Coronavirus Disease 2019 (COVID-19): A Pediatric Perspective

"KAIZEN" method realizing implementation of deep-learning models for COVID-19 CT diagnosis in real world hospitals

Numerous COVID-19 diagnostic imaging Artificial Intelligence (AI) studies exist. However, none of their models were of potential clinical use, primarily owing to methodological defects and the lack of implementation considerations for inference. In this study, all development processes of the deep-learning models are performed based on strict criteria of the "KAIZEN checklist", which is proposed based on previous AI development guidelines to overcome the deficiencies mentioned above. We develop and evaluate two binary-classification deep-learning models to triage COVID-19: a slice model examining a Computed Tomography (CT) slice to find COVID-19 lesions; a series model examining a series of CT images to find an infected patient. We collected 2,400,200 CT slices from twelve emergency centers in Japan. Area Under Curve (AUC) and accuracy were calculated for classification performance. The inference time of the system that includes these two models were measured. For validation data, the slice and series models recognized COVID-19 with AUCs and accuracies of 0.989 and 0.982, 95.9% and 93.0% respectively. For test data, the models' AUCs and accuracies were 0.958 and 0.953, 90.0% and 91.4% respectively. The average inference time per case was 2.83 s. Our deep-learning system realizes accuracy and inference speed high enough for practical use. The systems have already been implemented in four hospitals and eight are under progression. We released an application software and implementation code for free in a highly usable state to allow its use in Japan and globally.

Conditional cell reprogramming for modeling host-virus interactions and human viral diseases

Conventional cancer and transformed cell lines are widely used in cancer biology and other fields within biology. These cells usually have abnormalities from the original tumor itself, but may also develop abnormalities due to genetic manipulation, or genetic and epigenetic changes during long-term passages. Primary cultures may maintain lineage functions as the original tissue types, yet they have a very limited life span or population doubling time because of the nature of cellular senescence. Primary cultures usually have very low yields, and the high variability from any original tissue specimens, largely limiting their applications in research. Animal models are often used for studies of virus infections, disease modeling, development of antiviral drugs, and vaccines. Human viruses often need a series of passages in vivo to adapt to the host environment because of variable receptors on the cell surface and may have intracellular restrictions from the cell types or host species. Here, we describe a long-term cell culture system, conditionally reprogrammed cells (CRCs), and its applications in modeling human viral diseases and drug discovery. Using feeder layer coculture in presence of Y-27632 (conditional reprogramming, CR), CRCs can be obtained and rapidly propagated from surgical specimens, core or needle biopsies, and other minimally invasive or noninvasive specimens, for example, nasal cavity brushing. CRCs preserve their lineage functions and provide biologically relevant and physiological conditions, which are suitable for studies of viral entry and replication, innate immune responses of host cells, and discovery of antiviral drugs. In this review, we summarize the applications of CR technology in modeling host-virus interactions and human viral diseases including severe acute respiratory syndrome coronavirus-2 and coronavirus disease-2019, and antiviral discovery.