Miles N. Wernick

A support vector machine approach for detection of microcalcifications

We investigate an approach based on support vector machines (SVMs) for detection of microcalcification (MC) clusters in digital mammograms, and propose a successive enhancement learning scheme for improved performance. SVM is a machine-learning method, based on the principle of structural risk minimization, which performs well when applied to data outside the training set. We formulate MC detection as a supervised-learning problem and apply SVM to develop the detection algorithm. We use the SVM to detect at each location in the image whether an MC is present or not. We tested the proposed method using a database of 76 clinical mammograms containing 1120 MCs. We use free-response receiver operating characteristic curves to evaluate detection performance, and compare the proposed algorithm with several existing methods. In our experiments, the proposed SVM framework outperformed all the other methods tested. In particular, a sensitivity as high as 94% was achieved by the SVM method at an error rate of one false-positive cluster per image. The ability of SVM to outperform several well-known methods developed for the widely studied problem of MC detection suggests that SVM is a promising technique for object detection in a medical imaging application.

A similarity learning approach to content-based image retrieval: application to digital mammography

In this paper, we describe an approach to content-based retrieval of medical images from a database, and provide a preliminary demonstration of our approach as applied to retrieval of digital mammograms. Content-based image retrieval (CBIR) refers to the retrieval of images from a database using information derived from the images themselves, rather than solely from accompanying text indices. In the medical-imaging context, the ultimate aim of CBIR is to provide radiologists with a diagnostic aid in the form of a display of relevant past cases, along with proven pathology and other suitable information. CBIR may also be useful as a training tool for medical students and residents. The goal of information retrieval is to recall from a database information that is relevant to the users query. The most challenging aspect of CBIR is the definition of relevance (similarity), which is used to guide the retrieval machine. In this paper, we pursue a new approach, in which similarity is learned from training examples provided by human observers. Specifically, we explore the use of neural networks and support vector machines to predict the users notion of similarity. Within this framework we propose using a hierarchal learning approach, which consists of a cascade of a binary classifier and a regression module to optimize retrieval effectiveness and efficiency. We also explore how to incorporate online human interaction to achieve relevance feedback in this learning framework. Our experiments are based on a database consisting of 76 mammograms, all of which contain clustered microcalcifications (MCs). Our goal is to retrieve mammogram images containing similar MC clusters to that in a query. The performance of the retrieval system is evaluated using precision-recall curves computed using a cross-validation procedure. Our experimental results demonstrate that: 1) the learning framework can accurately predict the perceptual similarity reported by human observers, thereby serving as a basis for CBIR; 2) the learning-based framework can significantly outperform a simple distance-based similarity metric; 3) the use of the hierarchical two-stage network can improve retrieval performance; and 4) relevance feedback can be effectively incorporated into this learning framework to achieve improvement in retrieval precision based on online interaction with users; and 5) the retrieved images by the network can have predicting value for the disease condition of the query.

Multiple-image radiography

Conventional radiography produces a single image of an object by measuring the attenuation of an x-ray beam passing through it. When imaging weakly absorbing tissues, x-ray attenuation may be a suboptimal signature of disease-related information. In this paper we describe a new phase-sensitive imaging method, called multiple-image radiography (MIR), which is an improvement on a prior technique called diffraction-enhanced imaging (DEI). This paper elaborates on our initial presentation of the idea in Wernick et al (2002 Proc. Int. Symp. Biomed. Imaging pp 129-32). MIR simultaneously produces several images from a set of measurements made with a single x-ray beam. Specifically, MIR yields three images depicting separately the effects of refraction, ultra-small-angle scatter and attenuation by the object. All three images have good contrast, in part because they are virtually immune from degradation due to scatter at higher angles. MIR also yields a very comprehensive object description, consisting of the angular intensity spectrum of a transmitted x-ray beam at every image pixel, within a narrow angular range. Our experiments are based on data acquired using a synchrotron light source; however, in preparation for more practical implementations using conventional x-ray sources, we develop and evaluate algorithms designed for Poisson noise, which is characteristic of photon-limited imaging. The results suggest that MIR is capable of operating at low photon count levels, therefore the method shows promise for use with conventional x-ray sources. The results also show that, in addition to producing new types of object descriptions, MIR produces substantially more accurate images than its predecessor, DEI. MIR results are shown in the form of planar images of a phantom and a biological specimen. A preliminary demonstration of the use of MIR for computed tomography is also presented.

