PDF icon Download PDF
Methods Inf Med 2014; 53(04): 245-249
DOI: 10.3414/ME13-01-0135
Original Articles
Schattauer GmbH

Investigating Recurrent Neural Networks for OCT A-scan Based Tissue Analysis

C. Otte
1   Institute of Medical Technology, Hamburg University of Technology, Hamburg, Germany
,
S. Otte
2   Cognitive Systems Group, Computer Science Department, University of Tuebingen, Tuebingen, Germany
,
L. Wittig
3   Medical Clinic III, University Hospital Schleswig Holstein, Luebeck, Germany
,
G. Hüttmann
4   Institute of Biomedical Optics, University of Luebeck, Luebeck, Germany
,
C. Kugler
5   Medical Clinic III, Department of Thoracic Surgery, LungenClinic Grosshansdorf, Grosshansdorf, Germany
,
D. Drömann
3   Medical Clinic III, University Hospital Schleswig Holstein, Luebeck, Germany
,
A. Zell
2   Cognitive Systems Group, Computer Science Department, University of Tuebingen, Tuebingen, Germany
,
A. Schlaefer
1   Institute of Medical Technology, Hamburg University of Technology, Hamburg, Germany
› Author Affiliations
Further Information

Publication History

received:03 December 2013

accepted:26 June 2014

Publication Date:
20 January 2018 (online)

   Permissions and Reprints
Preview

Summary

Objectives: 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.

Methods: 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.

Results: 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.

Conclusion: 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.

Keywords

Optical Coherence Tomography - biopsy guidance - optical soft tissue classification - identification of pulmonary nodules - Long Short Term Memory - recurrent eural nets