Noise Reduction in Biomedical Images

Tomographic imaging modalities MRI,OCT & PET are
proving to be immensely helpful in the diagnosis of
various pathologies in clinical
environments. Images from these modalities span the
range from organ level to tissue level and represent
the condensation of information associated with the
objects that are imaged. The potential of this
information increases
signi cantly, if these images are effectively
processed and analyzed. However, the
quality of medical images is,usually, compromised due
to the undesirable noise
which reduces the SNR of these images. The degradation of
images by noise in uences the diagnosis of critical
diseases in clinical environments.
A novel noise reduction method based on wavelets and
differential evolution has
been proposed in this work. The proposed method was
tested on the images that
were acquired from MRI, OCT and PET. The application
of this method improved
SNR by a cosiderable amount while the loss in CNR was
kept at minimum level. A comprehensive analysis of
processed images, on the basis of different image
quality indices, showed that the proposed method,
effectively,
reduces noise without sacri cing the image resolution.

Autorentext

Vikas Gupta was born in Bareilly(India). Motivated by his desireto study human biology from a technical perspective, he graduatedwith a degree in biomedical engineering from MIT,Manipal. Afterworking for a year in SIEMENS, he joined a masters researchprogram in NTU(Singapore) where he worked on noise reductionmethods for biomedical images.

Klappentext

Tomographic imaging modalities MRI,OCT & PET areproving to be immensely helpful in the diagnosis ofvarious pathologies in clinicalenvironments. Images from these modalities span therange from organ level to tissue level and representthe condensation of information associated with theobjects that are imaged. The potential of thisinformation increases signi cantly, if these images are effectivelyprocessed and analyzed. However, thequality of medical images is,usually, compromised dueto the undesirable noisewhich reduces the SNR of these images. The degradation ofimages by noise in uences the diagnosis of criticaldiseases in clinical environments.A novel noise reduction method based on wavelets anddifferential evolution hasbeen proposed in this work. The proposed method wastested on the images thatwere acquired from MRI, OCT and PET. The applicationof this method improvedSNR by a cosiderable amount while the loss in CNR waskept at minimum level. A comprehensive analysis ofprocessed images, on the basis of different imagequality indices, showed that the proposed method,effectively,reduces noise without sacri cing the image resolution.