of the University Hospital of Fuenlabrada (Madrid, Spain). The presence of antimicrobial multidrug-resistant (AMR) bacteria is one of the greatest threats to the health system in general and to the ICUs in particular due to the critical health status of the patients therein. Thus, early identification of bacteria at the ICU and early prediction of their antibiotic resistance are key for the patients’ prognosis. While intelligent data-based processing and learning schemes can contribute to this early prediction, their acceptance and deployment in the ICUs require the automatic schemes to be not only accurate but also understandable by clinicians. Accordingly, we have designed trustworthy intelligent models for the early prediction of AMR based on the combination of meaningful feature selection with interpretable recurrent neural networks. These models were created using irregularly sampled clinical measurements, both considering the health status of the patient and the global ICU environment. We explored several strategies to cope with strongly imbalance data, since only a few ICU patients are infected by AMR bacteria. It is worth noting that our approach exhibits a good balance between performance and interpretability, especially when considering the difficulty of the classification task at hand. A multitude of factors are involved in the emergence of AMR (several of them not fully understood), and the records only contain a subset of them. In addition, the limited number of patients, the imbalance between classes, and the irregularity of the data render the problem harder to solve. Our models are also enriched with SHAP post-hoc interpretability and validated by clinicians who considered model understandability and trustworthiness of paramount concern for pragmatic purposes. Moreover, we use linguistic fuzzy systems to provide clinicians with explanations in natural language. Such explanations are automatically generated from a pool of interpretable rules that describe the interaction among the most relevant features identified by SHAP. Notice tha... Electronic health records provide rich, heterogeneous data about the evolution of the patients’ health status. However, such data need to be processed carefully, with the aim of extracting meaningful information for clinical decision support. In this paper, we leverage interpretable (deep) learning and signal processing tools to deal with multivariate time-series data collected from the Intensive Care Unit (ICU) of the University Hospital of Fuenlabrada (Madrid, Spain). The presence of antimicrobial multidrug-resistant (AMR) bacteria is one of the greatest threats to the health system in general and to the ICUs in particular due to the critical health status of the patients therein. Thus, early identification of bacteria at the ICU and early prediction of their antibiotic resistance are key for the patients’ prognosis. While intelligent data-based processing and learning schemes can contribute to this early prediction, their acceptance and deployment in the ICUs require the automatic schemes to be not only accurate but also understandable by clinicians. Accordingly, we have designed trustworthy intelligent models for the early prediction of AMR based on the combination of meaningful feature selection with interpretable recurrent neural networks. These models were created using irregularly sampled clinical measurements, both considering the health status of the patient and the global ICU environment. We explored several strategies to cope with strongly imbalance data, since only a few ICU patients are infected by AMR bacteria. It is worth noting that our approach exhibits a good balance between performance and interpretability, especially when considering the difficulty of the classification task at hand. A multitude of factors are involved in the emergence of AMR (several of them not fully understood), and the records only contain a subset of them. In addition, the limited number of patients, the imbalance between classes, and the irregularity of the data render the problem harder to solve. Our models are also enriched with SHAP post-hoc interpretability and validated by clinicians who considered model understandability and trustworthiness of paramount concern for pragmatic purposes. Moreover, we use linguistic fuzzy systems to provide clinicians with explanations in natural language. Such explanations are automatically generated from a pool of interpretable rules that describe the interaction among the most relevant features identified by SHAP. Notice tha... 37M10 Elsevier 68T01 Elsevier 94D05 Elsevier 68T07 Elsevier 92C50 Elsevier Soguero-Ruiz, Cristina oth Alonso-Moral, Jose M. oth Mora-Jiménez, Inmaculada oth Álvarez-Rodríguez, Joaquín oth Marques, Antonio G. oth Enthalten in Elsevier Science Jiang, Yan-Ping ELSEVIER Surgeon-patient matching based on pairwise comparisons information for elective surgery 2020 Amsterdam [u.a.] (DE-627)ELV004280385 volume:133 year:2022 pages:68-83 extent:16 https://doi.org/10.1016/j.future.2022.02.021 Volltext GBV_USEFLAG_U GBV_ELV SYSFLAG_U 85.35 Fertigung VZ 54.80 Angewandte Informatik VZ AR 133 2022 68-83 16 |
| spelling |
10.1016/j.future.2022.02.021 doi /cbs_pica/cbs_olc/import_discovery/elsevier/einzuspielen/GBV00000000001931.pica (DE-627)ELV057490953 (ELSEVIER)S0167-739X(22)00064-4 DE-627 ger DE-627 rakwb eng 004 VZ 85.35 bkl 54.80 bkl Martínez-Agüero, Sergio verfasserin aut Interpretable clinical time-series modeling with intelligent feature selection for early prediction of antimicrobial multidrug resistance 2022transfer abstract 16 nicht spezifiziert zzz rdacontent nicht spezifiziert z rdamedia nicht spezifiziert zu rdacarrier Electronic health records provide rich, heterogeneous data about the evolution of the patients’ health status. However, such data need to be processed carefully, with the aim of extracting meaningful information for clinical decision support. In this paper, we leverage interpretable (deep) learning and signal processing tools to deal with multivariate time-series data collected from the Intensive Care Unit (ICU) of the University Hospital of Fuenlabrada (Madrid, Spain). The presence of antimicrobial multidrug-resistant (AMR) bacteria is one of the greatest threats to the health system in general and to the ICUs in particular due to the critical health status of the patients therein. Thus, early identification of bacteria at the ICU and early prediction of their antibiotic resistance are key for the patients’ prognosis. While intelligent data-based processing and learning schemes can contribute to this early prediction, their acceptance and deployment in the ICUs require the automatic schemes to be not only accurate but also understandable by clinicians. Accordingly, we have designed trustworthy intelligent models for the early prediction of AMR based on the combination of meaningful feature selection with interpretable recurrent neural networks. These models were created using irregularly sampled clinical measurements, both considering the health status of the patient and the global ICU environment. We explored several strategies to cope with strongly imbalance data, since only a few ICU patients are infected by AMR bacteria. It is worth noting that our approach exhibits a good balance between performance and interpretability, especially when considering the difficulty of the classification task at hand. A multitude of factors are involved in the emergence of AMR (several of them not fully understood), and the records only contain a subset of them. In addition, the limited number of patients, the imbalance between classes, and the irregularity of the data render the problem harder to solve. Our models are also enriched with SHAP post-hoc interpretability and validated by clinicians who considered model understandability and trustworthiness of paramount concern for pragmatic purposes. Moreover, we