Network models for monitoring high-dimensional image profiles

Modern industries are investing in advanced imaging technology to increase information visibility, address system complexity, and improve quality and integrity of complex systems. The proliferation of high-dimensional images pose significant challenges on traditional and next-generation innovation practices for process monitoring and control in manufacturing and healthcare. Traditional statistical process control (SPC) is not concerned with imaging data but key product or process characteristics, and is limited in its ability to readily address complex structures of high-dimensional imaging profiles. Realizing the full potential of advanced imaging technology for process monitoring and control hinges on the development of new SPC methodologies. This paper presents a novel dynamic network methodology for monitoring and control of high-dimensional imaging streams. Experimental results on biomanufacturing and machining imaging profiles show that the proposed approach not only captures complex image structures but also provides an effective online control charts for monitoring image profiles. New dynamic network monitoring schemes are shown to have strong potentials to be generally applicable to research problems in diverse fields with image profiles.