A Highlighted Paper in IEEE Transactions on Biomedical Engineering

My recent work, in collaboration with scientists and doctors from Johns Hopkins University, School of Medicine, Center for Tuberculosis Research Center, was accepted to be published in IEEE Transactions on Biomedical Engineering.

To our best of knowledge, this is the first study in quantifying small animal infectious disease model from PET images. Our proposed method is fully automated, faster, highly accurate, and it has certain advantages to state of the art PET image segmentation methods in the literature. Our novel PET segmentation algorithm relies on the Affinity Propagation (AP) Clustering framework.

Title: Segmentation of PET Images for Computer-Aided Functional Quantification of Tuberculosis in Small Animal Models

Authors:  B. Foster*, U. Bagci, Z.Xu, B. Dey, B. Luna, W. Bishai, S. Jain, D.J. Mollura

Journal: IEEE Transactions on Biomedical Engineering (Impact Factor: 2.347)

Vol 61, number 3, pp. 711-724, 2014.

(*: indicates my Post-BAC student who was under full supervision of me during the study).

Abstract of the paper is below:

Pulmonary infections often cause spatially diffuse  and multi-focal radiotracer uptake in positron emission tomography (PET) images, which makes accurate quantification of the disease extent challenging. Image segmentation plays a vital role in quantifying uptake due to distributed nature of immuno-pathology and associated metabolic activities in pulmonary infection, specifically tuberculosis (TB). For this task,thresholding-based segmentation methods may be better suited over other methods; however, performance of the thresholding-based methods depend on the selection of thresholding parameters, which are often sub-optimal. Several optimal thresholding techniques have been proposed in the literature, but there is currently no consensus on how to determine the optimal threshold for precise identification of spatially diffuse  and multi-focal radiotracer uptake. In this study, we propose a method to select optimal thresholding levels by utilizing a novel intensity affinity metric within the Affinity Propagation clustering framework. We tested the proposed method against 70 longitudinal PET images of rabbits infected with TB. The overall dice similarity coefficient (DSC) between the segmentation from the proposed method and two expert segmentations was found to be 91.25 +/- 8.01% with a sensitivity of 88.80 +/- 12.59% and a specificity of 96.01 +/- 9.20%. High accuracy and heightened efficiency of our proposed method, as compared to other PET image segmentation methods, were reported with various quantification metrics.

Earlier version of the manuscript was presented at ISBI 2013 conference, and the PDF of the ISBI 2013 article can be found HERE!

PDF of the IEEE TBME paper.

Click for MATLAB GUI for AP based PET Segmentation


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s