Jim Graham

• DSc
• Consultant at Manchester Imaging Ltd

Introduction Reversal of enamel-only proximal caries by non-invasive treatments is important in preventive dentistry. However, detecting such caries using bitewing radiography is difficult and the subtle patterns are often missed by dental practitioners. Aims To investigate whether the ability of dentists to detect enamel-only proximal caries is en...

Introduction Reversal of enamel-only proximal caries by non-invasive treatments is important in preventive dentistry. However, detecting such caries using bitewing radiography is difficult, and the subtle patterns are often missed by dental practitioners. Aims To investigate whether the ability of dentists to detect enamel-only proximal caries is...

Objective: Corneal confocal microscopy (CCM), an in vivo ophthalmic imaging modality, is a noninvasive and objective imaging biomarker for identifying small nerve fiber damage. We have evaluated the diagnostic performance of previously established CCM parameters to a novel automated measure of corneal nerve complexity called the corneal nerve fibe...

Objective: We describe and evaluate an automated software tool for nerve fibre detection and quantification in corneal confocal microscopy (CCM) images, combining sensitive nerve-fibre detection with morphological descriptors. Method: We have evaluated the tool for quantification of Diabetic Sensorimotor Polyneuropathy (DSPN) using both new and...

Recent developments have made available free software tools that would expedite local quality assurance (QA) in magnetic resonance imaging (MRI), incorporating established methodology. IQWorks is a software package for digital Imaging and communication in medicine image analysis, which was found to have highly appropriate functionality. Key advanta...

Quantitative assessment of small fiber damage is key to the early diagnosis and assessment of progression or regression of diabetic sensorimotor polyneuropathy (DSPN). Intraepidermal nerve fiber density (IENFD) is the current gold standard, but corneal confocal microscopy (CCM), an in vivo ophthalmic imaging modality, has the potential to be a noni...

One of the major factors governing implant success is the quantity and density of the host bone. The aim of this work was to determine whether mandibular bone texture and cortical width measurements on plain radiographs could be associated with implant failure and/or marginal bone loss. A statistical model was built to predict implant failure; it i...

In this chapter, we address the online and real-time segmentation propagation from one example onto similar images. We consider segmentation as a process consisting of two stages: the localization of the anatomy of interest and its boundary delineation. For each stage, we identify and evaluate different potential candidate methods. All methods are...

A method for detecting indications of dental caries in an image of a tooth. The method comprises fitting a model to the image to identify a part of the image representing an outside edge of the tooth. Image variations are analyzed in an area of the image to detect said indications of dental caries, the area of the image representing a part of the t...

Purpose: To assess the diagnostic validity of a fully automated image analysis algorithm of in vivo confocal microscopy images in quantifying corneal subbasal nerves to diagnose diabetic neuropathy. Methods: One hundred eighty-six patients with type 1 and type 2 diabetes mellitus (T1/T2DM) and 55 age-matched controls underwent assessment of neur...

Statistical analysis of motion patterns of body joints is potentially useful for detecting and quantifying pathologies. However, building a statistical motion model across different subjects remains a challenging task, especially for a complex joint like the wrist. We present a novel framework for simultaneous registration and segmentation of multi...

The main recent developments in radiotherapy have focused on improved treatment techniques in order to generate further significant improvements in patient prognosis. There is now an internationally recognised need to improve 3D verification of highly conformal radiotherapy treatments. This is because of the very high dose gradients used in modern...

Previous studies have shown an association between osteoporosis and automatic measurements of mandibular cortical width on dental panoramic radiographs (DPRs). In this study we show that additional image texture features increase this association and propose the combined features as a potential biomarker for osteoporosis. We used an existing datase...

Comparison of dose distributions using the 3D gamma method is anticipated to provide better indicators for the quality assurance process than the 2.5D (stacked 2D slice-by-slice) gamma calculation, especially for advanced radiotherapy technologies. This study compares the accuracy of the 3D and 2.5D gamma calculation methods. 3D and 2.5D gamma calc...

