Ex vivo full‐field cellular‐resolution optical coherence tomography of basal cell carcinomas: A pilot study of quality and feasibility of images and diagnostic accuracy in subtypes

• Published in: Skin research and technology Vol. 26; no. 2; pp. 308 - 316
• Main Authors: Wang, Yen‐Jen, Chang, Wei‐Chin, Wang, Jen‐Yu, Wu, Yu‐Hung
• Format: Journal Article
• Language: English
• Published: England John Wiley & Sons, Inc 01.03.2020
• Subjects: Accuracy; Basal cell carcinoma; Carcinoma; Carcinoma, Basal Cell - diagnostic imaging; Carcinoma, Basal Cell - pathology; cellular resolution; Dermatologists - education; Dermatologists - statistics & numerical data; Dermatology; Diagnostic systems; Equipment Design; Feasibility Studies; Feature recognition; full‐field; Humans; Image quality; Learning; Medical imaging; Optical Coherence Tomography; Pathology; Physicians; Pilot Projects; Sensitivity and Specificity; Skin - diagnostic imaging; Skin - pathology; Skin cancer; Skin Neoplasms - diagnostic imaging; Skin Neoplasms - pathology; Tomography; Tomography, Optical Coherence - methods; basal cell carcinoma; cellular resolution; optical coherence tomography; full-field
• ISSN: 0909-752X; 1600-0846; 1600-0846
• DOI: 10.1111/srt.12801

Background Studies have reported the application of conventional optical coherence tomography (OCT) in the diagnosis of basal cell carcinoma (BCC). The new OCT provides cellular details similar to those in pathology slides and may reduce user learning time. This study aimed to demonstrate the quality of ex vivo full‐field cellular‐resolution OCT images and compare the diagnostic accuracy between physicians with varying pathology experience. Materials and Methods Sixty histologically confirmed BCCs were selected. Tissue samples were sectioned and scanned using OCT, and their features were compared with those of hematoxylin and eosin (H&E)‐stained sections. Thirty images were selected for the test administered to dermatology residents, dermatopathology fellows, and board‐certified general pathologists without any OCT experience. The pretest learning included a 3‐min instruction and 10‐min self‐study of four BCC variants. Results Histopathological BCC and normal histological features were clearly recognizable on the OCT images. The pathological BCC features observed in the OCT images correlated with those found in the H&E‐stained sections. Seven participants completed the test. The correct answer rates of the residents, fellows, and pathologists were 71%, 68%, and 83% for BCC and 44%, 57%, and 57% for the BCC subtypes, respectively. Conclusion All the participants identified BCC in >70% cases with a learning time of only 13 minutes. The results indicated that cellular‐resolution OCT provided high‐quality images similar to the conventional pathology slides. Pathology experience did reflect the diagnostic accuracy. However, a longer training time is still needed at all levels to recognize the BCC subtypes correctly.