Hybrid active contour model of hc segmentation into 3d optical local maps

Rajiv Gandhi Institute of Technology
Cholanagar, RGC Campus, RT Nagar,
Bangalore, Karnataka 560 032
08:22 Dept. Of BME, Rgit

Reasoning
Decision execution
thinking
Planning

Advance emerging platform

D Vinayakam
(1rg10bm008)
Under the Guidance Of,
Prof. Ashok K (B.E , Mtech)
Head of department
Department of BME

Content:
 Introduction
 Literature Survey
 Hybrid Active Contour Model
(HACM)
 Purpose Of Implementation
 Objective
 Functional Block
 Advantages
 Application
 Future Advancements
 Conclusion
 Query
 Bibliography
 References
08:22 Dept. Of BME, Rgit Title :HC Image Segmentation Analysis And Assistive Platform 6

INTRODUCTION

Hippo
campus
Hippocampus
Definition
Function
Location
Inhibition Memory Space
08:22
Dept. Of BME, Rgit Title :HC Image Segmentation Analysis And Assistive Platform
8

Disease Population
Non Disease Population
Worlds Population Suffering Mental Illness
Disease Population Hippocampus Related Non Disease Population
Hippocampus

Hippocampus ailmentsAging
Stress
Epilepsy
Schizophrenia
Transient
Global amnesia

Hippocampus Ailments
 Aging
Alzheimer
1. Episodic Memory
2. Working Memory
 Stress
1. Post Traumatic Stress
Disorder(PTSD)
2. Schizophrenia
3. Severe Depression
 Epilepsy
Hippocampal Sclerosis
 Schizophrenia
 Transient Global Amnesia

Literature Survey
 Hippocampus Segmentation Using ACM- IEEE 2012
 Hippocampus Segmentation Using Optical Local Map- IEEE 2013

Hybrid Active Contour Model
(HACM)

Auto Segmentation Method
Atlas Based
Techniques
Active Contour
Models (ACM)
Active Appearance
Models (AAM)

Optical Local Maps (OLM)
(Old)
ACM Frameworks
Region Based Image
Terms
Prior Terms
(New)
Hybrid ACM
Frameworks
Region
Based Terms
Edge Base Terms
Prior
Terms
08:22
16

Purpose Of Implementation
 Feature Enhancement
 Accurate and Automatic
 Initial Level and Final Level Diagnostic
 Treatment Assistive Aid

Objectives
 Detail Image Description With Data Graph
 Data Description
 Full Optical 3D Model
 Treatment Procedure

HC SEGMENTATION
FIGURE 1. A sagittal slice of a 3D MR volume (a), zoomed version of (a) where the
hippocampal region is indicated with magenta colour (b), and the reconstructed 3D
model of the hippocampus (c).
08:22
19

FUNCTIONAL BLOCK

MRI Image Scan
Diagnostic
Phase
Processed
Output
Suggestive
Treatment
Phase
Physician End

Diagnostic Phase
08:22
Dept. Of BME, Rgit Title :HC Image
Segmentation Analysis And Assistive Platform
22

3D-Optical Local
Map
Hybrid Active
Contour Model
Pictorial Data

FIGURE 2. Overview of the proposed methodology for calculating the spatial distribution map L, the OLMs as
well as the ACM parameters for a target image. The resulting L, OLMs and ACM parameters are
incorporated into the ACM framework, to produce the final segmentation.

Methods
Data
Description
Overview
Prior
ACM
Evolution
Calculation
Adaptation
 Dice Similarity Coefficient
D=2. Pr. Re D Є [0, 1]
Pr + Re
Evaluation Framework
08:22
25

 Data Description
1. OASIS (Open Access Series Of Image Studies) Dataset
2. IBSR (Internet Brain Segmentation Respiratory) Dataset
 Morphometric
3. OASIS MICCAI Dataset
 Neuromorphometrics
 Overview
1. Edge Based Term
2. Region Base Term
3. Prior Term
 Prior Information
 ACM Evolution
 Calculation Through Graph Cuts
 Adaptation Through Registration
08:22
26

FIGURE 3. Training phase: Overview of the procedure for calculating the training OLMs and ACM
parameters for the training images via an optimization scheme.08:22
27

Experiment Result
FIGURE 4. Segmentation results for subjects 2 (1st row), 12 (2nd row), 14 (3rd row) from
(a) OLM-ACM Joint, (b) Multi-atlas Joint, and (c) the AAM method of . On the both 2D slices and
(b) 3D models, blue color represents false positives, green true positives and
magenta false negatives.
08:22
28

Figure 5: Resulting Graph and Edges For Precise Data Information
08:22
29

Suggestive Treatment Phase
08:22
30

Processed Output Comparators
Alzheimer
Schizopherma
PSTD
Amnesia
Treatments Available
MEDICATION
SURGERY
Diseases
Stages
08:22
31