Purpose: To establish intraobserver and interobserver repeatability, agreement, and symmetry of corneal nerve fiber (NF) morphology in healthy subjects using in vivo corneal confocal microscopy. Methods: Nineteen subjects underwent in vivo corneal confocal microscopy (Heidelberg Retinal Tomograph III Rostock Cornea Module) at baseline and 7 days...

We present a novel framework for estimating the 3D poses and shapes of the carpal bones from single view fluoroscopic sequences. A hybrid statistical model representing both the pose and shape variation of the carpal bones is built, based on a number of 3D CT data sets obtained from different subjects at different poses. Given a fluoroscopic sequen...

In this paper we address the problem of automated differential segmentation of the prostate in three dimensional (3-D) magnetic resonance images (MRI) of patients with benign prostatic hyperplasia (BPH). We suggest a framework that consists of two stages: in the first stage, a Random Forest classifier localizes the anatomy of interest. In the secon...

We have applied the method of statistical shape modelling to fluorescence ratio profiles from CGH images. The models are built from a training set of control profiles. Principal Component Analysis is used to extract the most significant modes of variation of the shape. The variability of the training set is expressed in terms of a weighted combinat...

We present a novel framework for inferring 3D carpal bone kinematics and bone shapes from a single view fluoroscopic sequence. A hybrid statistical model representing both the kinematics and shape variation of the carpal bones is built, based on a number of 3D CT data sets obtained from different subjects at different poses. Given a fluoroscopic se...

A novel framework is presented in this paper for simultaneous multi-label segmentation and registration of carpal bones which leads to efficient statistical model building. It combines the Grow Cut segmentation algorithm with rigid image registration for propagating the segmentation of bones to new poses or different individuals. The proposed frame...

Diabetic peripheral neuropathy (DPN) is one of the most common long-term complications of diabetes. The accurate detection and quantification of DPN are important for defining at-risk patients, anticipating deterioration, and assessing new therapies. Current methods of detecting and quantifying DPN, such as neurophysiology, lack sensitivity, requir...

In this paper we address the problem of segmentation propagation from one example onto similar images in an on-line and real-time fashion. As suggested in, we consider segmentation as a process consisting of two stages: the localization of the anatomy of interest and its boundary delineation. For each stage we identify and evaluate different potent...

We present a method to detect damage in time-lapsed, micro-CT images of progressively loaded bone. The method we have developed splits the image into regions and performs registration on each region individually. The procedure is repeated with progressively smaller regions until either a minimum size or a maximum number of levels is reached. The re...

A method of generating data providing a quantitative indication of the presence or absence of disease from a corneal image of a patient. The method comprises processing the image to identify at least one linear structure, processing the at least one identified linear structure to generate data indicating properties of the at least one identified li...

Automated software was used to measure the mandibular cortical width in a large sample of dental radiographs. We determined that cortical thinning normally starts in women at age 42.5 years and accelerates thereafter. We can estimate population referral rates and thus enable cost benefit analyses for osteoporosis detection by dentists. Previous stu...

Corneal Confocal Microscopy (CCM) imaging is a non-invasive surrogate of detecting, quantifying and monitoring diabetic peripheral neuropathy. This paper presents an automated method for detecting nerve-fibres from CCM images using a dual-model detection algorithm and compares the performance to well-established texture and feature detection method...

To analyze the repeatability of measuring nerve fiber length (NFL) from images of the human corneal subbasal nerve plexus using semiautomated software. Images were captured from the corneas of 50 subjects with type 2 diabetes mellitus who showed varying severity of neuropathy, using the Heidelberg Retina Tomograph 3 with Rostock Corneal Module. Sem...

Objectives: The OSTEODENT index is a predicted probability of osteoporosis derived from a combination of an automated analysis of a dental panoramic radiograph and clinical information. This index has been proposed as a suitable case-finding tool for identification of subjects with osteoporosis in primary dental care; however, no data exist on the...

