Modeling and prediction of pressure injury in hospitalized patients using artificial intelligence
Abstract Background Hospital-acquired pressure injuries (PIs) induce significant patient suffering, inflate healthcare costs, and increase clinical co-morbidities. PIs are mostly due to bed-immobility, sensory impairment, bed positioning, and length of hospital stay. In this study, we use electronic...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2021-08-01
|
| Series: | BMC Medical Informatics and Decision Making |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12911-021-01608-5 |
| _version_ | 1818408779418435584 |
|---|---|
| author | Christine Anderson Zerihun Bekele Yongkai Qiu Dana Tschannen Ivo D. Dinov |
| author_facet | Christine Anderson Zerihun Bekele Yongkai Qiu Dana Tschannen Ivo D. Dinov |
| author_sort | Christine Anderson |
| collection | DOAJ |
| description | Abstract Background Hospital-acquired pressure injuries (PIs) induce significant patient suffering, inflate healthcare costs, and increase clinical co-morbidities. PIs are mostly due to bed-immobility, sensory impairment, bed positioning, and length of hospital stay. In this study, we use electronic health records and administrative data to examine the contributing factors to PI development using artificial intelligence (AI). Methods We used advanced data science techniques to first preprocess the data and then train machine learning classifiers to predict the probability of developing PIs. The AI training was based on large, incongruent, incomplete, heterogeneous, and time-varying data of hospitalized patients. Both model-based statistical methods and model-free AI strategies were used to forecast PI outcomes and determine the salient features that are highly predictive of the outcomes. Results Our findings reveal that PI prediction by model-free techniques outperform model-based forecasts. The performance of all AI methods is improved by rebalancing the training data and by including the Braden in the model learning phase. Compared to neural networks and linear modeling, with and without rebalancing or using Braden scores, Random forest consistently generated the optimal PI forecasts. Conclusions AI techniques show promise to automatically identify patients at risk for hospital acquired PIs in different surgical services. Our PI prediction model provide a first generation of AI guidance to prescreen patients at risk for developing PIs. Clinical impact This study provides a foundation for designing, implementing, and assessing novel interventions addressing specific healthcare needs. Specifically, this approach allows examining the impact of various dynamic, personalized, and clinical-environment effects on PI prevention for hospital patients receiving care from various surgical services. |
| first_indexed | 2024-12-14T09:49:09Z |
| format | Article |
| id | doaj.art-725a6b1fb47f442298472135bd858e9d |
| institution | Directory Open Access Journal |
| issn | 1472-6947 |
| language | English |
| last_indexed | 2024-12-14T09:49:09Z |
| publishDate | 2021-08-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Medical Informatics and Decision Making |
| spelling | doaj.art-725a6b1fb47f442298472135bd858e9d2022-12-21T23:07:33ZengBMCBMC Medical Informatics and Decision Making1472-69472021-08-0121111310.1186/s12911-021-01608-5Modeling and prediction of pressure injury in hospitalized patients using artificial intelligenceChristine Anderson0Zerihun Bekele1Yongkai Qiu2Dana Tschannen3Ivo D. Dinov4School of Nursing, University of MichiganStatistics Online Computational Resource (SOCR), University of MichiganDepartment of Applied and Computational Mathematics and Statistics, University of Notre DameSchool of Nursing, University of MichiganSchool of Nursing, University of MichiganAbstract Background Hospital-acquired pressure injuries (PIs) induce significant patient suffering, inflate healthcare costs, and increase clinical co-morbidities. PIs are mostly due to bed-immobility, sensory impairment, bed positioning, and length of hospital stay. In this study, we use electronic health records and administrative data to examine the contributing factors to PI development using artificial intelligence (AI). Methods We used advanced data science techniques to first preprocess the data and then train machine learning classifiers to predict the probability of developing PIs. The AI training was based on large, incongruent, incomplete, heterogeneous, and time-varying data of hospitalized patients. Both model-based statistical methods and model-free AI strategies were used to forecast PI outcomes and determine the salient features that are highly predictive of the outcomes. Results Our findings reveal that PI prediction by model-free techniques outperform model-based forecasts. The performance of all AI methods is improved by rebalancing the training data and by including the Braden in the model learning phase. Compared to neural networks and linear modeling, with and without rebalancing or using Braden scores, Random forest consistently generated the optimal PI forecasts. Conclusions AI techniques show promise to automatically identify patients at risk for hospital acquired PIs in different surgical services. Our PI prediction model provide a first generation of AI guidance to prescreen patients at risk for developing PIs. Clinical impact This study provides a foundation for designing, implementing, and assessing novel interventions addressing specific healthcare needs. Specifically, this approach allows examining the impact of various dynamic, personalized, and clinical-environment effects on PI prevention for hospital patients receiving care from various surgical services.https://doi.org/10.1186/s12911-021-01608-5Health analyticsPrecision medicineClinical assessmentHuman–machine intelligence |
| spellingShingle | Christine Anderson Zerihun Bekele Yongkai Qiu Dana Tschannen Ivo D. Dinov Modeling and prediction of pressure injury in hospitalized patients using artificial intelligence BMC Medical Informatics and Decision Making Health analytics Precision medicine Clinical assessment Human–machine intelligence |
| title | Modeling and prediction of pressure injury in hospitalized patients using artificial intelligence |
| title_full | Modeling and prediction of pressure injury in hospitalized patients using artificial intelligence |
| title_fullStr | Modeling and prediction of pressure injury in hospitalized patients using artificial intelligence |
| title_full_unstemmed | Modeling and prediction of pressure injury in hospitalized patients using artificial intelligence |
| title_short | Modeling and prediction of pressure injury in hospitalized patients using artificial intelligence |
| title_sort | modeling and prediction of pressure injury in hospitalized patients using artificial intelligence |
| topic | Health analytics Precision medicine Clinical assessment Human–machine intelligence |
| url | https://doi.org/10.1186/s12911-021-01608-5 |
| work_keys_str_mv | AT christineanderson modelingandpredictionofpressureinjuryinhospitalizedpatientsusingartificialintelligence AT zerihunbekele modelingandpredictionofpressureinjuryinhospitalizedpatientsusingartificialintelligence AT yongkaiqiu modelingandpredictionofpressureinjuryinhospitalizedpatientsusingartificialintelligence AT danatschannen modelingandpredictionofpressureinjuryinhospitalizedpatientsusingartificialintelligence AT ivoddinov modelingandpredictionofpressureinjuryinhospitalizedpatientsusingartificialintelligence |