- Electrical Engineering and Computer Science
Top 3 Areas of Expertise:
I am a PhD student supervised by Dr. Polina Golland in the Medical Vision Group in CSAIL. During my undergrad and Masters at the University of Toronto I worked on several algorithms for computational biology. Since coming to MIT, I've become interested in algorithms and models for medical and computer vision, as well as the intersection of medical vision and computational biology. I am currently developing models for extracting neuroimaging phenotypes from low-quality clinical scans of stroke patients, and leveraging these phenotypes in identifying novel genetic targets. Similarly, we are working on levarging genetic profiles to predict neuroimaging phenotypes.
On the side, I've worked on algorithms for computational photography and video, with special attention to time-lapses. I serve as the president of the MICCAI Society Student Board, and participate in the MIT-MGH SITECOR program, which allows engineers to observe surgical procedures in an effort to improve O.R. technologies.
Expected date of graduation:
In the context of brain diseases, imaging genetics aims to characterize the effect of perturbations in relevant genes on brain phenotypes, as observed via imaging. In early studies, imaging was used as an endo-phenotype to perform naïve, univariate modeling of genetic data. Such analysis identifies relevant genetic markers that may affect variability of particular traits, such as stroke risk and treatment outcome. Current approaches largely ignore complex genetic structure and interactions, and often summarize imaging data through few coarse measures such as volume of anatomical structures or of pathology. As a result, groups of genetic factors working together to produce a greater effect than any individual marker are likely to be missed by such methods.
In my thesis, I am developing statistical models of interactions between genetics and brain anatomy in the context of stroke, explicitly accounting for the structure patterns present in both data domains. Our close collaborators, led by Prof. Jonathan Rosand at MGH, have obtained a large imaging and genetics dataset of approximately 1000 stroke patients. We are developing generative models and algorithms to automatically extract neuroimaging phenotypes that are directly relevant to stroke, such as white matter hyperintensity volume in FLAIR images and stroke volume in DWI scans. We will also expand current epistasis modeling methods and investigate co-clustering of genetic markers with the extracted phenotypic traits. Finally, we will combine the neuroimaging components with genetic clusters by developing appropriate methods for correlation analysis to identify combined features. The goal of this framework is to determine connections between clusters of genetic factors, such as genes or SNPs, simultaneously with properties of brain anatomy, and to further determine how these connections are related to stroke risk and outcome. For example, certain genetic profiles and (pre-stroke) brain anatomy features are expected to differ between healthy subjects and patients experiencing stroke, yielding new target groups that together predict or strongly contribute to the disease.
Top 5 Awards and honors (name of award, date received):
5 Recent Papers:
A.V. Dalca, R. Sridharan, L. Cloonan, K. M. Fitzpatrick, A. Kanakis, K.L. Furie, J.Rosand, O.Wu, M.Sabuncu, N.S. Rost, P.Golland. Segmentation of Cerebrovascular Pathologies in Stroke Patients with Spatial and Shape Priors. In Proc. MICCAI: International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), LNCS 8674, pp. 773–780, 2014.
A.V. Dalca, R. Sridharan, N.S. Rost, P. Golland. tipiX: Rapid Visualization of Large Image Collections In MICCAI-IMIC Interactive Medical Image Computing Workshop, 2014.
Best paper award for Impact and usability.
R. Sridharan, A.V. Dalca, P. Golland. An interactive visualization tool for Nipype medical imaging pipelines. In MICCAI-IMIC Interactive Medical Image Computing Workshop, 2014.
R. Sridharan‡, A.V. Dalca‡*, K.M. Fitzpatrick, L. Cloonan, A. Kanakis, O. Wu, K.L. Furie, J. Rosand, N.S. Rost, P. Golland. Quantification and Analysis of Large Multimodal Clinical Image Studies: Application to Stroke. In Proc. MICCAI International Workshop on Multimodal Brain Image Analysis (MBIA), pp. 18–30, 2013.
‡ equal contribution
K.N. Batmanghelich, A.V. Dalca, M.R. Sabuncu, P. Golland. Joint Modeling of Imaging and Genetics, In Proc. IPMI: International Conference on Information Processing and Medical Imaging, LNCS 7917, pp. 766–777, 2013.