Sina Farsiu, PhD

Sina Farsiu Farsio Super-Resolution superresolution optical coherence tomography oct sd-oct segmentation retina

Associate Professor
Department of Biomedical Engineering
Department of Ophthalmology
Department of Electrical & Computer Engineering
Department of Computer Science

Duke University

Mailing AddressBOX 3802, DUMC, Durham, NC, 27710
School of Engineering Office: Room 2575, CIEMAS Building
School of Medicine Office: Room 5014, AERI Building
Phone :919-684-6642

Email: sina DOT farsiu AT duke DOT edu


Automated Segmentation of Ocular Images

deep learning, cnn, farsiu, segmentation, amd, age related macular degeneration, graph search. graph cut, convolutional neural network

deep learning, cnn, farsiu, segmentation,  ,  mactel, convolutional neural network

CHIU SEGMENTATION SDOCT DME diabetic macular edema cyst cysts sina farsiu KRGTDP

Farsiu SEGMENTATION rod cone adaptive optics cunefare deep learning adaptive optics

CHIU SEGMENTATION SDOCT sina farsiucornea segmentation OCT layers sina farsiu

drusen geographic atrophy chiu farsiu sina segmentation automatic

mouse retina rhoko Rho(−/−) Pratul P. Srinivasan farsiu sina segmentation automatic

Macular Hole sina farsiu automatic  segmentation AMAL graph cut length adaptive graph search JBO brenton keller

Retinal pigment epithelium rpe cell sina farsiu automatic  segmentation confocal microscopy adaptive optics

  sina farsiu automatic  segmentation retina retinal vasculature vessel video indirect ophthalmoscopy Exploratory Dijkstra forest

  sina farsiu automatic  segmentation cell  cone rod SLO scanning laser ophthalmoscope  adaptive optics

  sina farsiu automatic  segmentation cell  cone rod split detector confocal SLO scanning laser ophthalmoscope  adaptive optics

  sina farsiu automatic  segmentation Hossein Rabbani fluorescein angiography (FA) Diabetic Retinopathy Diabetic macular edema DME

Artery-Vein Classification:

Sina Farsiu wide field  Artery-Vein Classification optos

Photonic Imaging:

Sina Farsiu Ballistic Photon

Ballistic Photon adaptive sampling compressive sensing

Speckle Variance Doppler Swept Source OCT Mosaic Capillary Registration

Image Compression    Compression sparse representation optical coherence tomography MPEG Sina Farsiu

Clinical Ophthalmic Research and Image Analysis

 Farsiu AMD age related macular degeneration ophthalmology retina sdoct control drusen geographi atrophy

 Farsiu deep learning handheld cell phone camera pediatric Open-source, machine and deep learning-based automated algorithm for gestational age estimation through smartphone lens imaging


Vision and Image Processing (VIP) Laboratory

VIP LAB DUKE UNIVERSITY SINA FARSIU  optical coherence tomography adaptive optics image processing segmentation denoising

VIP Lab members "occupying" a local sushi restaurant since 2011!


Image Reconstruction

Low-Quality Input Images on Left, High-Quality Reconstructed Images on Right


farsiu Interpolation  sparsity segmentation denoising graph optical coherence  compressive sensing AMD drusen geographic atrophy Sina Farsiu SD-OCT

Interpolation Denoising sparsity compressive sensing AMD drusen geographic atrophy Sina Farsiu SD-OCT

Denoising sparsity compressive sensing Sina Farsiu SDOCT

mosaicing vio video indirect ophthalmoscopy vessel segmentation registration Sina Farsiu

 video superresolution super resolution super-resolution kalman filter registration Sina Farsiu

  wavelet deblurring breast cancer digital tomosynthesis superresolution super resolution super-resolution xray  registration Sina Farsiu

   multiframe demosaicing demosaicking color superresolution super resolution super-resolution Sina Farsiu

  kernel regression interpolation denoising super-resolution  Sina Farsiu

   sdoct oct spectral domain optical coherence tomography vessel registration superresolution medical super-resolution  Sina Farsiu

Graph Theory    Tree Topology Estimation Carlo Tomasi Vessels  Sina Farsiu

Adaptive Optics Imaging

Handheld Adaptive Optics Scanning Laser Ophthalmoscoperetina cone haoslo cones sina farsiu

Compressed sensing wavefront sensor adaptive optics retina cone rod cones rods

  wavefront sensorless adaptive optics retina dme diabetic vasculature with wavefront sensorless adaptive optics optical coherence tomography angiography OCTA OCT-A