use linguistic fuzzy systems to provide clinicians with explanations in natural language. Such explanations are automatically generated from a pool of interpretable rules that describe the interaction among the most relevant features identified by SHAP. Notice tha... Electronic health records provide rich, heterogeneous data about the evolution of the patients’ health status. However, such data need to be processed carefully, with the aim of extracting meaningful information for clinical decision support. In this paper, we leverage interpretable (deep) learning and signal processing tools to deal with multivariate time-series data collected from the Intensive Care Unit (ICU) of the University Hospital of Fuenlabrada (Madrid, Spain). The presence of antimicrobial multidrug-resistant (AMR) bacteria is one of the greatest threats to the health system in general and to the ICUs in particular due to the critical health status of the patients therein. Thus, early identification of bacteria at the ICU and early prediction of their antibiotic resistance are key for the patients’ prognosis. While intelligent data-based processing and learning schemes can contribute to this early prediction, their acceptance and deployment in the ICUs require the automatic schemes to be not only accurate but also understandable by clinicians. Accordingly, we have designed trustworthy intelligent models for the early prediction of AMR based on the combination of meaningful feature selection with interpretable recurrent neural networks. These models were created using irregularly sampled clinical measurements, both considering the health status of the patient and the global ICU environment. We explored several strategies to cope with strongly imbalance data, since only a few ICU patients are infected by AMR bacteria. It is worth noting that our approach exhibits a good balance between performance and interpretability, especially when considering the difficulty of the classification task at hand. A multitude of factors are involved in the emergence of AMR (several of them not fully understood), and the records only contain a subset of them. In addition, the limited number of patients, the imbalance between classes, and the irregularity of the data render the problem harder to solve. Our models are also enriched with SHAP post-hoc interpretability and validated by clinicians who considered model understandability and trustworthiness of paramount concern for pragmatic purposes. Moreover, we use linguistic fuzzy systems to provide clinicians with explanations in natural language. Such explanations are automatically generated from a pool of interpretable rules that describe the interaction among the most relevant features identified by SHAP. Notice tha... 37M10 Elsevier 68T01 Elsevier 94D05 Elsevier 68T07 Elsevier 92C50 Elsevier Soguero-Ruiz, Cristina oth Alonso-Moral, Jose M. oth Mora-Jiménez, Inmaculada oth Álvarez-Rodríguez, Joaquín oth Marques, Antonio G. oth Enthalten in Elsevier Science Jiang, Yan-Ping ELSEVIER Surgeon-patient matching based on pairwise comparisons information for elective surgery 2020 Amsterdam [u.a.] (DE-627)ELV004280385 volume:133 year:2022 pages:68-83 extent:16 https://doi.org/10.1016/j.future.2022.02.021 Volltext GBV_USEFLAG_U GBV_ELV SYSFLAG_U 85.35 Fertigung VZ 54.80 Angewandte Informatik VZ AR 133 2022 68-83 16 |
| allfields_unstemmed |
10.1016/j.future.2022.02.021 doi /cbs_pica/cbs_olc/import_discovery/elsevier/einzuspielen/GBV00000000001931.pica (DE-627)ELV057490953 (ELSEVIER)S0167-739X(22)00064-4 DE-627 ger DE-627 rakwb eng 004 VZ 85.35 bkl 54.80 bkl Martínez-Agüero, Sergio verfasserin aut Interpretable clinical time-series modeling with intelligent feature selection for early prediction of antimicrobial multidrug resistance 2022transfer abstract 16 nicht spezifiziert zzz rdacontent nicht spezifiziert z rdamedia nicht spezifiziert zu rdacarrier Electronic health records provide rich, heterogeneous data about the evolution of the patients’ health status. However, such data need to be processed carefully, with the aim of extracting meaningful information for clinical decision support. In this paper, we leverage interpretable (deep) learning and signal processing tools to deal with multivariate time-series data collected from the Intensive Care Unit (ICU) of the University Hospital of Fuenlabrada (Madrid, Spain). The presence of antimicrobial multidrug-resistant (AMR) bacteria is one of the greatest threats to the health system in general and to the ICUs in particular due to the critical health status of the patients therein. Thus, early identification of bacteria at the ICU and early prediction of their antibiotic resistance are key for the patients’ prognosis. While intelligent data-based processing and learning schemes can contribute to this early prediction, their acceptance and deployment in the ICUs require the automatic schemes to be not only accurate but also understandable by clinicians. Accordingly, we have designed trustworthy intelligent models for the early prediction of AMR based on the combination of meaningful feature selection with interpretable recurrent neural networks. These models were created using irregularly sampled clinical measurements, both considering the health status of the patient and the global ICU environment. We explored several strategies to cope with strongly imbalance data, since only a few ICU patients are infected by AMR bacteria. It is worth noting that our approach exhibits a good balance between performance and interpretability, especially when considering the difficulty of the classification task at hand. A multitude of factors are involved in the emergence of AMR (several of them not fully understood), and the records only contain a subset of them. In addition, the limited number of patients, the imbalance between classes, and the irregularity of the data render the problem harder to solve. Our models are also enriched with SHAP post-hoc interpretability and validated by clinicians who considered model understandability and trustworthiness of paramount concern for pragmatic purposes. Moreover, we use linguistic fuzzy systems to provide clinicians with explanations in natural language. Such explanations are automatically generated from a pool of interpretable rules that describe the interaction among the most relevant features identified by SHAP. Notice tha... Electronic health records provide rich, heterogeneous data about the evolution of the patients’ health status. However, such data need to be processed carefully, with the aim of extracting meaningful information for clinical decision support. In this paper, we leverage interpretable (deep) learning and signal processing tools to deal with multivariate time-series data collected from the Intensive Care Unit (ICU) of the University Hospital of Fuenlabrada (Madrid, Spain). The presence of antimicrobial multidrug-resistant (AMR) bacteria is one of the greatest threats to the health system in general and to the ICUs in particular due to the critical health status of the patients therein. Thus, early identification of bacteria at the ICU and early prediction of their antibiotic resistance are key for the patients’ prognosis. While intelligent data-based processing and learning schemes can contribute to this early prediction, their acceptance and deployment in the ICUs require the automatic schemes to be not only accurate but also understandable by clinicians. Accordingly, we have designed trustworthy intelligent models for the early prediction of AMR based on the combination of meaningful feature selection with interpretable recurrent neural networks. These models were created using irregularly sampled clinical measurements, both considering the health status of the patient and the global ICU environment. We explored several strategies to cope with strongly imbalance data, since only a few ICU patients