[ Back to main page ]
 

Abstract

 
Abstract No.:C-F3179
Country:India
  
Title:A NEUROIMAGING TECHNIQUE FOR EARLY DIAGNOSIS OF CEREBRAL MALARIA AND DIFFERENTIAL EDEMA ZONES USING CT FOR EMERGENCY THERAPY IN DEVELOPING COUNTRIES
  
Authors/Affiliations:1 V. P. Subramanyam Rallabandi*; 1 Prasun Roy;
1 National Brain Research Centre, Manesar, India
  
Content:Objective: We devise a computed tomography-based image processing procedure for accurate early diagnosis and grading of cerebral malaria, and differentiating between umbra and penumbra region of brain oedema. The basic aim is to enable the clinician to diagnose the lesion load in cerebral malaria early, and thereby initiate aggressive treatment, as the disease has a very brief window of therapeutic opportunity, after which the disorder becomes intractably fatal. Cerebral malaria is the most lethal feature of malaria, a major health hazard in over half of the world, across tropical and subtropical Asia, Africa and Americas.

Materials and methods: Motivated by information theory and nonlinear dynamics principles, we develop a technique using pertubative image activation for enhancement of noisy or indistinct computed assisted tomography images of the brain lesions in cerebral malaria. We illustrate the image enhancement benefit both visually and quantitatively for the radiologist. The CT image of malaria patients were taken, who were clinically diagnosed to have cerebral malaria, and we considered the CT axial slices, through supratentorial and infratentorial regions at slice thickness of 1 cm and 0.8 cm, respectively. The image processing technique was applied to the slices, the mathematical parameters of the procedure was optimized, so as to maximally enhance the oedematous region and also give differential contrast between the two zones of oedema, namely the cytotoxiz edema and vasogenic edema zones.

Results: Even if the image noise or blurring is considerable, the proposed method enhances the contour of the lesion, delineates the edematous zones more clearly, and clarifies the latent structure of the lesions, along with aiding more efficient discrimination of the different regions and structurations of the lesion. The degree of enhancement was characterized by increase of the image contrast index of the lesion. We also estimated the image contrast gain (produced by the perturbative image processing technique), by a quotient scale between the pre-enhancement and post-enhancement image performance indices. The image processing technique was found to generate sufficient contrast between the two different zones of the cerebral oedema region. We found another advantage of the perturbative imaging approach, in that it can adapt to the local image texture and perform maximal contrast and enhancement if one properly tunes the dynamical image processing parameters.

Conclusion: The advantages of the proposed method are that it can simultaneously operate both as an enhancement process as well as a noise-reduction operation for the brain images, and that the method can also optimally enhance an image even if the noise level is considerable, which happens usually with CT scanners maintained in developing countries. Formation of cerebral edema is the most significant sign indicating that the impending pathogenesis of cerebral malaria is on its way, and that imperative treatment is needed for the patient. The size of the cytotoxic zone of the edema signifies the degree of aggressivity and urgency with which to institute the treatment. The proposed image processing technique helps in estimating both aspects early, thus enabling accurate timely therapeutic intervention for
  
Back