Deep Learning Based Semantic Segmentation and Quantification for MRD Biochip Images

•Deep learning-based object segmentation applied to bright field optical microscope images of patient samples.•No cell staining and no fluorescent labeling.•The bright-field optical microscopy images of the biochips include cells, micropads, and micro beads.•98.7% global accuracy, 86.1% mean IoU, 92.2% mean precision, 92.2% mean recall, and 92.2% mean f-1 score were reported.