are infected by AMR bacteria. It is worth noting that our approach exhibits a good balance between performance and interpretability, especially when considering the difficulty of the classification task at hand. A multitude of factors are involved in the emergence of AMR (several of them not fully understood), and the records only contain a subset of them. In addition, the limited number of patients, the imbalance between classes, and the irregularity of the data render the problem harder to solve. Our models are also enriched with SHAP post-hoc interpretability and validated by clinicians who considered model understandability and trustworthiness of paramount concern for pragmatic purposes. Moreover, we use linguistic fuzzy systems to provide clinicians with explanations in natural language. Such explanations are automatically generated from a pool of interpretable rules that describe the interaction among the most relevant features identified by SHAP. Notice tha... 37M10 Elsevier 68T01 Elsevier 94D05 Elsevier 68T07 Elsevier 92C50 Elsevier Soguero-Ruiz, Cristina oth Alonso-Moral, Jose M. oth Mora-Jiménez, Inmaculada oth Álvarez-Rodríguez, Joaquín oth Marques, Antonio G. oth Enthalten in Elsevier Science Jiang, Yan-Ping ELSEVIER Surgeon-patient matching based on pairwise comparisons information for elective surgery 2020 Amsterdam [u.a.] (DE-627)ELV004280385 volume:133 year:2022 pages:68-83 extent:16 https://doi.org/10.1016/j.future.2022.02.021 Volltext GBV_USEFLAG_U GBV_ELV SYSFLAG_U 85.35 Fertigung VZ 54.80 Angewandte Informatik VZ AR 133 2022 68-83 16 |
| allfieldsGer |
10.1016/j.future.2022.02.021 doi /cbs_pica/cbs_olc/import_discovery/elsevier/einzuspielen/GBV00000000001931.pica (DE-627)ELV057490953 (ELSEVIER)S0167-739X(22)00064-4 DE-627 ger DE-627 rakwb eng 004 VZ 85.35 bkl 54.80 bkl Martínez-Agüero, Sergio verfasserin aut Interpretable clinical time-series modeling with intelligent feature selection for early prediction of antimicrobial multidrug resistance 2022transfer abstract 16 nicht spezifiziert zzz rdacontent nicht spezifiziert z rdamedia nicht spezifiziert zu rdacarrier Electronic health records provide rich, heterogeneous data about the evolution of the patients’ health status. However, such data need to be processed carefully, with the aim of extracting meaningful information for clinical decision support. In this paper, we leverage interpretable (deep) learning and signal processing tools to deal with multivariate time-series data collected from the Intensive Care Unit (ICU) of the University Hospital of Fuenlabrada (Madrid, Spain). The presence of antimicrobial multidrug-resistant (AMR) bacteria is one of the greatest threats to the health system in general and to the ICUs in particular due to the critical health status of the patients therein. Thus, early identification of bacteria at the ICU and early prediction of their antibiotic resistance are key for the patients’ prognosis. While intelligent data-based processing and learning schemes can contribute to this early prediction, their acceptance and deployment in the ICUs require the automatic schemes to be not only accurate but also understandable by clinicians. Accordingly, we have designed trustworthy intelligent models for the early prediction of AMR based on the combination of meaningful feature selection with interpretable recurrent neural networks. These models were created using irregularly sampled clinical measurements, both considering the health status of the patient and the global ICU environment. We explored several strategies to cope with strongly imbalance data, since only a few ICU patients are infected by AMR bacteria. It is worth noting that our approach exhibits a good balance between performance and interpretability, especially when considering the difficulty of the classification task at hand. A multitude of factors are involved in the emergence of AMR (several of them not fully understood), and the records only contain a subset of them. In addition, the limited number of patients, the imbalance between classes, and the irregularity of the data render the problem harder to solve. Our models are also enriched with SHAP post-hoc interpretability and validated by clinicians who considered model understandability and trustworthiness of paramount concern for pragmatic purposes. Moreover, we use linguistic fuzzy systems to provide clinicians with explanations in natural language. Such explanations are automatically generated from a pool of interpretable rules that describe the interaction among the most relevant features identified by SHAP. Notice tha... Electronic health records provide rich, heterogeneous data about the evolution of the patients’ health status. However, such data need to be processed carefully, with the aim of extracting meaningful information for clinical decision support. In this paper, we leverage interpretable (deep) learning and signal processing tools to deal with multivariate time-series data collected from the Intensive Care Unit (ICU) of the University Hospital of Fuenlabrada (Madrid, Spain). The presence of antimicrobial multidrug-resistant (AMR) bacteria is one of the greatest threats to the health system in general and to the ICUs in particular due to the critical health status of the patients therein. Thus, early identification of bacteria at the ICU and early prediction of their antibiotic resistance are key for the patients’ prognosis. While intelligent data-based processing and learning schemes can contribute to this early prediction, their acceptance and deployment in the ICUs require the automatic schemes to be not only accurate but also understandable by clinicians. Accordingly, we have designed trustworthy intelligent models for the early prediction of AMR based on the combination of meaningful feature selection with interpretable recurrent neural networks. These models were created using irregularly sampled clinical measurements, both considering the health status of the patient and the global ICU environment. We explored several strategies to cope with strongly imbalance data, since only a few ICU patients are infected by AMR bacteria. It is worth noting that our approach exhibits a good balance between performance and interpretability, especially when considering the difficulty of the classification task at hand. A multitude of factors are involved in the emergence of AMR (several of them not fully understood), and the records only contain a subset of them. In addition, the limited number of patients, the imbalance between classes, and the irregularity of the data render the problem harder to solve. Our models are also enriched with SHAP post-hoc interpretability and validated by clinicians who considered model understandability and trustworthiness of paramount concern for pragmatic purposes. Moreover, we use linguistic fuzzy systems to provide clinicians with explanations in natural language. Such explanations are automatically generated from a pool of interpretable rules that describe the interaction among the most relevant features identified by SHAP. Notice tha... 37M10 Elsevier 68T01 Elsevier 94D05 Elsevier 68T07 Elsevier 92C50 Elsevier Soguero-Ruiz, Cristina oth Alonso-Moral, Jose M. oth Mora-Jiménez, Inmaculada oth Álvarez-Rodríguez, Joaquín oth Marques, Antonio G. oth Enthalten in Elsevier Science Jiang, Yan-Ping ELSEVIER Surgeon-patient matching based on pairwise comparisons information for elective surgery 2020 Amsterdam [u.a.] (DE-627)ELV004280385 volume:133 year:2022 pages:68-83 extent:16 https://doi.org/10.1016/j.future.2022.02.021 Volltext GBV_USEFLAG_U GBV_ELV SYSFLAG_U 85.35 Fertigung VZ 54.80 Angewandte Informatik VZ AR 133 2022 68-83 16 |
| allfieldsSound |
