User login
Scientists say they have developed an automated system that can identify shapes of red blood cells (RBCs).
The team found their system could classify sickled RBCs “with high accuracy,” which suggests it could be used to help monitor patients with sickle cell disease.
“We have developed the first deep learning tool that can automatically identify and classify red blood cell alteration, hence providing direct quantitative evidence of the severity of the disease,” said George Karniadakis, PhD, of Brown University in Providence, Rhode Island.
Dr Karniadakis and his colleagues described their tool in PLOS Computational Biology.
The researchers wanted to automate the process of identifying RBC shape. So they developed a computational framework that employs a machine-learning tool known as a deep convolutional neural network (CNN).
The framework uses 3 steps to classify the shapes of RBCs in microscopic images of blood.
First, it distinguishes RBCs from the background of each image and from each other. Then, for each cell detected, it zooms in or out until all cell images are a uniform size. Finally, it uses deep CNNs to categorize RBCs by shape.
The researchers validated their new tool using 7000 microscopy images from 8 patients with sickle cell disease. The method successfully classified RBC shape for both oxygenated and deoxygenated cells.
The researchers plan to further improve their deep CNN tool and test it in other diseases that alter the shape and size of RBCs, such as diabetes and HIV. They also plan to explore its usefulness in characterizing cancer cells. 
Scientists say they have developed an automated system that can identify shapes of red blood cells (RBCs).
The team found their system could classify sickled RBCs “with high accuracy,” which suggests it could be used to help monitor patients with sickle cell disease.
“We have developed the first deep learning tool that can automatically identify and classify red blood cell alteration, hence providing direct quantitative evidence of the severity of the disease,” said George Karniadakis, PhD, of Brown University in Providence, Rhode Island.
Dr Karniadakis and his colleagues described their tool in PLOS Computational Biology.
The researchers wanted to automate the process of identifying RBC shape. So they developed a computational framework that employs a machine-learning tool known as a deep convolutional neural network (CNN).
The framework uses 3 steps to classify the shapes of RBCs in microscopic images of blood.
First, it distinguishes RBCs from the background of each image and from each other. Then, for each cell detected, it zooms in or out until all cell images are a uniform size. Finally, it uses deep CNNs to categorize RBCs by shape.
The researchers validated their new tool using 7000 microscopy images from 8 patients with sickle cell disease. The method successfully classified RBC shape for both oxygenated and deoxygenated cells.
The researchers plan to further improve their deep CNN tool and test it in other diseases that alter the shape and size of RBCs, such as diabetes and HIV. They also plan to explore its usefulness in characterizing cancer cells.
Scientists say they have developed an automated system that can identify shapes of red blood cells (RBCs).
The team found their system could classify sickled RBCs “with high accuracy,” which suggests it could be used to help monitor patients with sickle cell disease.
“We have developed the first deep learning tool that can automatically identify and classify red blood cell alteration, hence providing direct quantitative evidence of the severity of the disease,” said George Karniadakis, PhD, of Brown University in Providence, Rhode Island.
Dr Karniadakis and his colleagues described their tool in PLOS Computational Biology.
The researchers wanted to automate the process of identifying RBC shape. So they developed a computational framework that employs a machine-learning tool known as a deep convolutional neural network (CNN).
The framework uses 3 steps to classify the shapes of RBCs in microscopic images of blood.
First, it distinguishes RBCs from the background of each image and from each other. Then, for each cell detected, it zooms in or out until all cell images are a uniform size. Finally, it uses deep CNNs to categorize RBCs by shape.
The researchers validated their new tool using 7000 microscopy images from 8 patients with sickle cell disease. The method successfully classified RBC shape for both oxygenated and deoxygenated cells.
The researchers plan to further improve their deep CNN tool and test it in other diseases that alter the shape and size of RBCs, such as diabetes and HIV. They also plan to explore its usefulness in characterizing cancer cells.