In this paper we report on the results of a systematic performance evaluation of three efficient image segmentation algorithms. namely Graph-Cuts, Random-Walker and Grow-Cut. The evaluation focuses on their function as the computational part of an interactive segmentation system. The implications caused by the human involvement in the overall proce...

We describe improvements to a method of detecting patients at risk of osteoporosis from automatic measurement of the inferior mandibular cortex on panoramic dental tomograms. Previous work had used an Active Shape Model (ASM) to locate the mandibular edges. However, the edge-based ASM has little lateral positioning information and in osteoporotic c...

To determine the cross-sectional association of the osteoporotic status of patients with the number of their teeth, with and without taking into account age and/or smoking. At four centres, the study recruited 665 females aged 45-70 years and the number of teeth was counted for 651 subjects. Bone density was measured at the total hip, femoral neck...

Recently CCM images have shown a great potential to be a non-invasive and hence reiterative surrogate for accurate and reliable detection and quantification of human diabetic neuropathy. This paper presents a compar-ison between two methods to detect the nerves fibres in the CCM images: a well documented line operator (Linop) and a method based on...

Interactive image segmentation is often employed in the context of medical image analysis, as an alternative to automatic and manual image segmentation. In the last decade it has attracted a lot of attention due to the advent of efficient algorithms able to perform in interactive speed even for large three dimensional (3D) images. However, the huma...

To determine if thinning (<3 mm width) of the lower cortical border of the mandible on dental panoramic radiographs, as well as other clinical risk factors, may provide a useful diagnostic test for osteoporosis in young postmenopausal women. Six hundred and fifty-two subjects (age range 45-70 years) were involved in this multi-centre, cross-section...

We describe a novel method of estimating reduced bone mineral density (BMD) from dental panoramic tomograms (DPTs), which show the entire mandible. Careful expert width measurement of the inferior mandibular cortex has been shown to be predictive of BMD in hip and spine osteopenia and osteoporosis. We have implemented a method of automatic measurem...

We describe a novel method of estimating reduced bone mineral density (BMD) from dental panoramic tomograms (DPTs) which show the entire mandible. Careful expert width measurement of the inferior mandibular cortex has been shown to be predictive of BMD in hip and spine osteopenia and osteoporosis. We have implemented a method of automatic measureme...

Point-matching is a widely applied image registration method and many algorithms have been developed. Registration of 2-D electrophoresis gels is an important problem in biological research that presents many of the technical difficulties that beset point-matching: large numbers of points with variable densities, large nonrigid transformations betw...

General dental practitioners use a vast amount of panoramic radiography in their routine clinical work, but valuable information about patients' osteoporotic status is not collected. There are many reasons for this, but one of the prime reasons must be the disruption involved in clinical routine with lengthy manual radiographic assessment. We have...

Benign prostatic hyperplasia (BPH) is a non-cancerous expansion of the prostate, the progress of which can be quantified by measuring the relative volumes of the prostate's peripheral zone and central gland. Here we describe a method of automatic segmentation of both regions of the prostate from MR images using a combination of grey-level voxel cla...

We use point distribution models (PDMs) to investigate lateral asymmetries in the shape of brain ventricles between control subjects and people with schizophrenia. Ventricle surfaces were extracted from T2-weighted MR images and PDMs generated using structural correspondences on the individual surfaces. Using paired linear discriminant analysis we...

Stroke patients may develop personal neglect, peripersonal neglect or both. Four new measures were tested in a sample of 42 right-handed inpatients (25 male; 17 female, median age 72 years). Participants removed keys from a rack, identified grocery items, washed their face, and cleaned a tray. Prior to this, they were classified as: no neglect (15)...

The purpose of this study was to investigate the effects of long-term oestrogen deficiency on mandibular cortical thickness in ovariectomized rats. Twelve female rats either underwent ovariectomy or received a sham-operation. One year later images of the left side of the mandible were made using a standardized radiographic protocol. Manual measurem...