10.1016/j.future.2022.02.021 doi /cbs_pica/cbs_olc/import_discovery/elsevier/einzuspielen/GBV00000000001931.pica (DE-627)ELV057490953 (ELSEVIER)S0167-739X(22)00064-4 DE-627 ger DE-627 rakwb eng 004 VZ 85.35 bkl 54.80 bkl Martínez-Agüero, Sergio verfasserin aut Interpretable clinical time-series modeling with intelligent feature selection for early prediction of antimicrobial multidrug resistance 2022transfer abstract 16 nicht spezifiziert zzz rdacontent nicht spezifiziert z rdamedia nicht spezifiziert zu rdacarrier Electronic health records provide rich, heterogeneous data about the evolution of the patients’ health status. However, such data need to be processed carefully, with the aim of extracting meaningful information for clinical decision support. In this paper, we leverage interpretable (deep) learning and signal processing tools to deal with multivariate time-series data collected from the Intensive Care Unit (ICU) of the University Hospital of Fuenlabrada (Madrid, Spain). The presence of antimicrobial multidrug-resistant (AMR) bacteria is one of the greatest threats to the health system in general and to the ICUs in particular due to the critical health status of the patients therein. Thus, early identification of bacteria at the ICU and early prediction of their antibiotic resistance are key for the patients’ prognosis. While intelligent data-based processing and learning schemes can contribute to this early prediction, their acceptance and deployment in the ICUs require the automatic schemes to be not only accurate but also understandable by clinicians. Accordingly, we have designed trustworthy intelligent models for the early prediction of AMR based on the combination of meaningful feature selection with interpretable recurrent neural networks. These models were created using irregularly sampled clinical measurements, both considering the health status of the patient and the global ICU environment. We explored several strategies to cope with strongly imbalance data, since only a few ICU patients are infected by AMR bacteria. It is worth noting that our approach exhibits a good balance between performance and interpretability, especially when considering the difficulty of the classification task at hand. A multitude of factors are involved in the emergence of AMR (several of them not fully understood), and the records only contain a subset of them. In addition, the limited number of patients, the imbalance between classes, and the irregularity of the data render the problem harder to solve. Our models are also enriched with SHAP post-hoc interpretability and validated by clinicians who considered model understandability and trustworthiness of paramount concern for pragmatic purposes. Moreover, we use linguistic fuzzy systems to provide clinicians with explanations in natural language. Such explanations are automatically generated from a pool of interpretable rules that describe the interaction among the most relevant features identified by SHAP. Notice tha... Electronic health records provide rich, heterogeneous data about the evolution of the patients’ health status. However, such data need to be processed carefully, with the aim of extracting meaningful information for clinical decision support. In this paper, we leverage interpretable (deep) learning and signal processing tools to deal with multivariate time-series data collected from the Intensive Care Unit (ICU) of the University Hospital of Fuenlabrada (Madrid, Spain). The presence of antimicrobial multidrug-resistant (AMR) bacteria is one of the greatest threats to the health system in general and to the ICUs in particular due to the critical health status of the patients therein. Thus, early identification of bacteria at the ICU and early prediction of their antibiotic resistance are key for the patients’ prognosis. While intelligent data-based processing and learning schemes can contribute to this early prediction, their acceptance and deployment in the ICUs require the automatic schemes to be not only accurate but also understandable by clinicians. Accordingly, we have designed trustworthy intelligent models for the early prediction of AMR based on the combination of meaningful feature selection with interpretable recurrent neural networks. These models were created using irregularly sampled clinical measurements, both considering the health status of the patient and the global ICU environment. We explored several strategies to cope with strongly imbalance data, since only a few ICU patients are infected by AMR bacteria. It is worth noting that our approach exhibits a good balance between performance and interpretability, especially when considering the difficulty of the classification task at hand. A multitude of factors are involved in the emergence of AMR (several of them not fully understood), and the records only contain a subset of them. In addition, the limited number of patients, the imbalance between classes, and the irregularity of the data render the problem harder to solve. Our models are also enriched with SHAP post-hoc interpretability and validated by clinicians who considered model understandability and trustworthiness of paramount concern for pragmatic purposes. Moreover, we use linguistic fuzzy systems to provide clinicians with explanations in natural language. Such explanations are automatically generated from a pool of interpretable rules that describe the interaction among the most relevant features identified by SHAP. Notice tha... 37M10 Elsevier 68T01 Elsevier 94D05 Elsevier 68T07 Elsevier 92C50 Elsevier Soguero-Ruiz, Cristina oth Alonso-Moral, Jose M. oth Mora-Jiménez, Inmaculada oth Álvarez-Rodríguez, Joaquín oth Marques, Antonio G. oth Enthalten in Elsevier Science Jiang, Yan-Ping ELSEVIER Surgeon-patient matching based on pairwise comparisons information for elective surgery 2020 Amsterdam [u.a.] (DE-627)ELV004280385 volume:133 year:2022 pages:68-83 extent:16 https://doi.org/10.1016/j.future.2022.02.021 Volltext GBV_USEFLAG_U GBV_ELV SYSFLAG_U 85.35 Fertigung VZ 54.80 Angewandte Informatik VZ AR 133 2022 68-83 16 |
| language |
English |
| source |
Enthalten in Surgeon-patient matching based on pairwise comparisons information for elective surgery Amsterdam [u.a.] volume:133 year:2022 pages:68-83 extent:16 |
| sourceStr |
Enthalten in Surgeon-patient matching based on pairwise comparisons information for elective surgery Amsterdam [u.a.] volume:133 year:2022 pages:68-83 extent:16 |
| format_phy_str_mv |
Article |
| bklname |
Fertigung Angewandte Informatik |
| institution |
findex.gbv.de |
| topic_facet |
37M10 68T01 94D05 68T07 92C50 |
| dewey-raw |
004 |
| isfreeaccess_bool |
false |
| container_title |
Surgeon-patient matching based on pairwise comparisons information for elective surgery |
| authorswithroles_txt_mv |
Martínez-Agüero, Sergio @@aut@@ Soguero-Ruiz, Cristina @@oth@@ Alonso-Moral, Jose M. @@oth@@ Mora-Jiménez, Inmaculada @@oth@@ Álvarez-Rodríguez, Joaquín @@oth@@ Marques, Antonio G. @@oth@@ |
| publishDateDaySort_date |
2022-01-01T00:00:00Z |
| hierarchy_top_id |
ELV004280385 |
| dewey-sort |
14 |
| id |
ELV057490953 |
| language_de |
englisch |
| fullrecord |
