Near real-time intraoperative brain tumor diagnosis using stimulated Raman histology and deep neural networks

Autor: Karin M. Muraszko, Randy S. D'Amico, Siri Sahib S Khalsa, Jay Kenneth Trautman, Jeffrey N. Bruce, Marc L. Otten, Ashish H. Shah, Akshay V. Save, Matija Snuderl, Tamara Marie, Daniel A. Orringer, B. Gregory Thompson, Parag G. Patil, Stephen E. Sullivan, Balaji Pandian, Todd C. Hollon, Michael E. Ivan, Petros Petridis, Ricardo J. Komotar, Timothy D. Johnson, Shawn L. Hervey-Jumper, Esteban Urias, Arjun R. Adapa, Sandra Camelo-Piragua, Erin L. McKean, Peter Canoll, Cormac O. Maher, Guy M. McKhann, Michael B. Sisti, Hugh J. L. Garton, Oren Sagher, Spencer Lewis, Honglak Lee, Christian W. Freudiger, Daniel G Eichberg, Jason Heth, Zia Farooq
Rok vydání: 2020
Předmět:
0301 basic medicine
Image Processing
H&E stain
Spectrum Analysis
Raman
Medical and Health Sciences
Convolutional neural network
Computer-Assisted
0302 clinical medicine
Image Processing
Computer-Assisted
Segmentation
Raman
Cancer
Intraoperative
Clinical Trials as Topic
screening and diagnosis
Artificial neural network
Brain Neoplasms
General Medicine
Detection
Feature (computer vision)
030220 oncology & carcinogenesis
Radiology
Algorithms
medicine.medical_specialty
Monitoring
Neural Networks
Immunology
education
Brain tumor
Image processing
Article
General Biochemistry
Genetics and Molecular Biology
Computer
03 medical and health sciences
Deep Learning
Computer Systems
Clinical Research
Monitoring
Intraoperative
medicine
Humans
Probability
business.industry
Spectrum Analysis
Deep learning
Neurosciences
medicine.disease
4.1 Discovery and preclinical testing of markers and technologies
030104 developmental biology
Neural Networks
Computer
Artificial intelligence
business
Zdroj: Nature medicine, vol 26, iss 1
Nature medicine
ISSN: 1546-170X
1078-8956
DOI: 10.1038/s41591-019-0715-9
Popis: Intraoperative diagnosis is essential for providing safe and effective care during cancer surgery1. The existing workflow for intraoperative diagnosis based on hematoxylin and eosin staining of processed tissue is time, resource and labor intensive2,3. Moreover, interpretation of intraoperative histologic images is dependent on a contracting, unevenly distributed, pathology workforce4. In the present study, we report a parallel workflow that combines stimulated Raman histology (SRH)5-7, a label-free optical imaging method and deep convolutional neural networks (CNNs) to predict diagnosis at the bedside in near real-time in an automated fashion. Specifically, our CNNs, trained on over 2.5 million SRH images, predict brain tumor diagnosis in the operating room in under 150 s, an order of magnitude faster than conventional techniques (for example, 20-30 min)2. In a multicenter, prospective clinical trial (n = 278), we demonstrated that CNN-based diagnosis of SRH images was noninferior to pathologist-based interpretation of conventional histologic images (overall accuracy, 94.6% versus 93.9%). Our CNNs learned a hierarchy of recognizable histologic feature representations to classify the major histopathologic classes of brain tumors. In addition, we implemented a semantic segmentation method to identify tumor-infiltrated diagnostic regions within SRH images. These results demonstrate how intraoperative cancer diagnosis can be streamlined, creating a complementary pathway for tissue diagnosis that is independent of a traditional pathology laboratory.
Databáze: OpenAIRE
Externí odkaz: