Interpretable generalized neural additive models for mortality prediction of COVID-19 hospitalized patients in Hamadan, Iran
Abstract Background The high number of COVID-19 deaths is a serious threat to the world. Demographic and clinical biomarkers are significantly associated with the mortality risk of this disease. This study aimed to implement Generalized Neural Additive Model (GNAM) as an interpretable machine learni...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2022-12-01
|
| Series: | BMC Medical Research Methodology |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12874-022-01827-y |
| _version_ | 1797973553564352512 |
|---|---|
| author | Samad Moslehi Hossein Mahjub Maryam Farhadian Ali Reza Soltanian Mojgan Mamani |
| author_facet | Samad Moslehi Hossein Mahjub Maryam Farhadian Ali Reza Soltanian Mojgan Mamani |
| author_sort | Samad Moslehi |
| collection | DOAJ |
| description | Abstract Background The high number of COVID-19 deaths is a serious threat to the world. Demographic and clinical biomarkers are significantly associated with the mortality risk of this disease. This study aimed to implement Generalized Neural Additive Model (GNAM) as an interpretable machine learning method to predict the COVID-19 mortality of patients. Methods This cohort study included 2181 COVID-19 patients admitted from February 2020 to July 2021 in Sina and Besat hospitals in Hamadan, west of Iran. A total of 22 baseline features including patients' demographic information and clinical biomarkers were collected. Four strategies including removing missing values, mean, K-Nearest Neighbor (KNN), and Multivariate Imputation by Chained Equations (MICE) imputation methods were used to deal with missing data. Firstly, the important features for predicting binary outcome (1: death, 0: recovery) were selected using the Random Forest (RF) method. Also, synthetic minority over-sampling technique (SMOTE) method was used for handling imbalanced data. Next, considering the selected features, the predictive performance of GNAM for predicting mortality outcome was compared with logistic regression, RF, generalized additive model (GAMs), gradient boosting decision tree (GBDT), and deep neural networks (DNNs) classification models. Each model trained on fifty different subsets of a train-test dataset to ensure a model performance. The average accuracy, F1-score and area under the curve (AUC) evaluation indices were used for comparison of the predictive performance of the models. Results Out of the 2181 COVID-19 patients, 624 died during hospitalization and 1557 recovered. The missing rate was 3 percent for each patient. The mean age of dead patients (71.17 ± 14.44 years) was statistically significant higher than recovered patients (58.25 ± 16.52 years). Based on RF, 10 features with the highest relative importance were selected as the best influential features; including blood urea nitrogen (BUN), lymphocytes (Lym), age, blood sugar (BS), serum glutamic-oxaloacetic transaminase (SGOT), monocytes (Mono), blood creatinine (CR), neutrophils (NUT), alkaline phosphatase (ALP) and hematocrit (HCT). The results of predictive performance comparisons showed GNAM with the mean accuracy, F1-score, and mean AUC in the test dataset of 0.847, 0.691, and 0.774, respectively, had the best performance. The smooth function graphs learned from the GNAM were descending for the Lym and ascending for the other important features. Conclusions Interpretable GNAM can perform well in predicting the mortality of COVID-19 patients. Therefore, the use of such a reliable model can help physicians to prioritize some important demographic and clinical biomarkers by identifying the effective features and the type of predictive trend in disease progression. |
| first_indexed | 2024-04-11T04:06:06Z |
| format | Article |
| id | doaj.art-353678e8e4aa479bb60b45dc40361abf |
| institution | Directory Open Access Journal |
| issn | 1471-2288 |
| language | English |
| last_indexed | 2024-04-11T04:06:06Z |
| publishDate | 2022-12-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Medical Research Methodology |
| spelling | doaj.art-353678e8e4aa479bb60b45dc40361abf2023-01-01T12:22:17ZengBMCBMC Medical Research Methodology1471-22882022-12-0122111410.1186/s12874-022-01827-yInterpretable generalized neural additive models for mortality prediction of COVID-19 hospitalized patients in Hamadan, IranSamad Moslehi0Hossein Mahjub1Maryam Farhadian2Ali Reza Soltanian3Mojgan Mamani4Department of Biostatistics, School of Public Health, Hamadan University of Medical SciencesDepartment of Biostatistics, School of Public Health, Research Center for Health Sciences, Hamadan University of Medical SciencesDepartment of Biostatistics, School of Public Health, Research Center for Health Sciences, Hamadan University of Medical SciencesDepartment of Biostatistics, School of Public Health, Modeling of Noncommunicable Diseases Research Center, Hamadan University of Medical SciencesBrucellosis Research Center, Hamadan University of Medical