Automated frame selection process for high-resolution microendoscopy

Abstract. We developed an automated frame selection algorithm for high-resolution microendoscopy video sequences. The algorithm rapidly selects a representative frame with minimal motion artifact from a short video sequence, enabling fully automated image analysis at the point-of-care. The algorithm was evaluated by quantitative comparison of diagnostically relevant image features and diagnostic classification results obtained using automated frame selection versus manual frame selection. A data set consisting of video sequences collected in vivo from 100 oral sites and 167 esophageal sites was used in the analysis. The area under the receiver operating characteristic curve was 0.78 (automated selection) versus 0.82 (manual selection) for oral sites, and 0.93 (automated selection) versus 0.92 (manual selection) for esophageal sites. The implementation of fully automated high-resolution microendoscopy at the point-of-care has the potential to reduce the number of biopsies needed for accurate diagnosis of precancer and cancer in low-resource settings where there may be limited infrastructure and personnel for standard histologic analysis.

[1]  Jeffrey J Rodríguez,et al.  Computer-aided identification of ovarian cancer in confocal microendoscope images. , 2008, Journal of biomedical optics.

[2]  Rolf-Rainer Grigat,et al.  Keyframe selection for robust pose estimation in laparoscopic videos , 2012, Medical Imaging.

[3]  Yixuan Yuan,et al.  Hierarchical key frames extraction for WCE video , 2013, 2013 IEEE International Conference on Mechatronics and Automation.

[4]  Michael L Wilson,et al.  Improvement of pathology in sub-Saharan Africa. , 2013, The Lancet. Oncology.

[5]  Eric J. Seibel,et al.  Surface Mosaics of the Bladder Reconstructed From Endoscopic Video for Automated Surveillance , 2012, IEEE Transactions on Biomedical Engineering.

[6]  Timothy J Muldoon,et al.  Subcellular-resolution molecular imaging within living tissue by fiber microendoscopy. , 2007, Optics express.

[7]  Timothy J Muldoon,et al.  High-resolution imaging in Barrett's esophagus: a novel, low-cost endoscopic microscope. , 2008, Gastrointestinal endoscopy.

[8]  Jenq-Neng Hwang,et al.  Object-based video abstraction for video surveillance systems , 2002, IEEE Trans. Circuits Syst. Video Technol..

[9]  Du-Yih Tsai,et al.  Information Entropy Measure for Evaluation of Image Quality , 2008, Journal of Digital Imaging.

[10]  Paul Over,et al.  Video shot boundary detection: Seven years of TRECVid activity , 2010, Comput. Vis. Image Underst..

[11]  Vijayashree S. Bhattar,et al.  Accuracy of In Vivo Multimodal Optical Imaging for Detection of Oral Neoplasia , 2012, Cancer Prevention Research.

[12]  Daniel C Barbosa,et al.  Automatic small bowel tumor diagnosis by using multi-scale wavelet-based analysis in wireless capsule endoscopy images , 2012, Biomedical engineering online.

[13]  P. Boyle,et al.  World Cancer Report 2008 , 2009 .

[14]  Luc Van Gool,et al.  Speeded-Up Robust Features (SURF) , 2008, Comput. Vis. Image Underst..

[15]  Max Q.-H. Meng,et al.  WCE video abstracting based on novel color and texture features , 2011, 2011 IEEE International Conference on Robotics and Biomimetics.

[16]  Sun Young Park,et al.  A Colon Video Analysis Framework for Polyp Detection , 2012, IEEE Transactions on Biomedical Engineering.

[17]  S. Dawsey,et al.  Quantitative analysis of high-resolution microendoscopic images for diagnosis of esophageal squamous cell carcinoma. , 2015, Clinical gastroenterology and hepatology : the official clinical practice journal of the American Gastroenterological Association.

[18]  Jeffrey D Sachs,et al.  Expansion of cancer care and control in countries of low and middle income: a call to action , 2010, The Lancet.

[19]  David Beymer,et al.  Automatic Selection of Keyframes from Angiogram Videos , 2010, 2010 20th International Conference on Pattern Recognition.

[20]  Sung Wook Baik,et al.  Video summarization based tele-endoscopy: a service to efficiently manage visual data generated during wireless capsule endoscopy procedure , 2014, Journal of Medical Systems.

[21]  Chung-Feng Jeffrey Kuo,et al.  Using image processing technology and mathematical algorithm in the automatic selection of vocal cord opening and closing images from the larynx endoscopy video , 2013, Comput. Methods Programs Biomed..

[22]  William E. Higgins,et al.  Construction of a multimodal CT-video chest model , 2014, Medical Imaging.

[23]  K Cook,et al.  Comparison of autofocus methods for automated microscopy. , 1991, Cytometry.

[24]  Tie-Yan Liu,et al.  Dynamic selection and effective compression of key frames for video abstraction , 2003, Pattern Recognit. Lett..

[25]  Rebecca Richards-Kortum,et al.  High-resolution Fiber-optic Microendoscopy for in situ Cellular Imaging , 2011, Journal of visualized experiments : JoVE.

[26]  Timothy J Muldoon,et al.  Noninvasive imaging of oral neoplasia with a high-resolution fiber-optic microendoscope , 2012, Head & neck.

[27]  Dimitrios K. Iakovidis,et al.  Reduction of capsule endoscopy reading times by unsupervised image mining , 2010, Comput. Medical Imaging Graph..

[28]  Alfredo Ruggeri,et al.  Super-image mosaic of infant retinal fundus: Selection and registration of the best-quality frames from videos , 2013, 2013 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).