Premium
Automated diagnosis of colon cancer using hyperspectral sensing
Author(s) -
Beaulieu Robert J.,
Goldstein Seth D.,
Singh Jasvinder,
Safar Bashar,
Banerjee Amit,
Ahuja Nita
Publication year - 2018
Publication title -
the international journal of medical robotics and computer assisted surgery
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.556
H-Index - 53
eISSN - 1478-596X
pISSN - 1478-5951
DOI - 10.1002/rcs.1897
Subject(s) - colorectal cancer , hyperspectral imaging , medicine , cancer , stage (stratigraphy) , resection , radiology , oncology , surgery , computer science , artificial intelligence , biology , paleontology
Background Surgical management of colorectal cancer relies on accurate identification of tumor and possible metastatic disease. Hyperspectral (HS) sensing is a passive, non‐ionizing diagnostic method that has been considered for multiple tumor types. The ability to use HS for identification of tumor specimens during surgical resection of colorectal cancers was explored. Methods Patients with colorectal cancer who underwent operative resection were enrolled. HS measurements were performed both intra‐ and extra‐luminally. Spectral results were correlated with pathologic evaluation. Results Fifteen patient specimens were analyzed. For patients with confirmed colorectal cancer, extraluminal spectra analysis yielded 61.68% sensitivity with 90% specificity. For intraluminal specimens, sensitivity increased to 91.97% with 90% specificity. Conclusions Hyperspectral sensing can reliably detect tumors in resected colon specimens. This research offers promising results for a diagnostic technology that is non‐ionizing and does not require the use of contrast agents to achieve accurate colorectal cancer detection.