<?xml version="1.0" encoding="UTF-8"?><collection xmlns="http://www.loc.gov/MARC21/slim"><record><leader>01000caa a22002652 4500</leader><controlfield tag="001">ELV057490953</controlfield><controlfield tag="003">DE-627</controlfield><controlfield tag="005">20230626045218.0</controlfield><controlfield tag="007">cr uuu---uuuuu</controlfield><controlfield tag="008">220808s2022 xx |||||o 00| ||eng c</controlfield><datafield tag="024" ind1="7" ind2=" "><subfield code="a">10.1016/j.future.2022.02.021</subfield><subfield code="2">doi</subfield></datafield><datafield tag="028" ind1="5" ind2="2"><subfield code="a">/cbs_pica/cbs_olc/import_discovery/elsevier/einzuspielen/GBV00000000001931.pica</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-627)ELV057490953</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(ELSEVIER)S0167-739X(22)00064-4</subfield></datafield><datafield tag="040" ind1=" " ind2=" "><subfield code="a">DE-627</subfield><subfield code="b">ger</subfield><subfield code="c">DE-627</subfield><subfield code="e">rakwb</subfield></datafield><datafield tag="041" ind1=" " ind2=" "><subfield code="a">eng</subfield></datafield><datafield tag="082" ind1="0" ind2="4"><subfield code="a">004</subfield><subfield code="q">VZ</subfield></datafield><datafield tag="084" ind1=" " ind2=" "><subfield code="a">85.35</subfield><subfield code="2">bkl</subfield></datafield><datafield tag="084" ind1=" " ind2=" "><subfield code="a">54.80</subfield><subfield code="2">bkl</subfield></datafield><datafield tag="100" ind1="1" ind2=" "><subfield code="a">Martínez-Agüero, Sergio</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="245" ind1="1" ind2="0"><subfield code="a">Interpretable clinical time-series modeling with intelligent feature selection for early prediction of antimicrobial multidrug resistance</subfield></datafield><datafield tag="264" ind1=" " ind2="1"><subfield code="c">2022transfer abstract</subfield></datafield><datafield tag="300" ind1=" " ind2=" "><subfield code="a">16</subfield></datafield><datafield tag="336" ind1=" " ind2=" "><subfield code="a">nicht spezifiziert</subfield><subfield code="b">zzz</subfield><subfield code="2">rdacontent</subfield></datafield><datafield tag="337" ind1=" " ind2=" "><subfield code="a">nicht spezifiziert</subfield><subfield code="b">z</subfield><subfield code="2">rdamedia</subfield></datafield><datafield tag="338" ind1=" " ind2=" "><subfield code="a">nicht spezifiziert</subfield><subfield code="b">zu</subfield><subfield code="2">rdacarrier</subfield></datafield><datafield tag="520" ind1=" " ind2=" "><subfield code="a">Electronic health records provide rich, heterogeneous data about the evolution of the patients’ health status. However, such data need to be processed carefully, with the aim of extracting meaningful information for clinical decision support. In this paper, we leverage interpretable (deep) learning and signal processing tools to deal with multivariate time-series data collected from the Intensive Care Unit (ICU) of the University Hospital of Fuenlabrada (Madrid, Spain). The presence of antimicrobial multidrug-resistant (AMR) bacteria is one of the greatest threats to the health system in general and to the ICUs in particular due to the critical health status of the patients therein. Thus, early identification of bacteria at the ICU and early prediction of their antibiotic resistance are key for the patients’ prognosis. While intelligent data-based processing and learning schemes can contribute to this early prediction, their acceptance and deployment in the ICUs require the automatic schemes to be not only accurate but also understandable by clinicians. Accordingly, we have designed trustworthy intelligent models for the early prediction of AMR based on the combination of meaningful feature selection with interpretable recurrent neural networks. These models were created using irregularly sampled clinical measurements, both considering the health status of the patient and the global ICU environment. We explored several strategies to cope with strongly imbalance data, since only a few ICU patients are infected by AMR bacteria. It is worth noting that our approach exhibits a good balance between performance and interpretability, especially when considering the difficulty of the classification task at hand. A multitude of factors are involved in the emergence of AMR (several of them not fully understood), and the records only contain a subset of them. In addition, the limited number of patients, the imbalance between classes, and the irregularity of the data render the problem harder to solve. Our models are also enriched with SHAP post-hoc interpretability and validated by clinicians who considered model understandability and trustworthiness of paramount concern for pragmatic purposes. Moreover, we use linguistic fuzzy systems to provide clinicians with explanations in natural language. Such explanations are automatically generated from a pool of interpretable rules that describe the interaction among the most relevant features identified by SHAP. Notice tha...</subfield></datafield><datafield tag="520" ind1=" " ind2=" "><subfield code="a">Electronic health records provide rich, heterogeneous data about the evolution of the patients’ health status. However, such data need to be processed carefully, with the aim of extracting meaningful information for clinical decision support. In this paper, we leverage interpretable (deep) learning and signal processing tools to deal with multivariate time-series data collected from the Intensive Care Unit (ICU) of the University Hospital of Fuenlabrada (Madrid, Spain). The presence of antimicrobial multidrug-resistant (AMR) bacteria is one of the greatest threats to the health system in general and to the ICUs in particular due to the critical health status of the patients therein. Thus, early identification of bacteria at the ICU and early prediction of their antibiotic resistance are key for the patients’ prognosis. While intelligent data-based processing and learning schemes can contribute to this early prediction, their acceptance and deployment in the ICUs require the automatic schemes to be not only accurate but also understandable by clinicians. Accordingly, we have designed trustworthy intelligent models for the early prediction of AMR based on the combination of meaningful feature selection with interpretable recurrent neural networks. These models were created using irregularly sampled clinical measurements, both considering the health status of the patient and the global ICU environment. We explored several strategies to cope with strongly imbalance data, since only a few ICU patients are infected by AMR bacteria. It is worth noting that our approach exhibits a good balance between performance and interpretability, especially when considering the difficulty of the classification task at hand. A multitude of factors are involved in the emergence of AMR (several of them not fully understood), and the records only contain a subset of them. In addition, the limited number of patients, the imbalance between classes, and the irregularity of the data render the problem harder to solve. Our models are also enriched with SHAP post-hoc interpretability and validated by clinicians who considered model understandability and trustworthiness of paramount concern for pragmatic purposes. Moreover, we use linguistic fuzzy systems to provide clinicians with explanations in natural language. Such explanations are automatically generated from a pool of interpretable rules that describe the interaction among the most relevant features identified by SHAP. Notice tha...</subfield></datafield><datafield tag="650" ind1=" " ind2="7"><subfield code="a">37M10</subfield><subfield code="2">Elsevier</subfield></datafield><datafield tag="650" ind1=" " ind2="7"><subfield code="a">68T01</subfield><subfield code="2">Elsevier</subfield></datafield><datafield tag="650" ind1=" " ind2="7"><subfield code="a">94D05</subfield><subfield code="2">Elsevier</subfield></datafield><datafield tag="650" ind1=" " ind2="7"><subfield code="a">68T07</subfield><subfield code="2">Elsevier</subfield></datafield><datafield tag="650" ind1=" " ind2="7"><subfield code="a">92C50</subfield><subfield code="2">Elsevier</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Soguero-Ruiz, Cristina</subfield><subfield code="4">oth</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Alonso-Moral, Jose M.</subfield><subfield code="4">oth</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Mora-Jiménez, Inmaculada</subfield><subfield code="4">oth</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Álvarez-Rodríguez, Joaquín</subfield><subfield code="4">oth</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Marques, Antonio G.