Gender Based Analyzing Iron Deficiency Anemia Using Sequential Random Splitting in Machine Learning Model

There are a lot of physical and visual indications of anaemia, making the diagnostic process laborious and resource intensive. There are various kinds of anaemia, each of which can be identified by a unique set of symptoms. Complete blood count (CBC) tests can detect anaemia, but they can't tell you what form of anaemia you have because they're so fast, cheap, and easy to get. Consequently, to set a benchmark for the kind of anaemia in a patient, additional tests are necessary. These tests are not commonly performed in smaller healthcare facilities due to the high cost of the equipment needed. Iron deficiency anaemia (IDA), and combination anaemias can be hard to distinguish, even if there are several RBC formulas and indices with different ideal cutoff levels. This is because there are multiple forms of anaemia in people, and differentiating between BTT, IDA, HbE, and combinations of these forms can be challenging. To assist clinicians in identifying IDA efficiently, here proposed an automated prediction model that demonstrates promising accuracy and specificity based on our dataset. In this paper, proposed an algorithm to detect the Iron Deficiency Anemia. This was followed by the measurement of the performance using the confusion matrix and 15,300 data representing the five classes of anemia, and the results showed 92.71% accuracy, 89% sensitivity, 85% precision, and F1 score of 87%.