Registering histologic and MR images of prostate for image-based cancer detection

Semi-automatic deformable registration of prostate MR images to pathological slices

PURPOSE

To present a semi-automatic deformable registration algorithm for co-registering T2-weighted (T2w) images of the prostate with whole-mount pathological sections of prostatectomy specimens.

MATERIALS AND METHODS

Twenty-four patients underwent 1.5 Tesla (T) endorectal MR imaging before radical prostatectomy with whole-mount step-section pathologic analysis of surgical specimens. For each patient, the T2w imaging containing the largest area of tumor was manually matched with the corresponding pathologic slice. The prostate was co-registered using a free-form deformation (FFD) algorithm based on B-splines. Registration quality was assessed through differences between prostate diameters measured in right-left (RL) and anteroposterior (AP) directions on T2w images and pathologic slices and calculation of the Dice similarity coefficient, D, for the whole prostate (WP), the peripheral zone (PZ) and the transition zone (TZ).

RESULTS

The mean differences in diameters measured on pathology and MR imaging in the RL direction and the AP direction were 0.49 cm and -0.63 cm, respectively, before registration and 0.10 cm and -0.11 cm, respectively, after registration. The mean D values for the WP, PZ and TZ, were 0.76, 0.65, and 0.77, respectively, before registration and increased to 0.91, 0.76, and 0.85, respectively, after registration. The improvements in D were significant for all three tissues (P < 0.001 for all).

CONCLUSION

The proposed semi-automatic method enabled successful co-registration of anatomical prostate MR images to pathologic slices.

Determining histology-MRI slice correspondences for defining MRI-based disease signatures of prostate cancer

Mapping the spatial disease extent in a certain anatomical organ/tissue from histology images to radiological images is important in defining the disease signature in the radiological images. One such scenario is in the context of men with prostate cancer who have had pre-operative magnetic resonance imaging (MRI) before radical prostatectomy. For these cases, the prostate cancer extent from ex vivo whole-mount histology is to be mapped to in vivo MRI. The need for determining radiology-image-based disease signatures is important for (a) training radiologist residents and (b) for constructing an MRI-based computer aided diagnosis (CAD) system for disease detection in vivo. However, a prerequisite for this data mapping is the determination of slice correspondences (i.e. indices of each pair of corresponding image slices) between histological and magnetic resonance images. The explicit determination of such slice correspondences is especially indispensable when an accurate 3D reconstruction of the histological volume cannot be achieved because of (a) the limited tissue slices with unknown inter-slice spacing, and (b) obvious histological image artifacts (tissue loss or distortion). In the clinic practice, the histology-MRI slice correspondences are often determined visually by experienced radiologists and pathologists working in unison, but this procedure is laborious and time-consuming. We present an iterative method to automatically determine slice correspondence between images from histology and MRI via a group-wise comparison scheme, followed by 2D and 3D registration. The image slice correspondences obtained using our method were compared with the ground truth correspondences determined via consensus of multiple experts over a total of 23 patient studies. In most instances, the results of our method were very close to the results obtained via visual inspection by these experts.

A PET/CT Directed, 3D Ultrasound-Guided Biopsy System for Prostate Cancer

Prostate cancer affects 1 in 6 men in the USA. Systematic transrectal ultrasound (TRUS)-guided biopsy is the standard method for a definitive diagnosis of prostate cancer. However, this "blind" biopsy approach can miss at least 20% of prostate cancers. In this study, we are developing a PET/CT directed, 3D ultrasound image-guided biopsy system for improved detection of prostate cancer. In order to plan biopsy in three dimensions, we developed an automatic segmentation method based wavelet transform for 3D TRUS images of the prostate. The segmentation was tested in five patients with a DICE overlap ratio of more than 91%. In order to incorporate PET/CT images into ultrasound-guided biopsy, we developed a nonrigid registration algorithm for TRUS and PET/CT images. The registration method has been tested in a prostate phantom with a target registration error (TRE) of less than 0.4 mm. The segmentation and registration methods are two key components of the multimodality molecular image-guided biopsy system.