Multimodal Handheld Photonics Imaging    sdoct oct spectral domain optical coherence tomography  scanning laser ophthalmoscope SLO handheld  Sina Farsiu

OCT Registeration

wide-field Optical coherence tomography  registration Sina Farsiu boe jose lezama

Generalized Pseudo-Polar Fourier Grids

Generalized Pseudo-Polar Fourier registration Sina Farsiu

Foundamental Bounds in Detection and Segmentation

cramer rao bounds fluorescent genetically encoded calcium sensors Sina Farsiu

cramer rao bounds OCT Segmentation noise model Sina Farsiu



I am the director of the Vision and Image Processing (VIP) Laboratory and an associate professor of Biomedical Engineering and Ophthalmology with secondary appointments at the Departments of Electrical and Computer Engineering and Computer Science at Duke University. I am the Senior Area Editor of IEEE Transactions on Image Processing and the Associate Editor of Biomedical Optics Express and SIAM Journal on Imaging Sciences.

At VIP lab, our long-term goal is to improve the overall health and vision outcomes of at-risk patients with ocular and neurological diseases through earlier and better-directed therapy. To achieve this goal, we take advantage of recent advances in image processing and optics as an integrated technology to capture ocular images with higher resolution and better motion stability compared to the state-of-the-art imaging systems. Once these high-quality images are captured, we provide objective tools to quantitatively measure novel imaging biomarkers of the onset and progression of ophthalmic and neurological diseases.

VIP Lab’s Research Focus

1.      Image Analysis Software Development for Ophthalmology and Vision Sciences:

A major focus of our lab is development of fully automated software, using deep learning and other advanced machine learning technologies, to objectively detect and evaluate the biomarkers for onset and progression of ocular and neurological diseases in adults (e.g. diabetic retinopathy, age-related macular degeneration (AMD), Glaucoma, and Alzheimer's disease) and children (e.g. retinopathy or prematurity (ROP)). We also develop automatic segmentation algorithms to detect/segment/quantify ocular anatomical/pathological structures seen on ophthalmic imaging systems such as Optical Coherence Tomography (OCT) and adaptive optics scanning laser ophthalmoscopy (AO-SLO).

2.      Image Processing Theory and Application:

We study efficient signal processing based methods to overcome the theoretical and practical limitations that constrain the achievable resolution of any imaging device. Our approach, which is based on adaptive extraction and robust fusion of relevant information from the expensive and sophisticated as well as simple and cheap sensors, has found wide applications in improving the quality of imaging systems such as ophthalmic SD-OCT, video indirect ophthalmoscopy, digital X-ray mammography, electronic and optical microscopes, and commercial digital camcorders. When I am not busy developing a mathematical model of the procrastination theory, I play with some statistical signal processing ideas, mainly super-resolution, demosaicing/deblurring/denoising, motion estimation, compressive sensing/adaptive sampling, segmentation, and sensor fusion.

3.    Advanced Ophthalmic Imaging Hardware Development:

In collaboration with our colleagues at the department of biomedical engineering, especially the Laboratory for Biophotonicswe develop the next generation ophthalmic imaging systems, including advanced handheld OCT and adaptive optics ocular imaging systems.


November 2018: News release about Arjun Desai's paper on our open source automated software for detection of infants' gestational age through smartphone lens imaging.

November 2018: Call for papers in Neurophotonics special issue on "Advanced Retinal Imaging: Instruments, Methods, and Applications" which I'll be guest co-editing.

August 2018: News release about our paper on handheld adaptive optics scanning laser ophthalmoscope (HAOSLO). Also, we have made the Zemax optical design, Solidworks mechanical design, and LabView software, controlling the HAOSLO hardware open source and freely available online.

July 2018: Congrats to PhD Student Jessica Loo for receiving the John Chambers Scholar Award.

June 2018: Congrats to PhD Student Jessica Loo for receiving the First prize for outstanding research at the annual Duke Ophthalmology Trainee day Scientific Symposium.

March 2018:  Congrats to PhD student Jessica Loo for receiving the 2nd Place poster award for her work on deep learning based OCT image analysis at theFitzpatrick Institute for Photonics Annual Meeting.

Jan. 2018:  Congrats to PhD student David Cunefare for receiving the Retina Research Foundation/ Joseph M. and Eula C. Lawrence Travel Grant to present his exciting research on deep learning based adaptive optics image analysis at the ARVO 2018 Annual Meeting in Honolulu, Hawaii.

