Investigating recurrent neural networks for OCT A-scan based tissue analysis

• Published in: Methods of information in medicine Vol. 53; no. 4; p. 245
• Main Authors: Otte, C, Otte, S, Wittig, L, Hüttmann, G, Kugler, C, Drömann, D, Zell, A, Schlaefer, A
• Format: Journal Article
• Language: English
• Published: Germany 01.01.2014
• Subjects: Biopsy, Needle; Humans; Image Interpretation, Computer-Assisted; Image-Guided Biopsy; Lung - pathology; Multiple Pulmonary Nodules - classification; Multiple Pulmonary Nodules - pathology; Neural Networks (Computer); Sensitivity and Specificity; Software; Tomography, Optical Coherence; Optical Coherence Tomography; Long Short Term Memory; recurrent eural nets; optical soft tissue classification; identification of pulmonary nodules; biopsy guidance
• ISSN: 2511-705X
• DOI: 10.3414/ME13-01-0135

Optical Coherence Tomography (OCT) has been proposed as a high resolution image modality to guide transbronchial biopsies. In this study we address the question, whether individual A-scans obtained in needle direction can contribute to the identification of pulmonary nodules. OCT A-scans from freshly resected human lung tissue specimen were recorded through a customized needle with an embedded optical fiber. Bidirectional Long Short Term Memory networks (BLSTMs) were trained on randomly distributed training and test sets of the acquired A-scans. Patient specific training and different pre-processing steps were evaluated. Classification rates from 67.5% up to 76% were archived for different training scenarios. Sensitivity and specificity were highest for a patient specific training with 0.87 and 0.85. Low pass filtering decreased the accuracy from 73.2% on a reference distribution to 62.2% for higher cutoff frequencies and to 56% for lower cutoff frequencies. The results indicate that a grey value based classification is feasible and may provide additional information for diagnosis and navigation. Furthermore, the experiments show patient specific signal properties and indicate that the lower and upper parts of the frequency spectrum contribute to the classification.