In recent years as science and technology has progressed, deep learning has emerged as an important tool in multiple fields to solve challenging problems. Deep learning allows to derive information from a vast amount of data in a relatively short period, which wasn't possible previously. Also, deep learning avoids the need for manual extraction of features, which reduces the burden from the user as well removes the possibility of error or biases during the feature extraction. These techniques have the potential for medical imaging, diagnostics, and medical data analysis.
In drug discovery process, cell-based assays are widely used. Historically, two dimensional (2D) monolayer cell cultures are being used to estimate drug responses. Though, it is been established that 2D cultures have limitations as they cannot mimic the tissue-specific architecture, mechanical and cell-to-cell interaction, which hinders the drug responses for some diseases like cancer. To overcome this issue, nowadays 3D cell cultures are used which allow the better predictability of efficacy and toxicity of the drugs at the earlier stages of drug development cycle. Another technique, cell painting also is gaining popularity in identifying targets, toxicity in compounds. Deep learning techniques can help in analyzing and deriving information from such complex datasets.
The aim of this Research Topic is to cover the promising, novel, and recent deep learning techniques in analyzing 3D cell cultures and cell paint datasets. Areas to be covered in this Research Topic may include, but are not limited to:
• Correctly identifying and label the different cellular components in 3D cell culture and cell paint data set, to quantify multiple phenotypic parameters to understand the effect of treated compound.
• Predict cell positions based on spatial patterns of cells in 3D dataset
• Analyze and segmentation of 3D complex cellular phenotypes
In recent years as science and technology has progressed, deep learning has emerged as an important tool in multiple fields to solve challenging problems. Deep learning allows to derive information from a vast amount of data in a relatively short period, which wasn't possible previously. Also, deep learning avoids the need for manual extraction of features, which reduces the burden from the user as well removes the possibility of error or biases during the feature extraction. These techniques have the potential for medical imaging, diagnostics, and medical data analysis.
In drug discovery process, cell-based assays are widely used. Historically, two dimensional (2D) monolayer cell cultures are being used to estimate drug responses. Though, it is been established that 2D cultures have limitations as they cannot mimic the tissue-specific architecture, mechanical and cell-to-cell interaction, which hinders the drug responses for some diseases like cancer. To overcome this issue, nowadays 3D cell cultures are used which allow the better predictability of efficacy and toxicity of the drugs at the earlier stages of drug development cycle. Another technique, cell painting also is gaining popularity in identifying targets, toxicity in compounds. Deep learning techniques can help in analyzing and deriving information from such complex datasets.
The aim of this Research Topic is to cover the promising, novel, and recent deep learning techniques in analyzing 3D cell cultures and cell paint datasets. Areas to be covered in this Research Topic may include, but are not limited to:
• Correctly identifying and label the different cellular components in 3D cell culture and cell paint data set, to quantify multiple phenotypic parameters to understand the effect of treated compound.
• Predict cell positions based on spatial patterns of cells in 3D dataset
• Analyze and segmentation of 3D complex cellular phenotypes