November 2017:  You want to know if there is room for improving the accuracy of layer segmentation algorithms? Theo Dubose’s exciting new paper shows the theoretical lower bounds on the accuracy of retinal layer segmentation methods. His paper also accurately derives retinal layer specific statistical models of signal and noise in OCT imaging. PREPRINT 

July 2017:  David Cunefare's Open source software for automatic detection of cone photoreceptors in adaptive optics ophthalmoscopy using convolutional neural networks is now freely availabel online (click here for the code and dataset )

May 2017:  Congrats and best wishes to Dr. Rolando Estrada for being appointed as a tenure-track Assistant Professor of Computer Science at the Georgia State University. Congrats to our ECE Pratt Fellow Eli Cole for graduating with distinction and winning the 2017 Charles Ernest Seager Memorial Award recognizing the most outstanding undergraduate research project in the ECE Department, and wishing him the best in the EE PhD program at Caltech. Congrats to our BME Pratt Fellow Leon Cai for graduating with distinction and winning the 2017 Clark Award in the BME Department, and wishing him the best success in the MD/PhD program at Vanderbilt University.

Jan. 2017:  Excited about our new paper on predicting the response to anti-VEGF treatment!.

Jan. 2017:  Congrats to our outstanding undergrad Pratt Fellow researcher, Eli Cole, for receiving the 1st place award in the Fall 2016 Duke ECE Independent Study Poster Session.

Jan. 2017:  Excited and honored to receive the " the Association for Research in Vision and Ophthalmology (ARVO) 2017 Pfizer Ophthalmics Carl Camras Translational Research Award".

Nov. 2016:  Jose Lezama has made publically available source code for his work on "Registration of wide field-of-view retinal optical coherence tomography volumes"

Sept. 2016:  Leyuan Fang has made publically available sofwtaer for his work on "Segmentation based sparse reconstruction of retinal optical coherence tomography volumes"

June 2016:  Congrats to PhD student Somayyeh Soltanian-Zadeh for being awarded Pre-doctoral NIH Fellowship in the Medical Imaging Training Program (MITP).

June 2015: Congrats to PhD Student David Cunefare Jr. for receiving the $94,000 John T. Chambers Scholar award.

June 2015:  Rolando Estrada has made publically available two datasets of annotated wide field-of-view color retinal images utilized in his recent IEEE PAMI paper on tree topology estimation and IEEE TMI paper on artery vein classification.

March 2015: Stephanie Chiu's solution to one of the most challenging automated OCT segmentation problems (layers and cysts in low-quality diabetic meacular edema images) is published and dataset of annotated images is now freely available online.

March 2015: Hossein Rabbani has made the full dataset of annotated fluorescein angiography images from diabetic meacular edema patients utilized in his paper featured on the cover of IOVS freely available online.

February 2015: We made large datasets of annotated images from multiple studies on diabetic meacular edema (SD-OCT ) freely available online.

December 2014: Congratulations to our outstanding High School research intern Elizabeth Chiu for being admitted to Duke University.

December 2014: Rolando Estrada's fundamental work on graph-theory is now accepted for publication in IEEE T-PAMI. This paper provides a practical solution for estimating the 3-D topology of tree-like structures (e.g. vessels and plant roots) from a single 2D image.

September 2014: The second first authored journal paper of our recently graduated outstanding undergrad Pratt Fellow, Pratul Srinivasan, is now published .

August 2014: New patent no. 8,811,745 issued: Segmentation and identification of layered structures in images.”

May 2014: Congratulations to Dr. Stephanie Chiu for officially receiving her well deserved PhD degree after co-authoring over 20 journal papers during her tenure at VIP lab.

May 2014: Congratulations to our outstanding Pratt Fellow Pratul Srinivasan for being admitted to all top ranking ECE PhD programs in USA.

May 2014: Congratulations to our outstanding High School research intern Alec V. Arshavsky for being admitted to Stanford University.

February 2014:  James Polans put the fast compressed sensing based wavefront measurement software freely available online.

January 2014:  Congrats to our high school research intern, Alec V. Arshavsky for winning the North Carolina International Science Challenge; he will represent USA in the next round of competition in Beijing, China. Also, congrats to Alec for being selected as a finalist in the Intel Science Talent Search (Intel STS).

December 2013:  Yeay! Our team of collaborators from different disciplines received Bass Connections funding for our proposed project "Art, Vision and the Brain: An Exploration of Color and Brightness". It will be fun!