The aim of this study was to develop a system of computerised analysis of digital images of posterior capsule opacification (PCO) that is evidence based, objective and freely available. The paper will present evidence for the reliability and validity of the developed system. The system of PCO analysis was developed considering current published evi...

Recently, proteomics research has become a large growth area in the bio-sciences. Studies involve differential analysis of large sets of 2-D Elec-trophoresis (2-DE) gels. We present a robust and accurate 2-DE gel align-ment algorithm which combines point matching and local image-based re-finement. The algorithm uses a novel combination of Euclidian...

We have carried out an evaluation of feature matching algorithms and developed an automatic method of 2-D electrophoresis gel registration. We show that bi-partite graph matching is not appropriate in this context, and have used a combination of shape context and euclidian distance measures together with closest point correspondence estimation in o...

We have carried out an evaluation of feature matching algorithms and developed an automatic method of 2-D Electrophoresis gel registration. We show that bi-partite graph matching is not appropriate in this context, and have used a combination of shape context and euclidian distance measures together with closest point correspondence estimation in o...

We present results of morphometric analysis of the lateral ventricles of a group of schizophrenic and control subjects to investigate possible shape differences associated with schizophrenia. Our results show shape changes localised to three regions : the temporal horn (its tip near the amygdala, and along its body near the parahippocampal fissure)...

Proteomics research relies heavily on electrophoresis gels, which are complex images containing many protein spots. The identification and quantification of these spots is a bottleneck in the proteomics workflow. We describe a statistical model of protein spot appearance that is both general enough to represent unusual spots, and specific enough to...

Measurement of motion in transabdominal ultrasound sequences of the small bowel is hampered by the unstructured, textured appearance of the images. We investigate a method of measuring Optical Flow in the sequences, based on an analysis of the spatio-temporal frequency spectrum. The method requires no unrealistic assumptions about the distribution...

Protein spot detection is central to the analysis of two-dimensional electrophoresis gel images. There are many commercially available packages, each implementing a protein spot detection algorithm. Despite this, there have been relatively few studies comparing the performance characteristics of the different packages. This is in part due to the fa...

In image analysis of two-dimensional electrophoresis gels, individual spots need to be identified and quantified. Two classes of algorithms are commonly applied to this task. Parametric methods rely on a model, making strong assumptions about spot appearance, but are often insufficiently flexible to adequately represent all spots that may be presen...

Resolving chromosome overlaps is an unsolved problem in automated chromosome analysis. We propose a method that combines evidence from classification and shape, based on trainable shape models. In evaluation using synthesized overlaps, certain cases are resolvable using shape evidence alone, but where this is misleading, classification evidence imp...

In order to construct 3D Point Distribution models (PDMs) of the lateral ventricles of the brains, we need to establish point correspondences among members of a training set. The matched points should, as far as possible, represent physically corresponding features. We use crest lines and crest points on the ventricular surface obtained using the m...

Active shape models (ASMs) have been shown to be a powerful tool to aid the interpretation of images. ASM model parameter estimation is based on the assumption that residuals between model fit and data have a Gaussian distribution. However, in many real applications, specifically those found in the area of medical image analysis, this assumption ma...

Nerve capillary images have a complex textured appearance, which makes segmentation difficult.

Active shape models (ASMs) have been shown to be a powerful tool to aid the interpretation of images. ASM model parameter estimation is based on the assumption that residuals between model fit and data have a Gaussian distribution. However, in many real applications, specifically those found in the area of medical image analysis, this assumption ma...

Point distribution models have been successful in describing the shape constraints on two dimensional objects for shape description and image search. It is often the case that a class of objects to be modelled contains certain features which may be wholly present or absent in different instances. Moustaches on faces are a common example. Here we de...

Point distribution models have been successful in describing the shape constraints on two dimensional objects for shape description and image search. It is often the case that a class of objects to be modelled contains certain features which may be wholly present or absent in di#erent instances. Moustaches on faces are a common example. Here we des...

