Quaternion-Valued Convolutional Neural Network Applied for Acute Lymphoblastic Leukemia Diagnosis

The field of neural networks has seen significant advances in recent years with the development deep and convolutional networks. Although many current works address real-valued models, studies reveal that hypercomplex-valued parameters can better capture, generalize, represent complexity multidimensional data. This paper explores quaternion-valued network application for a pattern recognition task from medicine, namely, diagnosis acute lymphoblastic leukemia. Precisely, we compare performance to classify lymphoblasts peripheral blood smear microscopic images. achieved or similar than its corresponding but using only 34% parameters. result confirms quaternion algebra allows capturing extracting information color image fewer