August 2013:  Congrats to our new PhD student David Cunefare for being awarded Pre-doctoral NIH Fellowship in the Medical Imaging Training Program (MITP).

June 2013:  Yeay! Our team of collaborators from different disciplines received DIBS pilot funding to study ophthalmic imaging biomarkers in early Alzheimer’s disease!

May 2013:  Complete study dataset including automated and manual markings for our AO-SLO cone photoreceptor automatic segmentation paper is available online.

May 2013:  Congrats to Stephanie Chiu for advancing to PhD candidacy after co-authoring 18 (published/submitted) journal papers

July 2012:  Yeay! We got NIH R01 funding to develop software for automated classification of diabetic macular edema.

June 2012: PhD Student Stephanie Chiu received First prize for outstanding research at the annual Duke Ophthalmology Trainee day Scientific Symposium for the record "third times" in a row.

April. 2012: Yeay! We received NCBC funding to build a novel Ultrahigh-Resolution Adaptive Optics Optical Coherence Tomography/Scanning Laser Ophthalmoscopy System systems at DUEC.

April 2012: Stephanie Chiu has put the  complete study dataset including our automated and manual markings for her May 2012 BOE paper "Automatic segmentation of closed-contour features in ophthalmic images using graph theory and dynamic programming" online.

January 2012:  Yeay! We got NIH funding to build the next generation of handheld SD-OCT systems.

June 2011: PhD Student Stephanie Chiu received First prize for outstanding research at the annual Duke Ophthalmology Trainee day Scientific Symposium.

Feb. 2011: PhD Student Stephanie Chiu received the National Eye Institute travel award for her ARVO abstract.

Feb. 2011: 2010 IEEE Signal Processing Society -Best Paper Award  (Kernel Regression for Image Processing and Reconstruction, in IEEE Transactions on Image Processing). 

Nov. 2010: PhD Student Stephanie Chiu received $80,000 Chambers Fellowship

Oct. 2010: PhD Student Stephanie Chiu  received 3rd prize for best poster at the Annual Meeting of the Fitzpatrick Institute for Photonics

June 2010: PhD Student Stephanie Chiu received First prize for outstanding research at the annual Duke Ophthalmology Trainee day Scientific Symposium.

Apr. 2010: PhD Student Rolando Estrada  received AFER/Retina Research Foundation Student Travel Award for his ARVO abstract.


Emails without appropriate codes may directly go to my SPAM box!


2018 PhD Student Applicants:  I expect to recruit one BME PhD student with interests in image processing/photonics/ophthalmology related projects (I will also consider exceptional CompSci PhD student). l will only recruit students with exceptionally high achievements during undergraduate studies as evident by GPA from highly selective undergrad schools and/or high-impact peer-reviewed publications.

US residents with minimum 3.6GPA from a top US school may contact me directly via email (please type the code "USGS-2018" in the subject of your email).

International students: I receive hundreds of inquiries from students interested in joining my group, thus I maybe unable to answer each person individually. You are certainly welcome to contact me only if you have "already" submitted your application to Duke. In that case, please let me know that you are interested in working with me and I will look at your full application at Duke's website.

Exception is for the international students with financial aid from their own country. I will review their CV in PDF format before applying to Duke (please type the code "i-USGS-2018" in the subject of your email).

Current Duke Undergraduate/Graduate Students: Current Duke students (BME, EE, CE, or CS) interested in image processing/ophthalmology related projects can contact me directly via email (please type the code "i-MSGS-2018" if you are an MS and "i-DUS-2018" if you are an undergrad in the subject of your email).

Duke's 3rd Year Medical Students: Medical students interested in participating in ophthalmic imaging related projects may contact me directly. Please send me an email (please type the code "DMS-2018" in the subject of your email).  

Postdoctoral Research Associate #1: The minimum requirement is multiple first-authored high-quality image-processing journal publications (e.g. papers in IEEE TIP, CVPR/ICCV, IEEE PAMI, and Biomedical Optics Express) in the past three years. I am mainly interested in image analysis/segmentation expertise, esepcially deep learning. You must have a GoogleScholar page before applying for this position. Please type the code "PD-2017" in the subject of your email.

Postdoctoral Research Associate #2: Applicants with optical/photonics design expertise especially in the field of ophthalmic adaptive optics imaging are welcome to apply for a postdoc position jointly with Prof. Joseph A. Izatt. Please type the code "PD-2017" in the subject of your email.