Currently, Chronic Kidney Disease (CKD) is experiencing a globally increasing incidence and high cost to health systems. A delayed recognition implies premature mortality due to progressive loss of kidney function. The employment of data mining to discover subtle patterns in CKD indicators would contribute achieving early diagnosis. This work presents the development and evaluation of an explainable prediction model that would support clinicians in the early diagnosis of CKD patients. The model development is based on a data management pipeline that detects the best combination of ensemble trees algorithms and features selected concerning classification performance. The results obtained through the pipeline equals the performance of the best CKD prediction models identified in the literature. Furthermore, the main contribution of the paper involves an explainability-driven approach that allows selecting the best prediction model maintaining a balance between accuracy and explainability. Therefore, the most balanced explainable prediction model of CKD implements an XGBoost classifier over a group of 4 features (packed cell value, specific gravity, albumin, and hypertension), achieving an accuracy of 98.9% and 97.5% with cross-validation technique and with new unseen data respectively. In addition, by analysing the models explainability by means of different post-hoc techniques, the packed cell value and the specific gravity are determined as the most relevant features that influence the prediction results of the model. This small number of feature selected results in a reduced cost of the early diagnosis of CKD implying a promising solution for developing countries.
تحميل البحث