Single cell classification based on microscopic images is a well-known task tackled with various approaches, including Deep Learning algorithms. Typically, datasets consist of marked cells images altered by specific marker response. This simplifies classification but is resource-intensive. The pipeline consists of multiple deep neural networks and explainability techniques applied to cells extracted from the LIVECell dataset, generally used for segmentation, showing the accuracy and robustness of this approach. The generalization ability and classification performance of current deep models represent a key tool for label-free cell classification.
Label-Free Nervous System Single Cell Classification Using Pretrained VGG and ResNet Networks
Fiore, Pierpaolo;Bardozzo, Francesco;Tagliaferri, Roberto
2024-01-01
Abstract
Single cell classification based on microscopic images is a well-known task tackled with various approaches, including Deep Learning algorithms. Typically, datasets consist of marked cells images altered by specific marker response. This simplifies classification but is resource-intensive. The pipeline consists of multiple deep neural networks and explainability techniques applied to cells extracted from the LIVECell dataset, generally used for segmentation, showing the accuracy and robustness of this approach. The generalization ability and classification performance of current deep models represent a key tool for label-free cell classification.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.