Machine Learning in Medical Imaging

This article will discuss very different ways of using machine learning that may be less familiar, and we will demonstrate through examples the role of these concepts in medical imaging. Although the term machine learning is relatively recent, the ideas of machine learning have been applied to medical imaging for decades, perhaps most notably in the areas of computer-aided diagnosis (CAD) and functional brain mapping. We will not attempt in this brief article to survey the rich literature of this field. Instead our goals will be 1) to acquaint the reader with some modern techniques that are now staples of the machine-learning field and 2) to illustrate how these techniques can be employed in various ways in medical imaging.

Relevance vector machine for automatic detection of clustered microcalcifications

Clustered microcalcifications (MC) in mammograms can be an important early sign of breast cancer in women. Their accurate detection is important in computer-aided detection (CADe). In this paper, we propose the use of a recently developed machine-learning technique - relevance vector machine (RVM) - for detection of MCs in digital mammograms. RVM is based on Bayesian estimation theory, of which a distinctive feature is that it can yield a sparse decision function that is defined by only a very small number of so-called relevance vectors. By exploiting this sparse property of the RVM, we develop computerized detection algorithms that are not only accurate but also computationally efficient for MC detection in mammograms. We formulate MC detection as a supervised-learning problem, and apply RVM as a classifier to determine at each location in the mammogram if an MC object is present or not. To increase the computation speed further, we develop a two-stage classification network, in which a computationally much simpler linear RVM classifier is applied first to quickly eliminate the overwhelming majority, non-MC pixels in a mammogram from any further consideration. The proposed method is evaluated using a database of 141 clinical mammograms (all containing MCs), and compared with a well-tested support vector machine (SVM) classifier. The detection performance is evaluated using free-response receiver operating characteristic (FROC) curves. It is demonstrated in our experiments that the RVM classifier could greatly reduce the computational complexity of the SVM while maintaining its best detection accuracy. In particular, the two-stage RVM approach could reduce the detection time from 250 s for SVM to 7.26 s for a mammogram (nearly 35-fold reduction). Thus, the proposed RVM classifier is more advantageous for real-time processing of MC clusters in mammograms.

Fast spatio-temporal image reconstruction for dynamic PET

In tomographic imaging, dynamic images are typically obtained by reconstructing the frames of a time sequence independently, one by one. A disadvantage of this frame-by-frame reconstruction approach is that it fails to account. For temporal correlations in the signal. Ideally, one should treat the entire image sequence as a single spatio-temporal signal. However, the resulting reconstruction task becomes computationally intensive. Fortunately, as the authors show in this paper, the spatio-temporal reconstruction problem call be greatly simplified by first applying a temporal Karhunen-Loeve (KL) transformation to the imaging equation. The authors show that if the regularization operator is chosen to be separable into space and time components, penalized weighted least squares reconstruction of the entire image sequence is approximately equivalent to frame-by-frame reconstruction in the space-KL domain. By this approach, spatio-temporal reconstruction can be achieved at reasonable computational cost. One can achieve further computational savings by discarding high-order KL components to avoid reconstructing them. Performance of the method is demonstrated through statistical evaluations of the bias-variance tradeoff obtained by computer simulation reconstruction.

Preliminary assessment of extrastriatal dopamine d-2 receptor binding in the rodent and nonhuman primate brains using the high affinity radioligand, 18F-fallypride

We have identified the value of 18F-fallypride [(S)-N-[(1-allyl-2-pyrrolidinyl)methyl]-5-(3-[18F]fluoropropyl)-2, 3-dimethoxybenzamide], as a dopamine D-2 receptor radiotracer for the study of striatal and extrastriatal receptors. Fallypride exhibits high affinities for D-2 and D-3 subtypes and low affinity for D-4 (3H-spiperone IC50s: D-2 = 0.05 nM [rat striata], D-3 = 0.30 nM [SF9 cell lines, rat recombinant], and D-4 = 240 nM [CHO cell lines, human recombinant]). Biodistribution in the rat brain showed localization of 18F-fallypride in striata and extrastriatal regions such as the frontal cortex, parietal cortex, amygdala, hippocampus, thalamus, and hypothalamus. In vitro autoradiographic studies in sagittal slices of the rat brain showed localization of 18F-fallypride in striatal and several extrastriatal regions, including the medulla. Positron emission tomography (PET) experiments with 18F-fallypride in male rhesus monkeys were carried out in a PET VI scanner. In several PET experiments, apart from the specific binding seen in the striatum, specific binding of 18F-fallypride was also identified in extracellular regions (in a lower brain slice, possibly the thalamus). Specific binding in the extrastriata was, however, significantly lower compared with that observed in the striata of the monkeys (extrastriata/cerebellum = 2, striata/cerebellum = 10). Postmortem analysis of the monkey brain revealed significant 18F-fallypride binding in the striata, whereas binding was also observed in extrastriatal regions such as the thalamus, cortical areas, and brain stem.

