A Machine learning model for differentiations of cells in the bone marrow

Project Summary

The aim of this project is to develop a machine learning model that can accurately differentiate various types of cells in the bone marrow. The bone marrow is a complex tissue composed of different cell types, including hematopoietic stem cells, myeloid cells, and lymphoid cells. Abnormalities in the distribution and morphology of these cells can indicate various diseases.

Current methods for identifying and differentiating these cells in the bone marrow require manual inspection by trained hematopathologists, which can be time-consuming and subject to inter-observer variability. By leveraging machine learning techniques, we aim to develop a model that can automate this process and provide accurate and consistent cell differentiation.

Current projects use CNN to build the models which has limited efficacy when the data is unbalances. The promise of this project is to use novel algorithms such as transformer algorithms.

Dataset

⁠

https://wiki.cancerimagingarchive.net/pages/viewpage.action?pageId=101941770⁠

⁠

Loading ashpublications.org⁠

⁠

Project outline

We will use the above publicly available dataset as a training cohort for the model.

We will use state of the art transformer image algorithms as well as zero-shot algorithms to train the new model

We will use several algorithms to explain the model output

We will evaluate the model's accuracy using metrics such as precision, recall, and F1 score, and compare it to existing state-of-the-art approaches.

Want to print your doc?
This is not the way.

Try clicking the ⋯ next to your doc name or using a keyboard shortcut (

CtrlP

) instead.