Dynamic Contrast Enhanced (DCE) MRI : Hierarchical Clustering-based Segmentation (HCS) as an aid to diagnosis

SELVAN, Arul and WRIGHT, Chris (2014). Dynamic Contrast Enhanced (DCE) MRI : Hierarchical Clustering-based Segmentation (HCS) as an aid to diagnosis. In: The 20th Annual Scientific Meeting of the British Chapter of the International Society for Magnetic Resonance in Medicine (ISMRM), Edinburgh, UK, 4-5 September 2014.

[img]
Preview
PDF (Poster)
ISMRM_Conference_Poster.pdf - Accepted Version
All rights reserved.

Download (2MB) | Preview
Official URL: https://ismrmannualmeeting.wordpress.com/

Abstract

Dynamic Contrast Enhanced Magnetic Resonance Imaging (DCE-MRI) has become an important component in the diagnostic imaging pathway and is emerging as a useful clinical technique for evaluating the severity, location, and extent of primary and recurrent cancer. But DCE-MRI typically generates around 30 images per section and image interpretation requires substantial experience to detect and categorize lesions.

Tissue abnormality is usually related to a dissimilar part of an otherwise homogeneous image. Choosing the optimal post processing threshold value can be difficult because the image composition may vary depending on the acquisition parameters and the type of tissue. Hierarchical Clustering-based Segmentation (HCS) is an approach to Computer Aided Monitoring (CAM) that enables a user to define a region of interest and the process generates a hierarchy of segmentation results to highlight the varied dissimilarities that might be present.

This new HCS process based CAM system offers a versatile and flexible environment by allowing the user to derive the maximum benefit from the computational capability (perception) of the machine. At the same time, the user is able to incorporate their own interpretation in the appropriate place and thus limit the machine's interpretive function to achieve a complementary synthesis of both computer and human strengths.

As a diagnostic aid for the analysis of DCE-MRI image data, the process starts with the user defining a rough region of interest (ROI) on a section/slice of choice. Within the user defined ROI, the HCS process is applied to all the DCE-MRI temporal frames of the slice. HCS process output provides heat map images based on the normalised average pixel value of the various dissimilar regions within the ROI. Time intensity curves of the contrast wash-in, wash-out process are then plotted for suspicious regions confirmed by the user.

Early results suggest that the HCS process based CAM system offers increased capability to differentiate suspicious areas by combining users' expertise and computer system's processing capability. The application is useful at the point of first diagnosis because often lesions are not solitary in nature, which can result in an incomplete treatment regime and affect prognosis; and also in monitoring the effects of drugs or radiotherapy.

Ongoing research applications of the HCS process based CAM system are prostate, breast, knee, brain, and liver imaging. An example of the HCS process based CAM system for prostate is outlined below.

Item Type: Conference or Workshop Item (Poster)
Research Institute, Centre or Group - Does NOT include content added after October 2018: Materials and Engineering Research Institute > Modelling Research Centre > Microsystems and Machine Vision Laboratory
Depositing User: Arul Selvan
Date Deposited: 28 Apr 2015 09:17
Last Modified: 18 Mar 2021 13:37
URI: https://shura.shu.ac.uk/id/eprint/9586

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year

View more statistics