---

Eppright Chair and University Distinguished Professor of Biomedical Engineering, Texas A&M University, USA

An exciting instrumental innovation… The combination of and creative uses of cell phone optics and data transmission with deep learning methods for data analysis – and especially image analysis for multidimensional data – is truly phenomenal. Prescreening of x-rays and MRI images is already being accomplished using AI tools. The same approaches are being extended to cell cytometry and environmental analyses. The cell phone imaging combined with AI-based analytics will open new doors for providing rapid, on-site analytical information for food safety, precision agriculture, and environmental safety as well as medical diagnostics.

Missing from the analytical toolbox? Inexpensive methods for nondestructive analysis of living systems in 4 dimensions at the cell and molecular level, including deep in tissues. Many investigators are currently tackling this problem, employing technologies such as IR, 2-photon, multispectral, impedance or quantum spectroscopy, usually with fairly expensive equipment.  Useful solutions will be inexpensive, flexible and may not even require direct contact with the tissue. Lots of challenges remain to solve this problem.

Spending a $1 Billion research grant… Set up a virtual institute with a 10-year lifetime to create, manufacture and deploy sensors that can measure complex toxic components in aquatic species, animals, plants and people over diverse parts of the earth. Analytics would be developed to provide data on the impact of manufacturing, energy creation, agricultural practices, climate change and human habitation. The institute would include researchers from relevant fields of engineering and science, lawyers, communication and policy specialists, and industrial partners.

Most memorable advice? “Serendipity happens in the cracks between the disciplines” Joel Schur, circa 1985. The greatest inventions emerge from the cracks between the disciplines. Reading widely and learning broadly is the best way to prepare to “commit invention.” Knowing what your predecessors have accomplished across your field and others can spark a better appreciation of capability gaps and generate new uses for analytical devices employing seemingly unrelated technologies.