</subfield><subfield code="4">oth</subfield></datafield><datafield tag="773" ind1="0" ind2="8"><subfield code="i">Enthalten in</subfield><subfield code="n">Elsevier Science</subfield><subfield code="a">Jiang, Yan-Ping ELSEVIER</subfield><subfield code="t">Surgeon-patient matching based on pairwise comparisons information for elective surgery</subfield><subfield code="d">2020</subfield><subfield code="g">Amsterdam [u.a.]</subfield><subfield code="w">(DE-627)ELV004280385</subfield></datafield><datafield tag="773" ind1="1" ind2="8"><subfield code="g">volume:133</subfield><subfield code="g">year:2022</subfield><subfield code="g">pages:68-83</subfield><subfield code="g">extent:16</subfield></datafield><datafield tag="856" ind1="4" ind2="0"><subfield code="u">https://doi.org/10.1016/j.future.2022.02.021</subfield><subfield code="3">Volltext</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_USEFLAG_U</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ELV</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SYSFLAG_U</subfield></datafield><datafield tag="936" ind1="b" ind2="k"><subfield code="a">85.35</subfield><subfield code="j">Fertigung</subfield><subfield code="q">VZ</subfield></datafield><datafield tag="936" ind1="b" ind2="k"><subfield code="a">54.80</subfield><subfield code="j">Angewandte Informatik</subfield><subfield code="q">VZ</subfield></datafield><datafield tag="951" ind1=" " ind2=" "><subfield code="a">AR</subfield></datafield><datafield tag="952" ind1=" " ind2=" "><subfield code="d">133</subfield><subfield code="j">2022</subfield><subfield code="h">68-83</subfield><subfield code="g">16</subfield></datafield></record></collection>
|
| author |
Martínez-Agüero, Sergio |
| spellingShingle |
Martínez-Agüero, Sergio ddc 004 bkl 85.35 bkl 54.80 Elsevier 37M10 Elsevier 68T01 Elsevier 94D05 Elsevier 68T07 Elsevier 92C50 Interpretable clinical time-series modeling with intelligent feature selection for early prediction of antimicrobial multidrug resistance |
| authorStr |
Martínez-Agüero, Sergio |
| ppnlink_with_tag_str_mv |
@@773@@(DE-627)ELV004280385 |
| format |
electronic Article |
| dewey-ones |
004 - Data processing & computer science |
| delete_txt_mv |
keep |
| author_role |
aut |
| collection |
elsevier |
| remote_str |
true |
| illustrated |
Not Illustrated |
| topic_title |
004 VZ 85.35 bkl 54.80 bkl Interpretable clinical time-series modeling with intelligent feature selection for early prediction of antimicrobial multidrug resistance 37M10 Elsevier 68T01 Elsevier 94D05 Elsevier 68T07 Elsevier 92C50 Elsevier |
| topic |
ddc 004 bkl 85.35 bkl 54.80 Elsevier 37M10 Elsevier 68T01 Elsevier 94D05 Elsevier 68T07 Elsevier 92C50 |
| topic_unstemmed |
ddc 004 bkl 85.35 bkl 54.80 Elsevier 37M10 Elsevier 68T01 Elsevier 94D05 Elsevier 68T07 Elsevier 92C50 |
| topic_browse |
ddc 004 bkl 85.35 bkl 54.80 Elsevier 37M10 Elsevier 68T01 Elsevier 94D05 Elsevier 68T07 Elsevier 92C50 |
| format_facet |
Elektronische Aufsätze Aufsätze Elektronische Ressource |
| format_main_str_mv |
Text Zeitschrift/Artikel |
| carriertype_str_mv |
zu |
| author2_variant |
c s r csr j m a m jma jmam i m j imj j á r jár a g m ag agm |
| hierarchy_parent_title |
Surgeon-patient matching based on pairwise comparisons information for elective surgery |
| hierarchy_parent_id |
ELV004280385 |
| dewey-tens |
000 - Computer science, knowledge & systems |
| hierarchy_top_title |
Surgeon-patient matching based on pairwise comparisons information for elective surgery |
| isfreeaccess_txt |
false |
| familylinks_str_mv |
(DE-627)ELV004280385 |
| title |
Interpretable clinical time-series modeling with intelligent feature selection for early prediction of antimicrobial multidrug resistance |
| ctrlnum |
(DE-627)ELV057490953 (ELSEVIER)S0167-739X(22)00064-4 |
| title_full |
Interpretable clinical time-series modeling with intelligent feature selection for early prediction of antimicrobial multidrug resistance |
| author_sort |
Martínez-Agüero, Sergio |
| journal |
Surgeon-patient matching based on pairwise comparisons information for elective surgery |
| journalStr |
Surgeon-patient matching based on pairwise comparisons information for elective surgery |
| lang_code |
eng |
| isOA_bool |
false |
| dewey-hundreds |
000 - Computer science, information & general works |
| recordtype |
marc |
| publishDateSort |
2022 |
| contenttype_str_mv |
zzz |
| container_start_page |
68 |
| author_browse |
Martínez-Agüero, Sergio |
| container_volume |
133 |
| physical |
16 |
| class |
004 VZ 85.35 bkl 54.80 bkl |
| format_se |
Elektronische Aufsätze |
| author-letter |
Martínez-Agüero, Sergio |
| doi_str_mv |
10.1016/j.future.2022.02.021 |
| dewey-full |
004 |
| title_sort |
interpretable clinical time-series modeling with intelligent feature selection for early prediction of antimicrobial multidrug resistance |
| title_auth |
Interpretable clinical time-series modeling with intelligent feature selection for early prediction of antimicrobial multidrug resistance |
| abstract |
Electronic health records provide rich, heterogeneous data about the evolution of the patients’ health status. However, such data need to be processed carefully, with the aim of extracting meaningful information for clinical decision support. In this paper, we leverage interpretable (deep) learning and signal processing tools to deal with multivariate time-series data collected from the Intensive Care Unit (ICU) of the University Hospital of Fuenlabrada (Madrid, Spain). The presence of antimicrobial multidrug-resistant (AMR) bacteria is one of the greatest threats to the health system in general and to the ICUs in particular due to the critical health status of the patients therein. Thus, early identification of bacteria at the ICU and early prediction of their antibiotic resistance are key for the patients’ prognosis. While intelligent data-based processing and learning schemes can contribute to this early prediction, their acceptance and deployment in the ICUs require the automatic schemes to be not only accurate but also understandable by clinicians. Accordingly, we have designed trustworthy intelligent models for the early prediction of AMR based on the combination of meaningful feature selection with interpretable recurrent neural networks. These models were created using irregularly sampled clinical measurements, both considering the health status of the patient and the global ICU environment. We explored several strategies to cope with strongly imbalance data, since only a few ICU patients are infected by AMR bacteria. It is worth noting that our approach exhibits a good balance between performance and interpretability, especially when considering the difficulty of the classification task at hand. A multitude of factors are involved in the emergence of AMR (several of them not fully understood), and the records only contain a subset of them. In addition, the limited number of patients, the imbalance between classes, and the irregularity of the data render the problem harder to solve. Our models are also enriched with SHAP post-hoc interpretability and validated by clinicians who considered model understandability and trustworthiness of paramount concern for pragmatic purposes. Moreover, we use linguistic fuzzy systems to provide clinicians with explanations in natural language. Such explanations are automatically generated from a pool of interpretable rules that describe the interaction among the most relevant features identified by SHAP. Notice tha... |
| abstractGer |