Advantages
 Suggestive Treatments
 User Friendly Platform
 Details Data Description
 Realistic Model
08:22
32

Application
 Pre and Post Diagnosis Of Therapy
 Real Time Interactive Environment
 MRI
08:22
33

Future Advancement
+
 CT
 Interfacing Robotics
08:22
34

08:22
35

Conclusion
 The incorporation of OLMs into a hybrid ACM, to be used on top of the multi-atlas
concept for HC segmentation.
 The evidence favours the inclusion of HC volumetric in clinical practice, to
enhance disease diagnosis, within a decision support system.
 The system provides platform for Treatment suggesting platform using graphics
08:22
36

08:22
37

Bibliography
DIMITRIOS ZARPALAS
Ph.D.
Laboratory of Medical
Informatics,
The Medical School,
Aristotle University of
Thessaloniki
POLYXENI GKONTRA
M.Sc.
Medical Informatics,
The Aristotle University Of
Thessaloniki,
Greece
PETROS DARAS
Researcher Grade B
The Information Technologies
Institute,
Centre for Research and
Technology, Hellas
NICOS MAGLAVERAS
The M.Sc. and Ph.D.
Electrical Engineering And Computer
Science From North Western
University, Evanston, IL, USA
08:22
38

References:
1. Multimodality operating platform-IEEE 2013
2. Pan American Health Organization. (2009). Strategy and plan of action on mental health.
Washington, DC, USA. Tech. Rep. CD 49/11, 2009 [Online]. Available:
http://www2.paho.org/hq/dmdocuments/2009/MENTAL_HEALTH.pdf
3. N. A. DeCarolis and A. J. Eisch, ‘‘Hippocampal neurogenesis as a target for the treatment of
mental illness: A critical evaluation,’’ Neuropharma- cology, vol. 58, no. 6, pp. 884–893, 2010.
4. H. U. Wittchen, F. Jacobi, J. Rehm, A. Gustavsson, M. Svensson, B. Jo¨nsson, et al., ‘‘The size
and burden of mental disorders and other disorders of the brain in Europe 2010,’’ Eur.
Neuropsychopharmacol., vol. 21, pp. 655–679, 2011. VOLUME 2, 2014 1800116Zarpalas et al.:
Accurate and Fully Automatic HC Segmentation
5. B. Bogerts, M. Ashtari, G. Degreef, J. M. Alvir, R. M. Bilder, and J.A.Lieberman, ‘‘Reduced
temporal limbic structure volumes on magnetic resonance images in ﬁrst episode schizophrenia,’’
Psychiatry Res, vol. 35, no. 1, pp. 1–13, 1990.
6. A. Breier, R. W. Buchanan, A. Elkashef, R. C. Munson, B. Kirkpatrick, and F. Gellad, ‘‘Brain
morphology and schizophrenia. A magnetic resonance imaging study of limbic, prefrontal cortex,
and caudate structures,’’Archives General Psychiatry, vol. 49, no. 12, pp. 921–926, 1992
08:22
39

6. D.Velakoulis, S.J.Wood, M.T.Wong, P.D.McGorry, A.Yung, L.Phillips, et al., ‘‘Hippocampal and
Amygdala volumes according to Psychosis stage and diagnosis: A magnetic resonance imaging
study of chronic schizophrenia, first-episode psychosis, and ultra-high-risk individuals,’’Archives
General Psychiatry, vol. 63, no. 2, pp. 139–49, 2006.
7. A. Sumich, X. A. Chitnis, D. G. Fannon, S. O’Ceallaigh, V. C. Doku, and A. Falrowicz,
‘‘Temporal lobe abnormalities in first-episode psychosis,’’Amer. J. Psychiatry, vol. 159, no. 7, pp.
1232–1235, 2002.
8. P. Brambilla, J. P. Hatch, and J. C. Soares, ‘‘Limbic changes identified by imaging in bipolar
patients,’’ Current Psychiatry Rep., vol. 10, no. 6, pp. 505–509, 2008.
9. C. Langan and C. McDonald, ‘‘Neurobiological trait abnormalities in bipolar disorder,’’ Mol.
Psychiatry, vol. 14, no. 9, pp. 833–846, 2009.
10. H. P. Blumberg, J. Kaufman, A. Martin, R. Whiteman, J. H. Zhang, J.C.Gore, etal., ‘‘Amygdala
and hippocampal volumes in adolescents and adults with bipolar disorder’’,
ArchivesGeneralPsychiatry,vol.60,no.12, pp. 1201–1208, 2003.
11. https://www.ushb.com
12. https://www.wikepedia.com/hippocampus
13. https://WWW.NHB.com
08:22
40

08:22
41

Hybrid active contour model of hc segmentation into 3d optical local maps

More Related Content

What's hot (20)

Similar to Hybrid active contour model of hc segmentation into 3d optical local maps (20)

Recently uploaded (20)

Hybrid active contour model of hc segmentation into 3d optical local maps