General Events

Back to Listing

CAII Fall Seminar Series: "Chemical Imaging for an Expanded View of the Pathologic Basis of Disease: A Challenge for AI"

Event Type
Center for Artificial Intelligence Innovation
Nov 9, 2020   11:00 am  
Rohit Bhargava, Professor, University of Illinois at Urbana-Champaign
Originating Calendar
Center for Artificial Intelligence Innovation

Rohit Bhargava, Professor at the Grainger College of Engineering at the University of Illinois at Urbana-Champaign, will present "Chemical Imaging for an Expanded View of the Pathologic Basis of Disease: A Challenge for AI" on Monday, November 9 at 11:00 a.m.

Abstract: Measurement science has traditionally driven improvements in instruments and the improved quality of recorded data has determined, in turn, the quality of information we can access. Since the signal could be directly estimated from recorded data and its uncertainty characterized, metrics such as the limits of detection, signal to noise ratio (SNR), and spatial resolution could be well defined and are widely adopted. As measurements become more complex, the push for higher precision continues and larger data sets become widely available, use of data science techniques becomes imperative to extract information and assure its quality. Here, we provide examples of the co-evolution of measurement and artificial intelligence (AI) pipelines using infrared (IR) spectroscopic imaging as an example. First, we discuss the fundamental advances in theory and instrumentation, as well as novel concepts that have led to a remarkable advance in measurement technology. We describe cases in which the performance of modern instruments can exceed those of the decades-dominant Fourier transform infrared (FT-IR) instruments, while optical designs are providing much finer, micron-scale data. The advances in measurements are highly synergistic with the advances in data storage and processing capability and a newfound public interest in the relation of data and knowledge using AI methods. New methods, processes, algorithms and systems to extract knowledge and insights from measured data are not only leading to improved understanding but also raise fundamental questions about the capabilities and limits of modern instrumentation. We provide examples of where AI tools have enabled new applications in cancer pathology and all-digital molecular analyses for routine histopathology.

Register to attend this webinar.

Seminar Zoom link.

link for robots only