학술논문
Rapid and Label-Free Histopathology of Oral Lesions Using Deep Learning Applied to Optical and Infrared Spectroscopic Imaging Data.
Document Type
Academic Journal
Author
Confer MP; Beckman Institute for Advanced Science and Technology, University of Illinois Urbana Champaign, Urbana, IL 61820, USA.; Falahkheirkhah K; Beckman Institute for Advanced Science and Technology, University of Illinois Urbana Champaign, Urbana, IL 61820, USA.; Department of Chemical and Biomolecular Engineering, University of Illinois Urbana Champaign, Urbana, IL 61820, USA.; Surendran S; Head & Neck Surgery, Roswell Park Comprehensive Cancer Center, Buffalo, NY 14263, USA.; Sunny SP; Head and Neck Surgery, Mazumdar Shaw Medical Foundation, Narayana Health City, Bangalore 560099, India.; Yeh K; Beckman Institute for Advanced Science and Technology, University of Illinois Urbana Champaign, Urbana, IL 61820, USA.; Department of Bioengineering, University of Illinois Urbana Champaign, Urbana, IL 61820, USA.; Liu YT; Beckman Institute for Advanced Science and Technology, University of Illinois Urbana Champaign, Urbana, IL 61820, USA.; Department of Electrical and Computer Engineering, University of Illinois Urbana Champaign, Urbana, IL 61820, USA.; Sharma I; Beckman Institute for Advanced Science and Technology, University of Illinois Urbana Champaign, Urbana, IL 61820, USA.; Orr AC; Beckman Institute for Advanced Science and Technology, University of Illinois Urbana Champaign, Urbana, IL 61820, USA.; Lebovic I; Department of Bioengineering, University of Illinois Urbana Champaign, Urbana, IL 61820, USA.; Magner WJ; Head & Neck Surgery, Roswell Park Comprehensive Cancer Center, Buffalo, NY 14263, USA.; Sigurdson SL; Head & Neck Surgery, Roswell Park Comprehensive Cancer Center, Buffalo, NY 14263, USA.; Aguirre A; Oral Diagnostic Sciences, University at Buffalo School of Dental Medicine, Buffalo, NY 14215, USA.; Markiewicz MR; Head & Neck Surgery, Roswell Park Comprehensive Cancer Center, Buffalo, NY 14263, USA.; Oral and Maxillofacial Surgery, University at Buffalo School of Dental Medicine, Buffalo, NY 14215, USA.; Suresh A; Head & Neck Surgery, Roswell Park Comprehensive Cancer Center, Buffalo, NY 14263, USA.; Head and Neck Surgery, Mazumdar Shaw Medical Foundation, Narayana Health City, Bangalore 560099, India.; Hicks WL Jr; Head & Neck Surgery, Roswell Park Comprehensive Cancer Center, Buffalo, NY 14263, USA.; Birur P; KLE Society Institute of Dental Sciences, Bangalore 560099, India.; Kuriakose MA; Head & Neck Surgery, Roswell Park Comprehensive Cancer Center, Buffalo, NY 14263, USA.; Head and Neck Surgery, Mazumdar Shaw Medical Foundation, Narayana Health City, Bangalore 560099, India.; Karkinos Healthcare, Kochi 682017, India.; Bhargava R; Beckman Institute for Advanced Science and Technology, University of Illinois Urbana Champaign, Urbana, IL 61820, USA.; Department of Chemical and Biomolecular Engineering, University of Illinois Urbana Champaign, Urbana, IL 61820, USA.; Department of Bioengineering, University of Illinois Urbana Champaign, Urbana, IL 61820, USA.; Department of Electrical and Computer Engineering, University of Illinois Urbana Champaign, Urbana, IL 61820, USA.; Department of Chemistry, University of Illinois Urbana Champaign, Urbana, IL 61820, USA.; Department of Mechanical Science and Engineering, University of Illinois Urbana Champaign, Urbana, IL 61820, USA.; Cancer Center at Illinois, University of Illinois Urbana Champaign, Urbana, IL 61820, USA.
Source
Publisher: MDPI AG Country of Publication: Switzerland NLM ID: 101602269 Publication Model: Electronic Cited Medium: Print ISSN: 2075-4426 (Print) Linking ISSN: 20754426 NLM ISO Abbreviation: J Pers Med Subsets: PubMed not MEDLINE
Subject
Language
English
ISSN
2075-4426
Abstract
Oral potentially malignant disorders (OPMDs) are precursors to over 80% of oral cancers. Hematoxylin and eosin (H&E) staining, followed by pathologist interpretation of tissue and cellular morphology, is the current gold standard for diagnosis. However, this method is qualitative, can result in errors during the multi-step diagnostic process, and results may have significant inter-observer variability. Chemical imaging (CI) offers a promising alternative, wherein label-free imaging is used to record both the morphology and the composition of tissue and artificial intelligence (AI) is used to objectively assign histologic information. Here, we employ quantum cascade laser (QCL)-based discrete frequency infrared (DFIR) chemical imaging to record data from oral tissues. In this proof-of-concept study, we focused on achieving tissue segmentation into three classes (connective tissue, dysplastic epithelium, and normal epithelium) using a convolutional neural network (CNN) applied to three bands of label-free DFIR data with paired darkfield visible imaging. Using pathologist-annotated H&E images as the ground truth, we demonstrate results that are 94.5% accurate with the ground truth using combined information from IR and darkfield microscopy in a deep learning framework. This chemical-imaging-based workflow for OPMD classification has the potential to enhance the efficiency and accuracy of clinical oral precancer diagnosis.