DRAMMS: deformable registration via attribute matching and mutual-saliency weighting

A general-purpose deformable registration algorithm referred to as "DRAMMS" is presented in this paper. DRAMMS adds to the literature of registration methods that bridge between the traditional voxel-wise methods and landmark/feature-based methods. In particular, DRAMMS extracts Gabor attributes at each voxel and selects the optimal components, so that they form a highly distinctive morphological signature reflecting the anatomical context around each voxel in a multi-scale and multi-resolution fashion. Compared with intensity or mutual-information based methods, the high-dimensional optimal Gabor attributes render different anatomical regions relatively distinctively identifiable and therefore help establish more accurate and reliable correspondence. Moreover, the optimal Gabor attribute vector is constructed in a way that generalizes well, i.e., it can be applied to different registration tasks, regardless of the image contents under registration. A second characteristic of DRAMMS is that it is based on a cost function that weights different voxel pairs according to a metric referred to as "mutual-saliency", which reflects the uniqueness (reliability) of anatomical correspondences implied by the tentative transformation. As a result, image voxels do not contribute equally to the optimization process, as in most voxel-wise methods, or in a binary selection fashion, as in most landmark/feature-based methods. Instead, they contribute according to a continuously-valued mutual-saliency map, which is dynamically updated during the algorithm's evolution. The general applicability and accuracy of DRAMMS are demonstrated by experiments in simulated images, inter-subject images, single-/multi-modality images, and longitudinal images, from human and mouse brains, breast, heart, and prostate.

Sampling the spatial patterns of cancer: optimized biopsy procedures for estimating prostate cancer volume and Gleason Score

Prostate biopsy is the current gold-standard procedure for prostate cancer diagnosis. Existing prostate biopsy procedures have been mostly focusing on detecting cancer presence. However, they often ignore the potential use of biopsy to estimate cancer volume (CV) and Gleason Score (GS, a cancer grade descriptor), the two surrogate markers for cancer aggressiveness and the two crucial factors for treatment planning. To fill up this vacancy, this paper assumes and demonstrates that, by optimally sampling the spatial patterns of cancer, biopsy procedures can be specifically designed for estimating CV and GS. Our approach combines image analysis and machine learning tools in an atlas-based population study that consists of three steps. First, the spatial distributions of cancer in a patient population are learned, by constructing statistical atlases from histological images of prostate specimens with known cancer ground truths. Then, the optimal biopsy locations are determined in a feature selection formulation, so that biopsy outcomes (either cancer presence or absence) at those locations could be used to differentiate, at the best rate, between the existing specimens having different (high vs. low) CV/GS values. Finally, the optimized biopsy locations are utilized to estimate whether a new-coming prostate cancer patient has high or low CV/GS values, based on a binary classification formulation. The estimation accuracy and the generalization ability are evaluated by the classification rates and the associated receiver-operating-characteristic (ROC) curves in cross validations. The optimized biopsy procedures are also designed to be robust to the almost inevitable needle displacement errors in clinical practice, and are found to be robust to variations in the optimization parameters as well as the training populations.

Multispectral analysis of multimodal images

INTRODUCTION

An increasing number of multimodal images represent a valuable increase in available image information, but at the same time it complicates the extraction of diagnostic information across the images. Multispectral analysis (MSA) has the potential to simplify this problem substantially as unlimited number of images can be combined, and tissue properties across the images can be extracted automatically.

MATERIALS AND METHODS

We have developed a software solution for MSA containing two algorithms for unsupervised classification, an EM-algorithm finding multinormal class descriptions and the k-means clustering algorithm, and two for supervised classification, a Bayesian classifier using multinormal class descriptions and a kNN-algorithm. The software has an efficient user interface for the creation and manipulation of class descriptions, and it has proper tools for displaying the results.

RESULTS

The software has been tested on different sets of images. One application is to segment cross-sectional images of brain tissue (T1- and T2-weighted MR images) into its main normal tissues and brain tumors. Another interesting set of images are the perfusion maps and diffusion maps, derived images from raw MR images. The software returns segmentations that seem to be sensible.

DISCUSSION

The MSA software appears to be a valuable tool for image analysis with multimodal images at hand. It readily gives a segmentation of image volumes that visually seems to be sensible. However, to really learn how to use MSA, it will be necessary to gain more insight into what tissues the different segments contain, and the upcoming work will therefore be focused on examining the tissues through for example histological sections.

Nonrigid Registration of Brain Tumor Resection MR Images Based on Joint Saliency Map and Keypoint Clustering

This paper proposes a novel global-to-local nonrigid brain MR image registration to compensate for the brain shift and the unmatchable outliers caused by the tumor resection. The mutual information between the corresponding salient structures, which are enhanced by the joint saliency map (JSM), is maximized to achieve a global rigid registration of the two images. Being detected and clustered at the paired contiguous matching areas in the globally registered images, the paired pools of DoG keypoints in combination with the JSM provide a useful cluster-to-cluster correspondence to guide the local control-point correspondence detection and the outlier keypoint rejection. Lastly, a quasi-inverse consistent deformation is smoothly approximated to locally register brain images through the mapping the clustered control points by compact support radial basis functions. The 2D implementation of the method can model the brain shift in brain tumor resection MR images, though the theory holds for the 3D case.