Enhancing Generalization in Sickle Cell Disease Diagnosis through Ensemble Methods and Feature Importance Analysis
This work provides incremental improvements in diagnostic support for Sickle Cell Disease patients through better generalization in medical imaging.
This paper tackles the problem of diagnosing Sickle Cell Disease from blood smear images by developing an ensemble method with feature importance analysis, achieving an F1-score of 90.71% and SDS-score of 93.33%, outperforming previous methods.
This work presents a novel approach for selecting the optimal ensemble-based classification method and features with a primarly focus on achieving generalization, based on the state-of-the-art, to provide diagnostic support for Sickle Cell Disease using peripheral blood smear images of red blood cells. We pre-processed and segmented the microscopic images to ensure the extraction of high-quality features. To ensure the reliability of our proposed system, we conducted an in-depth analysis of interpretability. Leveraging techniques established in the literature, we extracted features from blood cells and employed ensemble machine learning methods to classify their morphology. Furthermore, we have devised a methodology to identify the most critical features for classification, aimed at reducing complexity and training time and enhancing interpretability in opaque models. Lastly, we validated our results using a new dataset, where our model overperformed state-of-the-art models in terms of generalization. The results of classifier ensembled of Random Forest and Extra Trees classifier achieved an harmonic mean of precision and recall (F1-score) of 90.71\% and a Sickle Cell Disease diagnosis support score (SDS-score) of 93.33\%. These results demonstrate notable enhancement from previous ones with Gradient Boosting classifier (F1-score 87.32\% and SDS-score 89.51\%). To foster scientific progress, we have made available the parameters for each model, the implemented code library, and the confusion matrices with the raw data.