Deep learning in digital holography for biomedical applications

Quantitative optical imaging techniques represent a new highly promising approach to identify such cellular biomarkers in particular when combining with artificial intelligence (AI) technologies for scientific, industrial, and most importantly biomedical applications. Among several new optical quantitative imaging techniques, digital holographic microscopy (DHM) have recently emerged as a powerful new technique well suited to non-invasively explore cell structure and dynamics with a nanometric axial sensitivity and hence to identify new cellular biomarkers. This overview paper provides explanations in the DHM to perform label-free phenotypic cellular assays. It further provides explanations of AI and deep learning pipelines for the development of an intelligent DHM that performs optical phase measurement, phase image processing, feature extraction, and classification. In addition, this paper provides some perspective on the use of the intelligent DHM in biomedical fields and shows its great potential for biomedical application.