Detection and classification of abnormal red blood cells with computational intelligence techniques: a review

Red blood cells (RBCs), or erythrocytes, are usually disc-shaped. However, pathological conditions can change their shape. The complete blood count (CBC) is a test that can detect abnormal RBCs. Yet it is a manual test susceptible to errors, so there are efforts to automate the detection and classification of abnormal cells. A total of 31 papers were reviewed, and all the selected studies focus on the detection or classification of abnormal RBCs. Different approaches were applied to tackle this issue, including image processing techniques, classification using machine learning, and convolutional neural networks for detection, among other methods. Furthermore, machine learning techniques are recently presenting promising results for abnormal RBCs detection and classification. This review also brings a discussion on the computational intelligence methodologies applied.