Electronic health records provide rich, heterogeneous data about the evolution of the patients’ health status. However, such data need to be processed carefully, with the aim of extracting meaningful information for clinical decision support. In this paper, we leverage interpretable (deep) learning and signal processing tools to deal with multivariate time-series data collected from the Intensive Care Unit (ICU) of the University Hospital of Fuenlabrada (Madrid, Spain). The presence of antimicrobial multidrug-resistant (AMR) bacteria is one of the greatest threats to the health system in general and to the ICUs in particular due to the critical health status of the patients therein. Thus, early identification of bacteria at the ICU and early prediction of their antibiotic resistance are key for the patients’ prognosis. While intelligent data-based processing and learning schemes can contribute to this early prediction, their acceptance and deployment in the ICUs require the automatic schemes to be not only accurate but also understandable by clinicians. Accordingly, we have designed trustworthy intelligent models for the early prediction of AMR based on the combination of meaningful feature selection with interpretable recurrent neural networks. These models were created using irregularly sampled clinical measurements, both considering the health status of the patient and the global ICU environment. We explored several strategies to cope with strongly imbalance data, since only a few ICU patients are infected by AMR bacteria. It is worth noting that our approach exhibits a good balance between performance and interpretability, especially when considering the difficulty of the classification task at hand. A multitude of factors are involved in the emergence of AMR (several of them not fully understood), and the records only contain a subset of them. In addition, the limited number of patients, the imbalance between classes, and the irregularity of the data render the problem harder to solve. Our models are also enriched with SHAP post-hoc interpretability and validated by clinicians who considered model understandability and trustworthiness of paramount concern for pragmatic purposes. Moreover, we use linguistic fuzzy systems to provide clinicians with explanations in natural language. Such explanations are automatically generated from a pool of interpretable rules that describe the interaction among the most relevant features identified by SHAP. Notice tha... |
| abstract_unstemmed |
Electronic health records provide rich, heterogeneous data about the evolution of the patients’ health status. However, such data need to be processed carefully, with the aim of extracting meaningful information for clinical decision support. In this paper, we leverage interpretable (deep) learning and signal processing tools to deal with multivariate time-series data collected from the Intensive Care Unit (ICU) of the University Hospital of Fuenlabrada (Madrid, Spain). The presence of antimicrobial multidrug-resistant (AMR) bacteria is one of the greatest threats to the health system in general and to the ICUs in particular due to the critical health status of the patients therein. Thus, early identification of bacteria at the ICU and early prediction of their antibiotic resistance are key for the patients’ prognosis. While intelligent data-based processing and learning schemes can contribute to this early prediction, their acceptance and deployment in the ICUs require the automatic schemes to be not only accurate but also understandable by clinicians. Accordingly, we have designed trustworthy intelligent models for the early prediction of AMR based on the combination of meaningful feature selection with interpretable recurrent neural networks. These models were created using irregularly sampled clinical measurements, both considering the health status of the patient and the global ICU environment. We explored several strategies to cope with strongly imbalance data, since only a few ICU patients are infected by AMR bacteria. It is worth noting that our approach exhibits a good balance between performance and interpretability, especially when considering the difficulty of the classification task at hand. A multitude of factors are involved in the emergence of AMR (several of them not fully understood), and the records only contain a subset of them. In addition, the limited number of patients, the imbalance between classes, and the irregularity of the data render the problem harder to solve. Our models are also enriched with SHAP post-hoc interpretability and validated by clinicians who considered model understandability and trustworthiness of paramount concern for pragmatic purposes. Moreover, we use linguistic fuzzy systems to provide clinicians with explanations in natural language. Such explanations are automatically generated from a pool of interpretable rules that describe the interaction among the most relevant features identified by SHAP. Notice tha... |
| collection_details |
GBV_USEFLAG_U GBV_ELV SYSFLAG_U |
| title_short |
Interpretable clinical time-series modeling with intelligent feature selection for early prediction of antimicrobial multidrug resistance |
| url |
https://doi.org/10.1016/j.future.2022.02.021 |
| remote_bool |
true |
| author2 |
Soguero-Ruiz, Cristina Alonso-Moral, Jose M. Mora-Jiménez, Inmaculada Álvarez-Rodríguez, Joaquín Marques, Antonio G. |
| author2Str |
Soguero-Ruiz, Cristina Alonso-Moral, Jose M. Mora-Jiménez, Inmaculada Álvarez-Rodríguez, Joaquín Marques, Antonio G. |
| ppnlink |
ELV004280385 |
| mediatype_str_mv |
z |
| isOA_txt |
false |
| hochschulschrift_bool |
false |
| author2_role |
oth oth oth oth oth |
| doi_str |
10.1016/j.future.2022.02.021 |
| up_date |
2024-07-06T23:23:10.978Z |
| _version_ |
1803873881663995904 |
| fullrecord_marcxml |
<?xml version="1.0" encoding="UTF-8"?><collection xmlns="http://www.loc.gov/MARC21/slim"><record><leader>01000caa a22002652 4500</leader><controlfield tag="001">ELV057490953</controlfield><controlfield tag="003">DE-627</controlfield><controlfield tag="005">20230626045218.0</controlfield><controlfield tag="007">cr uuu---uuuuu</controlfield><controlfield tag="008">220808s2022 xx |||||o 00| ||eng c</controlfield><datafield tag="024" ind1="7" ind2=" "><subfield code="a">10.1016/j.future.2022.02.021</subfield><subfield code="2">doi</subfield></datafield><datafield tag="028" ind1="5" ind2="2"><subfield code="a">/cbs_pica/cbs_olc/import_discovery/elsevier/einzuspielen/GBV00000000001931.pica</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-627)ELV057490953</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(ELSEVIER)S0167-739X(22)00064-4</subfield></datafield><datafield tag="040" ind1=" " ind2=" "><subfield code="a">DE-627</subfield><subfield code="b">ger</subfield><subfield code="c">DE-627</subfield><subfield code="e">rakwb</subfield></datafield><datafield tag="041" ind1=" " ind2=" "><subfield code="a">eng</subfield></datafield><datafield tag="082" ind1="0" ind2="4"><subfield code="a">004</subfield><subfield code="q">VZ</subfield></datafield><datafield tag="084" ind1=" " ind2=" "><subfield code="a">85.35</subfield><subfield code="2">bkl</subfield></datafield><datafield tag="084" ind1=" " ind2=" "><subfield code="a">54.80</subfield><subfield code="2">bkl</subfield></datafield><datafield tag="100" ind1="1" ind2=" "><subfield code="a">Martínez-Agüero, Sergio</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="245" ind1="1" ind2="0"><subfield code="a">Interpretable clinical time-series modeling with intelligent feature selection for early prediction of antimicrobial multidrug resistance</subfield></datafield><datafield tag="264" ind1=" " ind2="1"><subfield code="c">2022transfer abstract</subfield></datafield><datafield tag="300" ind1=" " ind2=" "><subfield code="a">16</subfield></datafield><datafield tag="336" ind1=" " ind2=" "><subfield code="a">nicht spezifiziert</subfield><subfield code="b">zzz</subfield><subfield code="2">rdacontent</subfield></datafield><datafield tag="337" ind1=" " ind2=" "><subfield code="a">nicht spezifiziert</subfield><subfield code="b">z</subfield><subfield code="2">rdamedia</subfield></datafield><datafield tag="338" ind1=" " ind2=" "><subfield code="a">nicht spezifiziert</subfield><subfield code="b">zu</subfield><subfield code="2">rdacarrier</subfield></datafield><datafield