TY - JOUR
T1 - Rapid Automated Analysis of Skull Base Tumor Specimens Using Intraoperative Optical Imaging and Artificial Intelligence
AU - Jiang, Cheng
AU - Bhattacharya, Abhishek
AU - Linzey, Joseph R.
AU - Joshi, Rushikesh S.
AU - Cha, Sung Jik
AU - Srinivasan, Sudharsan
AU - Alber, Daniel
AU - Kondepudi, Akhil
AU - Urias, Esteban
AU - Pandian, Balaji
AU - Al-Holou, Wajd N.
AU - Sullivan, Stephen E.
AU - Thompson, B. Gregory
AU - Heth, Jason A.
AU - Freudiger, Christian W.
AU - Khalsa, Siri Sahib S.
AU - Pacione, Donato R.
AU - Golfinos, John G.
AU - Camelo-Piragua, Sandra
AU - Orringer, Daniel A.
AU - Lee, Honglak
AU - Hollon, Todd C.
N1 - Publisher Copyright:
© 2022 Congress of Neurological Surgeons. All rights reserved.
PY - 2022/6/1
Y1 - 2022/6/1
N2 - BACKGROUND: Accurate specimen analysis of skull base tumors is essential for providing personalized surgical treatment strategies. Intraoperative specimen interpretation can be challenging because of the wide range of skull base pathologies and lack of intraoperative pathology resources. OBJECTIVE: To develop an independent and parallel intraoperative workflow that can provide rapid and accurate skull base tumor specimen analysis using label-free optical imaging and artificial intelligence. METHODS: We used a fiber laser-based, label-free, nonconsumptive, high-resolution microscopy method (<60 seconds per 1 × 1 mm 2), called stimulated Raman histology (SRH), to image a consecutive, multicenter cohort of patients with skull base tumor. SRH images were then used to train a convolutional neural network model using 3 representation learning strategies: cross-entropy, self-supervised contrastive learning, and supervised contrastive learning. Our trained convolutional neural network models were tested on a held-out, multicenter SRH data set. RESULTS: SRH was able to image the diagnostic features of both benign and malignant skull base tumors. Of the 3 representation learning strategies, supervised contrastive learning most effectively learned the distinctive and diagnostic SRH image features for each of the skull base tumor types. In our multicenter testing set, cross-entropy achieved an overall diagnostic accuracy of 91.5%, self-supervised contrastive learning 83.9%, and supervised contrastive learning 96.6%. Our trained model was able to segment tumor-normal margins and detect regions of microscopic tumor infiltration in meningioma SRH images. CONCLUSION: SRH with trained artificial intelligence models can provide rapid and accurate intraoperative analysis of skull base tumor specimens to inform surgical decision-making.
AB - BACKGROUND: Accurate specimen analysis of skull base tumors is essential for providing personalized surgical treatment strategies. Intraoperative specimen interpretation can be challenging because of the wide range of skull base pathologies and lack of intraoperative pathology resources. OBJECTIVE: To develop an independent and parallel intraoperative workflow that can provide rapid and accurate skull base tumor specimen analysis using label-free optical imaging and artificial intelligence. METHODS: We used a fiber laser-based, label-free, nonconsumptive, high-resolution microscopy method (<60 seconds per 1 × 1 mm 2), called stimulated Raman histology (SRH), to image a consecutive, multicenter cohort of patients with skull base tumor. SRH images were then used to train a convolutional neural network model using 3 representation learning strategies: cross-entropy, self-supervised contrastive learning, and supervised contrastive learning. Our trained convolutional neural network models were tested on a held-out, multicenter SRH data set. RESULTS: SRH was able to image the diagnostic features of both benign and malignant skull base tumors. Of the 3 representation learning strategies, supervised contrastive learning most effectively learned the distinctive and diagnostic SRH image features for each of the skull base tumor types. In our multicenter testing set, cross-entropy achieved an overall diagnostic accuracy of 91.5%, self-supervised contrastive learning 83.9%, and supervised contrastive learning 96.6%. Our trained model was able to segment tumor-normal margins and detect regions of microscopic tumor infiltration in meningioma SRH images. CONCLUSION: SRH with trained artificial intelligence models can provide rapid and accurate intraoperative analysis of skull base tumor specimens to inform surgical decision-making.
KW - Artificial intelligence
KW - Automated diagnosis
KW - Contrastive learning
KW - Skull base tumors
KW - Stimulated Raman histology
KW - Tumor margin delineation
UR - http://www.scopus.com/inward/record.url?scp=85130631769&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85130631769&partnerID=8YFLogxK
U2 - 10.1227/neu.0000000000001929
DO - 10.1227/neu.0000000000001929
M3 - Article
C2 - 35343469
AN - SCOPUS:85130631769
SN - 0148-396X
VL - 90
SP - 758
EP - 767
JO - Neurosurgery
JF - Neurosurgery
IS - 6
ER -