Automated Assessment of Erythrocyte Disorders Using Artificial Neural Network

Abstract

In this paper, we employ artificial neural network (ANN) together with image analysis techniques to automate the assessment of erythrocyte disorders using blood parameters such as red blood cell (RBC) count, hemoglobin (Hgb) level, and mean corpuscular hemoglobin (MCH). The neural network is trained using 800 blood sample images collected from the Prince George-BC, Hospital. The images are captured using a high-resolution digital camera mounted on a microscope. The red, green, and blue values of each image are fed as the input of the neural network. The Hospital RBC, Hgb values of the samples measured using hydrodynamic focused analyzer (CELL-DYN 3200 System) are provided as the target values during training. Several variations of the back propagationlearning algorithm were applied for training. The trained network is tested against 200 blood samples. The output results are compared with those of Hospital laboratory and found to be near identical, most of which are within 5% margin of error, and are much significantly better than those published. The proposed method is simple, fast, accurate, and can be a crucial step in automating laboratory reporting.

A Novel Regression Based Model for Detecting Anemia Using Color Microscopic Blood Images

“…None of the published paper attempted to model Hgb via regression models and images analysis. Zahir and Chowdhry [7] have presented a combined method that is based on artificial neural network (ANN) in conjunction with color image analysis. They presented results that are far better than those reported in [8].…”

Analysis of red blood cells from peripheral blood smear images for anemia detection: a methodological review

“…Zahir et al [162] presented an ANN-based method to detect RBC disorders anemia and polycythemia using Hb value, MCH and RBC count and obtained significant results for more than 90% of the 1000 blood samples with training time less than 15 minutes. Bacus et al [36,37] presented RBC classification method by extracting the features and obtained correlation coefficient of 0.965 for 100 cells from 4 specimens.…”