SciencesAbstract Background The high number of COVID-19 deaths is a serious threat to the world. Demographic and clinical biomarkers are significantly associated with the mortality risk of this disease. This study aimed to implement Generalized Neural Additive Model (GNAM) as an interpretable machine learning method to predict the COVID-19 mortality of patients. Methods This cohort study included 2181 COVID-19 patients admitted from February 2020 to July 2021 in Sina and Besat hospitals in Hamadan, west of Iran. A total of 22 baseline features including patients' demographic information and clinical biomarkers were collected. Four strategies including removing missing values, mean, K-Nearest Neighbor (KNN), and Multivariate Imputation by Chained Equations (MICE) imputation methods were used to deal with missing data. Firstly, the important features for predicting binary outcome (1: death, 0: recovery) were selected using the Random Forest (RF) method. Also, synthetic minority over-sampling technique (SMOTE) method was used for handling imbalanced data. Next, considering the selected features, the predictive performance of GNAM for predicting mortality outcome was compared with logistic regression, RF, generalized additive model (GAMs), gradient boosting decision tree (GBDT), and deep neural networks (DNNs) classification models. Each model trained on fifty different subsets of a train-test dataset to ensure a model performance. The average accuracy, F1-score and area under the curve (AUC) evaluation indices were used for comparison of the predictive performance of the models. Results Out of the 2181 COVID-19 patients, 624 died during hospitalization and 1557 recovered. The missing rate was 3 percent for each patient. The mean age of dead patients (71.17 ± 14.44 years) was statistically significant higher than recovered patients (58.25 ± 16.52 years). Based on RF, 10 features with the highest relative importance were selected as the best influential features; including blood urea nitrogen (BUN), lymphocytes (Lym), age, blood sugar (BS), serum glutamic-oxaloacetic transaminase (SGOT), monocytes (Mono), blood creatinine (CR), neutrophils (NUT), alkaline phosphatase (ALP) and hematocrit (HCT). The results of predictive performance comparisons showed GNAM with the mean accuracy, F1-score, and mean AUC in the test dataset of 0.847, 0.691, and 0.774, respectively, had the best performance. The smooth function graphs learned from the GNAM were descending for the Lym and ascending for the other important features. Conclusions Interpretable GNAM can perform well in predicting the mortality of COVID-19 patients. Therefore, the use of such a reliable model can help physicians to prioritize some important demographic and clinical biomarkers by identifying the effective features and the type of predictive trend in disease progression.https://doi.org/10.1186/s12874-022-01827-yCOVID-19Feature selectionLaboratory markersMachine learningGeneralized neural additivePrediction |
| spellingShingle | Samad Moslehi Hossein Mahjub Maryam Farhadian Ali Reza Soltanian Mojgan Mamani Interpretable generalized neural additive models for mortality prediction of COVID-19 hospitalized patients in Hamadan, Iran BMC Medical Research Methodology COVID-19 Feature selection Laboratory markers Machine learning Generalized neural additive Prediction |
| title | Interpretable generalized neural additive models for mortality prediction of COVID-19 hospitalized patients in Hamadan, Iran |
| title_full | Interpretable generalized neural additive models for mortality prediction of COVID-19 hospitalized patients in Hamadan, Iran |
| title_fullStr | Interpretable generalized neural additive models for mortality prediction of COVID-19 hospitalized patients in Hamadan, Iran |
| title_full_unstemmed | Interpretable generalized neural additive models for mortality prediction of COVID-19 hospitalized patients in Hamadan, Iran |
| title_short | Interpretable generalized neural additive models for mortality prediction of COVID-19 hospitalized patients in Hamadan, Iran |
| title_sort | interpretable generalized neural additive models for mortality prediction of covid 19 hospitalized patients in hamadan iran |
| topic | COVID-19 Feature selection Laboratory markers Machine learning Generalized neural additive Prediction |
| url | https://doi.org/10.1186/s12874-022-01827-y |
| work_keys_str_mv | AT samadmoslehi interpretablegeneralizedneuraladditivemodelsformortalitypredictionofcovid19hospitalizedpatientsinhamadaniran AT hosseinmahjub interpretablegeneralizedneuraladditivemodelsformortalitypredictionofcovid19hospitalizedpatientsinhamadaniran AT maryamfarhadian interpretablegeneralizedneuraladditivemodelsformortalitypredictionofcovid19hospitalizedpatientsinhamadaniran AT alirezasoltanian interpretablegeneralizedneuraladditivemodelsformortalitypredictionofcovid19hospitalizedpatientsinhamadaniran AT mojganmamani interpretablegeneralizedneuraladditivemodelsformortalitypredictionofcovid19hospitalizedpatientsinhamadaniran |