David Romo-Bucheli, PhD.

 

Postdoctoral Research Scientist

Christian Doppler Laboratory for Ophthalmic Image Analysis

Department of Ophthalmology

Medical University of Vienna

Email:  david.romo-bucheli(at)meduniwien.ac.at

Phone: +43 1 40400 67600

Office: Rektoratsgebäude (Building 88, floor 2, room 206)


Bio:

David Romo-Bucheli finished his BSc in Electronic Engineering at Universidad Nacional de Colombia in Bogota, Colombia. Subsequently, he successfully pursued a Master's degree in Biomedical Engineering. Afterwards, he finished his doctoral studies in 2017 at Universidad Nacional de Colombia. During his PhD, and also as a visiting researcher in the Center of Computational Imaging and Personalized Diagnostics (CCIPD) - Case Western Reserve University, he was involved in the development of algorithms for the automatic identification of pathological markers in breast cancer histological images. Additionally, he studied the correlation of these automatically detected markers with breast cancer risk. In November 2017, he joined as a postdoc to the Ophthalmic Image Analysis laboratory (OPTIMA) in Vienna, Austria. His current research interests are the application of machine learning techniques, including deep learning, to the representation and analysis of medical images. Specifically, he aims to investigate the predictive power and the prognostic value of computational imaging features in the medical domain.


Research Interests:

- Application of machine learning techniques
- Deep learning
- Representation and analysis of medical images

 

Selected Publications:

 

Conference Papers

David Romo-Bucheli, Germán Corredor, Juan D. García-Arteaga, Viviana Arias, and Eduardo Romero. Nuclei graph local features for basal cell carcinoma classification in  whole  slide  images, Proceedings of the 12th International Symposium on Medical Information Processing and Analysis (SIPAIM), pp. 101600Q-101600Q, 2016. doi:10.1117/12.2257386

 David Romo-Bucheli, Andrew Janowczyk, Eduardo Romero, Hannah Gilmore, Anant Madabhushi. Automated tubule nuclei quantification and correlation with oncotype DX risk  categories  in  ER+  breast  cancer  whole  slide  images,  Proceedings  of  the  SPIE Medical Imaging, 2016. doi:10.1117/12.2211368

Ricardo Moncayo, David Romo-Bucheli, and Eduardo Romero. A Grading Strategy for  Nuclear  Pleomorphism  in  Histopathological  Breast  Cancer  Images  Using  a Bag  of Features  (BOF), Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications, LNCS, 2015. doi:10.1007/978-3-319-25751-8_10

David Romo-Bucheli,  Ricardo  Moncayo,  Angel  Cruz-Roa  and  Eduardo  Romero. Identifying histological concepts on basal cell carcinoma images using nuclei based sampling  and  multi-scale  descriptors,  Proceedings  of  the  2015  IEEE  12th International Symposium on Biomedical Imaging (ISBI), New York, 2015. doi:10.1109/ISBI.2015.7164041

David E. Romo, Juan D. García-Arteaga, Pablo Arbeláez and Eduardo Romero. A Discriminant  Multi-Scale  Histopathology  Descriptor  using  Dictionary  Learning, Proceedings of the SPIE Medical Imaging, 2014. doi:10.1117/12.2043935

David E. Romo, Jonathan Tarquino, Juan D. García-Arteaga and Eduardo Romero. Virtual  slide  mosaicing  using  feature  descriptors  and  a  registration  consistency  measure, Proceedings of the IX International Seminar on Medical Information Processing and Analysis, 2013. doi:10.1117/12.2035463

David Romo, Eduardo Romero, Fabio Gonzalez. Learning Regions of Interest from a Low-Level Maps in Virtual Microscopy”, Diagnostic Pathology, 2011. doi:10.1186/1746-1596-6-S1-S22

 

Journal Articles

David Romo-Bucheli, Andrew Janowczyk, Eduardo Romero, Hannah Gilmore, Anant Madabhushi. A deep learning based strategy for ideintifying and associating mitotic activity  with  gene  expression  derived  risk  categories  in  estrogen  receptor positive  breast cancers, Cytometry Part A, 2017. doi:10.1002/cyto.a23065

G. Lee, D. Romo-Bucheli, A. Madabhushi. Adaptive Dimensionality Reduction with Semi-Supervision (AdDReSS): Classifying Multi-Attribute Biomedical Data, PLoS One, 11(7), 2016. doi:10.1371/journal.pone.0159088

David Romo-Bucheli, Andrew Janowczyk, Eduardo Romero, Hannah Gilmore, Anant Madabhushi. Automated tubule nuclei quantification and correlation with oncotype DX risk  categories  in  ER+  breast  cancer  whole  slide  images, Nature  Publishing  Group, Scientific Reports, vol. 6, no. 32706, 2016. doi:10.1038/srep32706

Raul Celis, David Romo, Eduardo Romero. Blind colour separation of H&E stained histological images by linearly transforming the colour space, Journal of Microscopy, vol. 260, no 3, p. 377-388, 2015. doi:10.1111/jmi.12304