TCKAN:A Novel Integrated Network Model for Predicting Mortality Risk in Sepsis Patients

TL;DR

The results confirm the model’s effectiveness and its potential to transform patient management and treatment optimization in clinical practice, as it surpasses existing machine learning and deep learning methods in accuracy, sensitivity, and specificity.

Abstract

Sepsis poses a major global health threat, accounting for millions of deaths annually and significant economic costs. Accurately predicting the risk of mortality in sepsis patients enables early identification, promotes the efficient allocation of medical resources, and facilitates timely interventions, thereby improving patient outcomes. Current methods typically utilize only one type of data--either constant, temporal, or ICD codes. This study introduces a novel approach, the Time-Constant Kolmogorov-Arnold Network (TCKAN), which uniquely integrates temporal data, constant data, and ICD codes within a single predictive model. Unlike existing methods that typically rely on one type of data, TCKAN leverages a multi-modal data integration strategy, resulting in superior predictive accuracy and robustness in identifying high-risk sepsis patients. Validated against the MIMIC-III and MIMIC-IV datasets, TCKAN surpasses existing machine learning and deep learning methods in accuracy, sensitivity, and specificity. Notably, TCKAN achieved AUCs of 87.76% and 88.07%, demonstrating superior capability in identifying high-risk patients. Additionally, TCKAN effectively combats the prevalent issue of data imbalance in clinical settings, improving the detection of patients at elevated risk of mortality and facilitating timely interventions. These results confirm the model's effectiveness and its potential to transform patient management and treatment optimization in clinical practice. Although the TCKAN model has already incorporated temporal, constant, and ICD code data, future research could include more diverse medical data types, such as imaging and laboratory test results, to achieve a more comprehensive data integration and further improve predictive accuracy.

Figures (10)

• Figure 1: Main flowchart of the TCKAN
• Figure 2: Process for extracting data from MIMIC-III and MIMIC-IV
• Figure 3: Sequential measurements of temperature and mean blood pressure (MBP) are recorded at specified time intervals
• Figure 4: The gated recurrent unit with decay (GRU-D) mechanism, highlighting the interactions between input data, hidden states, and decay factors. The symbols $\gamma$, $h$, $Z$, and $r$ represent decay factors, hidden state updates, update gates, and reset gates, respectively.
• Figure 5: The architectural layout of the Kolmogorov–Arnold Network (KAN), the network's unique structure that employs learnable B-spline activation functions.