tag="520" ind1=" " ind2=" "><subfield code="a">Electronic health records provide rich, heterogeneous data about the evolution of the patients’ health status. However, such data need to be processed carefully, with the aim of extracting meaningful information for clinical decision support. In this paper, we leverage interpretable (deep) learning and signal processing tools to deal with multivariate time-series data collected from the Intensive Care Unit (ICU) of the University Hospital of Fuenlabrada (Madrid, Spain). The presence of antimicrobial multidrug-resistant (AMR) bacteria is one of the greatest threats to the health system in general and to the ICUs in particular due to the critical health status of the patients therein. Thus, early identification of bacteria at the ICU and early prediction of their antibiotic resistance are key for the patients’ prognosis. While intelligent data-based processing and learning schemes can contribute to this early prediction, their acceptance and deployment in the ICUs require the automatic schemes to be not only accurate but also understandable by clinicians. Accordingly, we have designed trustworthy intelligent models for the early prediction of AMR based on the combination of meaningful feature selection with interpretable recurrent neural networks. These models were created using irregularly sampled clinical measurements, both considering the health status of the patient and the global ICU environment. We explored several strategies to cope with strongly imbalance data, since only a few ICU patients are infected by AMR bacteria. It is worth noting that our approach exhibits a good balance between performance and interpretability, especially when considering the difficulty of the classification task at hand. A multitude of factors are involved in the emergence of AMR (several of them not fully understood), and the records only contain a subset of them. In addition, the limited number of patients, the imbalance between classes, and the irregularity of the data render the problem harder to solve. Our models are also enriched with SHAP post-hoc interpretability and validated by clinicians who considered model understandability and trustworthiness of paramount concern for pragmatic purposes. Moreover, we use linguistic fuzzy systems to provide clinicians with explanations in natural language. Such explanations are automatically generated from a pool of interpretable rules that describe the interaction among the most relevant features identified by SHAP. Notice tha...</subfield></datafield><datafield tag="520" ind1=" " ind2=" "><subfield code="a">Electronic health records provide rich, heterogeneous data about the evolution of the patients’ health status. However, such data need to be processed carefully, with the aim of extracting meaningful information for clinical decision support. In this paper, we leverage interpretable (deep) learning and signal processing tools to deal with multivariate time-series data collected from the Intensive Care Unit (ICU) of the University Hospital of Fuenlabrada (Madrid, Spain). The presence of antimicrobial multidrug-resistant (AMR) bacteria is one of the greatest threats to the health system in general and to the ICUs in particular due to the critical health status of the patients therein. Thus, early identification of bacteria at the ICU and early prediction of their antibiotic resistance are key for the patients’ prognosis. While intelligent data-based processing and learning schemes can contribute to this early prediction, their acceptance and deployment in the ICUs require the automatic schemes to be not only accurate but also understandable by clinicians. Accordingly, we have designed trustworthy intelligent models for the early prediction of AMR based on the combination of meaningful feature selection with interpretable recurrent neural networks. These models were created using irregularly sampled clinical measurements, both considering the health status of the patient and the global ICU environment. We explored several strategies to cope with strongly imbalance data, since only a few ICU patients are infected by AMR bacteria. It is worth noting that our approach exhibits a good balance between performance and interpretability, especially when considering the difficulty of the classification task at hand. A multitude of factors are involved in the emergence of AMR (several of them not fully understood), and the records only contain a subset of them. In addition, the limited number of patients, the imbalance between classes, and the irregularity of the data render the problem harder to solve. Our models are also enriched with SHAP post-hoc interpretability and validated by clinicians who considered model understandability and trustworthiness of paramount concern for pragmatic purposes. Moreover, we use linguistic fuzzy systems to provide clinicians with explanations in natural language. Such explanations are automatically generated from a pool of interpretable rules that describe the interaction among the most relevant features identified by SHAP. Notice tha...</subfield></datafield><datafield tag="650" ind1=" " ind2="7"><subfield code="a">37M10</subfield><subfield code="2">Elsevier</subfield></datafield><datafield tag="650" ind1=" " ind2="7"><subfield code="a">68T01</subfield><subfield code="2">Elsevier</subfield></datafield><datafield tag="650" ind1=" " ind2="7"><subfield code="a">94D05</subfield><subfield code="2">Elsevier</subfield></datafield><datafield tag="650" ind1=" " ind2="7"><subfield code="a">68T07</subfield><subfield code="2">Elsevier</subfield></datafield><datafield tag="650" ind1=" " ind2="7"><subfield code="a">92C50</subfield><subfield code="2">Elsevier</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Soguero-Ruiz, Cristina</subfield><subfield code="4">oth</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Alonso-Moral, Jose M.</subfield><subfield code="4">oth</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Mora-Jiménez, Inmaculada</subfield><subfield code="4">oth</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Álvarez-Rodríguez, Joaquín</subfield><subfield code="4">oth</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Marques, Antonio G.</subfield><subfield code="4">oth</subfield></datafield><datafield tag="773" ind1="0" ind2="8"><subfield code="i">Enthalten in</subfield><subfield code="n">Elsevier Science</subfield><subfield code="a">Jiang, Yan-Ping ELSEVIER</subfield><subfield code="t">Surgeon-patient matching based on pairwise comparisons information for elective surgery</subfield><subfield code="d">2020</subfield><subfield code="g">Amsterdam [u.a.]</subfield><subfield code="w">(DE-627)ELV004280385</subfield></datafield><datafield tag="773" ind1="1" ind2="8"><subfield code="g">volume:133</subfield><subfield code="g">year:2022</subfield><subfield code="g">pages:68-83</subfield><subfield code="g">extent:16</subfield></datafield><datafield tag="856" ind1="4" ind2="0"><subfield code="u">https://doi.org/10.1016/j.future.2022.02.021</subfield><subfield code="3">Volltext</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_USEFLAG_U</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ELV</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SYSFLAG_U</subfield></datafield><datafield tag="936" ind1="b" ind2="k"><subfield code="a">85.35</subfield><subfield code="j">Fertigung</subfield><subfield code="q">VZ</subfield></datafield><datafield tag="936" ind1="b" ind2="k"><subfield code="a">54.80</subfield><subfield code="j">Angewandte Informatik</subfield><subfield code="q">VZ</subfield></datafield><datafield tag="951" ind1=" " ind2=" "><subfield code="a">AR</subfield></datafield><datafield tag="952" ind1=" " ind2=" "><subfield code="d">133</subfield><subfield code="j">2022</subfield><subfield code="h">68-83</subfield><subfield code="g">16</subfield></datafield></record></collection>
|
| score |
7.4009905 |