A DATASET AT A GLANCE...

A Dataset for Bone Marrow Cell Classification

Do you know that the site responsible for blood cell generation, be it white, red or platelets, is the bone marrow? To define it, it is a semi-solid tissue that is located within the cancellous (spongy) bones that generate blood cells. However, excess or deficit generation because of bone marrow tissue abnormality cannot be withstood, since it might result in various diseases that range from the curable iron deficiency anaemia to the fatal leukaemia or aplastic anaemia. Additionally, as per a recent research, bone marrow abnormality and the more common cardiovascular diseases are inter-linked. Hence, it is very essential to monitor the proper functioning of this tissue to keep the individuals away from fatal diseases, which are caused by blood cell abundancy or scarcity. The procedure that allows collection and examination of bone marrow is the bone marrow biopsy. However manual examination cannot be always precise. Hence, research on automatic diagnosis of various bone marrow abnormalities is encouraged for application on different patients with differing abnormalities. A researcher intending to serve this purpose can make use of the “Bone Marrow Cell Classification” dataset from Kaggle (https://www.kaggle.com/datasets/andrewmvd/bone-marrow-cell-classification). This publicly-available dataset is collected from 945 patients and there are over 170000 de-identified, expert-annotated cells of bone marrow smears. The processing, the scanning and the post-processing of samples were done at the Munich Leukemia Laboratory (MLL) and their acquisition was done using a brightfield microscope with 40x magnification and oil immersion. The researchers attempting to use this dataset have the citation request as: (i) Matek, C., Krappe, S., Münzenmayer, C., Haferlach, T., & Marr, C. (2021). An Expert-Annotated Dataset of Bone Marrow Cytology in Hematologic Malignancies [Data set]. The Cancer Imaging Archive. https://doi.org/10.7937/TCIA.AXH3-T579 and (ii) Matek, C., Krappe, S., Münzenmayer, C., Haferlach, T., and Marr, C. (2021). Highly accurate differentiation of bone marrow cell morphologies using deep neural networks on a large image dataset. https://doi.org/10.1182/blood.2020010568.

Image Courtesy: www.scientificanimations.com