Repeatable, quantitative assessment of intervertebral disc pathology requires accurate localization and labeling of the lumbar region discs. To that end, we propose a two-level probabilistic model for such disc localization and labeling. Our model integrates both pixel-level information, such as appearance, and object-level information, such as relative location. Utilizing both levels of information adds robustness to the ambiguous disc intensity signature and high structure variation. Yet, we are able to do efficient (and convergent) localization and labeling with generalized expectation-maximization. We present accurate results on 20 normal cases (96%) and a promising extension to a pathology case.

Abstract-Backbone anatomical structure detection and labeling is a necessary step for various analysis tasks of the vertebral column. Appearance, shape and geometry measurements are necessary for abnormality detection locally at each disc and vertebrae (such as herniation) as well as globally for the whole spine (such as spinal scoliosis). We propose a two-level probabilistic model for the localization of discs from clinical magnetic resonance imaging (MRI) data that captures both pixel-and object-level features. Using a Gibbs distribution, we model appearance and spatial information at the pixel level, and at the object level, we model the spatial distribution of the discs and the relative distances between them. We use generalized expectation-maximization for optimization, which achieves efficient convergence of disc labels. Our two-level model allows the assumption of conditional independence at the pixel-level to enhance efficiency while maintaining robustness. We use a dataset that contains 105 MRI clinical normal and abnormal cases for the lumbar area. We thoroughly test our model and achieve encouraging results on normal and abnormal cases.

Content-based image retrieval systems for digital pathology require sub-image retrieval rather than the whole image retrieval for the system to be of clinical use. Digital pathology images are huge in size and thus the pathologist is interested in retrieving specific structures from the whole images in the database along with the previous diagnosis of the retrieved sub-image. We propose a content-based sub-image retrieval system (sCBIR) framework for high resolution digital pathology images. We utilize scale-invariant feature extraction and present an efficient and robust searching mechanism for indexing the images as well as for query execution of sub-image retrieval. We present a working sCBIR system and show results of testing our system on a set of queries for specific structures of interest for pathologists in clinical use. The outcomes of the sCBIR system are compared to manual search and there is an 80% match in the top five searches.

Lumbar vertebral fractures vary greatly in types and causes and usually result from severe trauma or pathological conditions such as osteoporosis. Lumbar wedge compression fractures are amongst the most common ones where the vertebra is severely compressed forming a wedge shape and causing pain and pressure on the nerve roots and the spine. Since vertebral segmentation is the first step in any automated diagnosis task, we present a fully automated method for robustly localizing and segmenting the vertebrae for preparation of vertebral fracture diagnosis. Our segmentation method consists of five main steps towards the CAD(Computer-Aided Diagnosis) system: 1) Localization of the intervertebral discs. 2) Localization of the vertebral skeleton. 3) Segmentation of the individual vertebra. 4) Detection of the vertebrae center line and 5) Detection of the vertebrae major boundary points. Our segmentation results are promising with an average error of 1.5mm (modified Hausdorff distance metric) on 50 clinical CT cases i.e. a total of 250 lumbar vertebrae. We also present promising preliminary results for automatic wedge compression fracture diagnosis on 15 cases, 7 of which have one or more vertebral compression fracture, and obtain an accuracy of 97.33%.

We presented a fully automated CAD system that seamlessly integrates within the clinical work flow of the radiologist. Our clinically motivated features resulted in a great performance of both the supervised and unsupervised learners that we utilize to validate our CAD system. Our CAD system results are promising to serve in clinical applications after extensive validation.

We achieve high accuracy for detection of abnormal discs using our proposed model that incorporates disc appearance, location, and context. We show the extendability of our proposed model to subsequent diagnosis tasks specific to each intervertebral disc abnormality such as desiccation and herniation.

Digital pathology has been clinically approved for over a decade to replace traditional methods of diagnosis. Many challenges appear when digitising the whole slide scan into high resolution images including memory and time management. Whole slide images require huge memory space if the tissue is not pre-localised for the scanner. The authors propose a set of clinically motivated features representing colour, intensity, texture and location to segment and localise the tissue from the whole slide image. This step saves both the scanning time and the required memory space. On average, it reduces scanning time up to 40% depending on the tissue type. The authors propose, using unsupervised learning, to segment and localise tissue by clustering. Unlike supervised methods, this method does not require the ground truth which is time consuming for domain experts. The authors proposed method achieves an average of 96% localisation accuracy on a large dataset. Moreover, the authors outperform the previously proposed supervised learning results on the same data.

Abstract. Lower Back Pain (LBP) is the second most common neurological ailment in the United States after the headache. It costs over $100 Billion annually in treatment and related rehabilitation costs including worker compensation. In fact, it is the most common reason for lost wages and missed work days. Degenerative Disc Disease (DDD) is the major abnormality that causes LBP. Moreover, Magnetic Resonance Imaging (MRI) test is the main clinically approved noninvasive imaging modality for the diagnosis of DDD. However, there is over 50% inter-and intra-observer variability in the MRI interpretation that urges the need for standardized mechanisms in MRI interpretation. In this paper, we propose a Computer Aided Diagnosis (CAD) System for Disc Degenerative Disease detection from clinical Magnetic Resonance Imaging (MRI). This CAD produces a reproducible and clinically accurate diagnosis of the DDD for lumbar spine. We design a classifier to automatically detect degenerated disc (also clinically known as Herniation) using shape potentials. We extract these shape potentials by jointly applying an active shape model (ASM) and a gradient vector flow snake model (GVF-snake). The ASM roughly segments the disc by the detection of a certain point distribution around the disc. Then, we use this point distribution to initialize a GVF-snake model to delineate the posterior disc segment. We then extract the set of shape potentials for our Gibbs-based classifier. The whole work flow is fully automated given the full clinical MRI. We validate our model on 65 clinical MRI cases (6 discs each) and achieve an average of 93.9% classification accuracy. Our shape-based classifier is superior in classification accuracy compared to the state-of-the-art work on this problem that reports 86% and 91% on 34 and 33 cases, respectively.