Supervised and unsupervised methods for prostate cancer segmentation with multispectral MRI

PURPOSE Magnetic resonance imaging (MRI) has been proposed as a promising alternative to transrectal ultrasound for the detection and localization of prostate cancer and fusing the information from multispectral MR images is currently an active research area. In this study, the goal is to develop automated methods that combine the pharmacokinetic parameters derived from dynamic contrast enhanced (DCE) MRI with quantitative T2 MRI and diffusion weighted imaging (DWI) in contrast to most of the studies which were performed with human readers. The main advantages of the automated methods are that the observer variability is removed and easily reproducible results can be efficiently obtained when the methods are applied to a test data. The goal is also to compare the performance of automated supervised and unsupervised methods for prostate cancer localization with multispectral MRI. METHODS The authors use multispectral MRI data from 20 patients with biopsy-confirmed prostate cancer patients, and the image set consists of parameters derived from T2, DWI, and DCE-MRI. The authors utilize large margin classifiers for prostate cancer segmentation and compare them to an unsupervised method the authors have previously developed. The authors also develop thresholding schemes to tune support vector machines (SVMs) and their probabilistic counterparts, relevance vector machines (RVMs), for an improved performance with respect to a selected criterion. Moreover, the authors apply a thresholding method to make the unsupervised fuzzy Markov random fields method fully automatic. RESULTS The authors have developed a supervised machine learning method that performs better than the previously developed unsupervised method and, additionally, have found that there is no significant difference between the SVM and RVM segmentation results. The results also show that the proposed methods for threshold selection can be used to tune the automated segmentation methods to optimize results for certain criteria such as accuracy or sensitivity. The test results of the automated algorithms indicate that using multispectral MRI improves prostate cancer segmentation performance when compared to single MR images, a result similar to the human reader studies that were performed before. CONCLUSIONS The automated methods presented here can help diagnose and detect prostate cancer, and improve segmentation results. For that purpose, multispectral MRI provides better information about cancer and normal regions in the prostate when compared to methods that use single MRI techniques; thus, the different MRI measurements provide complementary information in the automated methods. Moreover, the use of supervised algorithms in such automated methods remain a good alternative to the use of unsupervised algorithms.

Extraction of extinction, refraction and absorption properties in diffraction enhanced imaging

Diffraction enhanced imaging is a radiographic technique that derives contrast from an objects x-ray absorption, refraction gradient and small angle scatter properties (extinction). In prior work, images obtained using two analyser settings were combined to obtain refraction angle and apparent absorption images. A more general method of determining independently the refraction, absorption and extinction of the object is presented. This approach has been used to model the transmission, refraction and scatter distribution of the sample and to visualize these three physical phenomena separately.

Prostate Cancer Localization With Multispectral MRI Using Cost-Sensitive Support Vector Machines and Conditional Random Fields

Prostate cancer is a leading cause of cancer death for men in the United States. Fortunately, the survival rate for early diagnosed patients is relatively high. Therefore, in vivo imaging plays an important role for the detection and treatment of the disease. Accurate prostate cancer localization with noninvasive imaging can be used to guide biopsy, radiotheraphy, and surgery as well as to monitor disease progression. Magnetic resonance imaging (MRI) performed with an endorectal coil provides higher prostate cancer localization accuracy, when compared to transrectal ultrasound (TRUS). However, in general, a single type of MRI is not sufficient for reliable tumor localization. As an alternative, multispectral MRI, i.e., the use of multiple MRI-derived datasets, has emerged as a promising noninvasive imaging technique for the localization of prostate cancer; however almost all studies are with human readers. There is a significant inter and intraobserver variability for human readers, and it is substantially difficult for humans to analyze the large dataset of multispectral MRI. To solve these problems, this study presents an automated localization method using cost-sensitive support vector machines (SVMs) and shows that this method results in improved localization accuracy than classical SVM. Additionally, we develop a new segmentation method by combining conditional random fields (CRF) with a cost-sensitive framework and show that our method further improves cost-sensitive SVM results by incorporating spatial information. We test SVM, cost-sensitive SVM, and the proposed cost-sensitive CRF on multispectral MRI datasets acquired from 21 biopsy-confirmed cancer patients. Our results show that multispectral MRI helps to increase the accuracy of prostate cancer localization when compared to single MR images; and that using advanced methods such as cost-sensitive SVM as well as the proposed cost-sensitive CRF can boost the performance significantly when compared to SVM.