Microangiopathy is one form of pathology associated with peripheral neuropathy in diabetes. Capillaries imaged by electron microscopy show a complex textured appearance, which makes segmentation difficult. Considerable variation occurs among boundaries manually positioned by human experts. Detection of region boundaries using Active Contour Models...

In an earlier paper [1] we have proposed a shape representation called the CLD (Chord Length Distribution) which possesses many of the often-quoted desirable properties of a shape representation. It also captures shape variability and complements an object location method using belief updating which integrates low-level evidence and shape constrain...

In a normal human cell there are 46 chromosomes which, at appropriate stages of cell division (prophase and metaphase) can be observed as separate objects using high resolution light microscopy. Figure 18.1 shows chromosomes in a metaphase cell. The chromosomes have been stained to exhibit a series of bands along their lengths. Each chromosome also...

A wide range of books on image processing and analysis provide comprehensive descriptions of mathematics and algorithms for image processing practitioners, or introductory material for engineering students. This volume is different in addressing the topic from the point of view of the "user". Standard algorithms, procedures and rules of thumb are e...

We propose and evaluate a mechanism for resolving the segmentation of overlapping chromosomes using trainable models of the expected banding appearance. The models consist of templates of sub-chromosome length band profiles. Candidate chromosome segments are classified according to their responses to the entire set of templates, and matched on the...

Neurological studies are often concerned with identifying abnormalities in brain structure affecting asymmetry between left and right hemispheres. This paper presents techniques which allow measurement and characterisation of differences between neuroanatomic structures due to variation in both shape and receptor distribution. This provides a poten...

In this study we have investigated the application of an Artificial Neural Net classifier to the diagnosis of vascular disease using Doppler ultrasound blood-velocity/time waveforms. A multi-layer perceptron network was trained with waveforms from control subjects and from patients with arterial disease. The diseased cases were confirmed by angiogr...

We present two applications of model based computer vision methods to measurement of image features significant in the diagnosis of diabetic neuropathy. The first involves the location of the boundaries of nerve fascicles in light microscope images. The second involves the segmentation of capillary cell regions using electron microscope images. In...

A method for building flexible shape models is presented in which a shape is represented by a set of labelled points. The technique determines the statistics of the points over a collection of example shapes. The mean positions of the points give an average shape and a number of modes of variation are deter# mined describing the main ways in which...

The usefulness of advanced machine vision techniques for automated inspection is restricted by long development cycles for the inspection software and a lack of general applicability of the resulting system. These difficulties arise because in the majority of cases the particu- lars of the inspection are embedded in the software structure resulting...

In an earlier paper [1] we have proposed a shape representation called the CLD (Chord Length Distribution) which possesses many of the often-quoted desirable properties of a shape representation. It also captures shape variability and complements an object location method using belief updating which integrates low-level evidence and shape constrain...

Model-based vision is firmly established as a robust approach to recognizing and locating known rigid objects in the presence of noise, clutter, and occlusion. It is more problematic to apply model-based methods to images of objects whose appearance can vary, though a number of approaches based on the use of flexible templates have been proposed. T...

This chapter differs from the majority in this book in that its subject matter is the application of computer image interpretation techniques to the analysis of metaphase chromosome spreads. Were we to follow the prescription of the remainder of the book, we might simply publish the code of a computer program together with a list of suitable equipm...

This paper describes a technique for building compact models of the shape and appearance of flexible objects seen in two-dimensional images. The models are derived from the statistics of sets of images of example objects with ‘landmark’ points labelled on each object. Each model consists of a flexible shape template, describing how the landmark poi...

This work presents an approach to the automatic classification of metaphase chromosomes using a multilayer perceptron neural network. Representation of the banding patterns by intuitively defined features is avoided. The inputs to the network are the chromosome size and centromeric index and a coarsely quantized representation of the chromosome ban...

Classification of banded metaphase chromosomes is an important step in automated clinical chromosome analysis. We have conducted a preliminary investigation of the application of artificial neural networks to this process, making use of a natural representation of the banding pattern. Two different network architectures have been compared: the Koho...