Interpretable Machine Learning for Inpatient COVID-19 Mortality Risk Assessments: Diabetes Mellitus Exclusive Interplay

Abstract

People with diabetes mellitus (DM) are at elevated risk of in-hospital mortality from coronavirus disease-2019 (COVID-19). This vulnerability has spurred efforts to pinpoint distinctive characteristics of COVID-19 patients with DM. In this context, the present article develops ML models equipped with interpretation modules for inpatient mortality risk assessments of COVID-19 patients with DM. To this end, a cohort of 156 hospitalised COVID-19 patients with pre-existing DM is studied. For creating risk assessment platforms, this work explores a pool of historical, on-admission, and during-admission data that are DM-related or, according to preliminary investigations, are exclusively attributed to the COVID-19 susceptibility of DM patients. First, a set of careful pre-modelling steps are executed on the clinical data, including cleaning, pre-processing, subdivision, and feature elimination. Subsequently, standard machine learning (ML) modelling analysis is performed on the cured data. Initially, a classifier is tasked with forecasting COVID-19 fatality from selected features. The model undergoes thorough evaluation analysis. The results achieved substantiate the efficacy of the undertaken data curation and modelling steps. Afterwards, SHapley Additive exPlanations (SHAP) technique is assigned to interpret the generated mortality risk prediction model by rating the predictors' global and local influence on the model's outputs. These interpretations advance the comprehensibility of the analysis by explaining the formation of outcomes and, in this way, foster the adoption of the proposed methodologies. Next, a clustering algorithm demarcates patients into four separate groups based on their SHAP values, providing a practical risk stratification method. Finally, a re-evaluation analysis is performed to verify the robustness of the proposed framework.

Keywords: COVID-19; SHAP; diabetes mellitus; machine learning.

Figure 1

A schematic of elbow analysis operated to decide the number of clusters.

Figure 2

Global interpretation plots for the developed inpatient COVID-19 mortality prediction model. (A) Bee swarm SHAP values plot, (B) SHAP summary importance plot. The bee swarm plot shows all SHAP values in accord with predictors values. The summary plot presents predictors in descending order based on their overall importance on the model’s outcomes derived from mean absolute SHAP values. Note. BGL: blood glucose level; CRP: c-reactive protein; FI: first inpatient; FiO2: fraction of inspired oxygen; HR: highest requirement; HV: highest value; LaV: last value; LAYBA: latest available within one year before admission; LV: lowest value; LYM: lymphocytes; MN: monocytes; NEUT: neutrophils; OA: on admission; O2: oxygen; PT: prothrombin time; RR: respiratory rate; SHAP: SHapley Additive exPlanations.

Figure 3

The local interpretation plots for the developed inpatient COVID-19 mortality prediction model. (A) an example of patients with survival as the outcome of admission, (B) an example of patients with death as the outcome of admission. The plots start from the bottom with a predefined prediction for the risk of death equal to the average death rate in the training set. Next, the arrows with an ascending order show how each feature has contributed to the formation of a final prediction specified for the given data instance shown at the top of the plot. Note. BGL: blood glucose level; CRP: c-reactive protein; FI: first inpatient; FiO2: fraction of inspired oxygen; HR: highest requirement; HV: highest value; LaV: last value; LAYBA: latest available within one year before admission; LV: lowest value; LYM: lymphocytes; MN: monocytes; NEUT: neutrophils; OA: on admission; O2: oxygen; PT: prothrombin time; RR: respiratory rate; SHAP: SHapley Additive exPlanations.

Figure 4

Global interpretation plots for the updated inpatient COVID-19 mortality prediction model. (A) Bee swarm SHAP values plot, (B) SHAP summary importance plot. The bee swarm plot shows all SHAP values in accord with predictors values. The summary plot presents predictors in descending order based on their overall importance on the model’s outcomes derived from mean absolute SHAP values. Note. BGV: blood gas value; Cl: chloride; CRP: c-reactive protein; DM: diabetes mellitus; FI: first inpatient; FiO2: fraction of inspired oxygen; HR: highest requirement; HV: highest value; LaV: last value; LV: lowest value; LYM: lymphocytes; Na: sodium; NEUT: neutrophils; O2: oxygen; RR: respiratory rate; SBP: systolic blood pressure; SHAP: SHapley Additive exPlanations.