Combining Sparse and Dense Features to Improve Multi-Modal Registration for Brain DTI Images

Abstract

A new solution to overcome the constraints of multimodality medical intra-subject image registration is proposed, using the mutual information (MI) of image histogram-oriented gradients as a new matching criterion. We present a rigid, multi-modal image registration algorithm based on linear transformation and oriented gradients for the alignment of T2-weighted (T2w) images (as a fixed reference) and diffusion tensor imaging (DTI) (b-values of 500 and 1250 s/mm2) as floating images of three patients to compensate for the motion during the acquisition process. Diffusion MRI is very sensitive to motion, especially when the intensity and duration of the gradient pulses (characterized by the b-value) increases. The proposed method relies on the whole brain surface and addresses the variability of anatomical features into an image stack. The sparse features refer to corners detected using the Harris corner detector operator, while dense features use all image pixels through the image histogram of oriented gradients (HOG) as a measure of the degree of statistical dependence between a pair of registered images. HOG as a dense feature is focused on the structure and extracts the oriented gradient image in the x and y directions. MI is used as an objective function for the optimization process. The entropy functions and joint entropy function are determined using the HOGs data. To determine the best image transformation, the fiducial registration error (FRE) measure is used. We compare the results against the MI-based intensities results computed using a statistical intensity relationship between corresponding pixels in source and target images. Our approach, which is devoted to the whole brain, shows improved registration accuracy, robustness, and computational cost compared with the registration algorithms, which use anatomical features or regions of interest areas with specific neuroanatomy. Despite the supplementary HOG computation task, the computation time is comparable for MI-based intensities and MI-based HOG methods.

An NLP-based technique to extract meaningful features from drug SMILES

“…Sparse features blended with dense features can help develop better-performing 27 models. The results of the PSD case studies in this section provide further evidence for the effectiveness of the NLP-based features in building effective ML models for drug screening.…”

Combining CNNs and Markov-like Models for Facial Landmark Detection with Spatial Consistency Estimates

“…[17] exploits a CNN as a feature extractor to learn the input for the optimization process of an ASM to ultimately detect landmarks. In [10], the authors advocate face detection using the deformable parts model (PDM) combined with a cascade shape regression using multiscale histogram of oriented gradients (HOG) features (also utilized for registration techniques, as in [18]), by incorporating a local refinement for the least accurate landmarks. In [19], a CNN combined with a conditional random field (CRF) were jointly trained to capture the variations due to pose and deformation in order to generate a structured probabilistic prediction of landmark locations.…”

A Medical Image Registration Method Based on Progressive Images

“…Generally, the bulk of medical image registration techniques can be divided into two categories: intensity-based registration and feature-based registration [ 9 – 11 ]. Intensity-based registration plays an important role in the diagnosis and treatment of medical diseases.…”