A Varian 4 MeV isocentric therapy accelerator has been modified to perform also as a CT scanner. The goal is to provide low cost computed tomography capability for use in radiotherapy. The system will have three principal uses. These are (i) to provide 2- and 3-dimensional maps of electron density distribution for CT assisted therapy planning, (ii) to aid in patient set up by providing sectional views of the treatment volume and high contrast scout-mode verification images and (iii) to provide a means for periodically checking the patients anatomical conformation against what was used to generate the original therapy plan. The treatment machine was modified by mounting an array of detectors on a frame bolted to the counter weight end of the gantry in such a manner as to define a 'third generation' CT Scanner geometry. The data gathering is controlled by a Z-80 based microcomputer system which transfers the x-ray transmission data to a general purpose PDP 11/34 for processing. There a series of calibration processes and a logarithmic conversion are performed to get projection data. After reordering the projection data to an equivalent parallel beam sinogram format a convolution algorithm is employed to construct the image from the equivalent parallel projection data. Results of phantom studies have shown a spatial resolution of 2.6 mm and an electron density discrimination of less than 1% which are sufficiently good for accurate therapy planning. Results also show that the system is linear to within the precision of our measurement (≈ .75%) over a wide range of electron densities corresponding to those found in body tissues. Animal and human images are also presented to demonstrate that the system's imaging capability is sufficient to allow the necessary visualization of anatomy.

Modern medical imaging devices are able to generate highly detailed three-dimensional (3D) images of the skeleton. Computerized image processing and analysis methods, combined with real-time volume visualization techniques, can greatly facilitate the interpretation of such images and are increasingly used in surgical planning to aid reconstruction of the skeleton after trauma or disease. Two key challenges are to accurately separate (segment) bone structures or cavities of interest from the rest of the image and to interact with the 3D data in an efficient way. This thesis presents efficient and precise interactive methods for segmenting, visualizing, and analysing 3D computed tomography (CT) images of the skeleton. The methods are validated on real CT datasets and are primarily intended to support planning and evaluation of cranio-maxillofacial (CMF) and orthopedic surgery. Two interactive methods for segmenting the orbit (eye-socket) are introduced. The first method implements a deformable model that is guided and fitted to the orbit via haptic 3D interaction, whereas the second method implements a user-steered volumetric brush that uses distance and gradient information to find exact object boundaries. The thesis also presents a semi-automatic method for measuring 3D angulation changes in wrist fractures. The fractured bone is extracted with interactive mesh segmentation, and the angulation is determined with a technique based on surface registration and RANSAC. Lastly, the thesis presents an interactive and intuitive tool for segmenting individual bones and bone fragments. This type of segmentation is essential for virtual surgery planning, but takes several hours to perform with conventional manual methods. The presented tool combines GPU-accelerated random walks segmentation with direct volume rendering and interactive 3D texture painting to enable quick marking and separation of bone structures. It enables the user to produce an accurate segmentation within a few minutes, thereby removing a major bottleneck in the planning procedure.

Preclinical micro-computed tomography (microCT) images are of utility for 3D morphological bone evaluation, which is of great interest in cancer detection and treatment development. This work introduces a compression strategy for microCTs that allocates specific substances in different Volumes of Interest (Vols). The allocation procedure is conducted by the Hounsfield scale. The Vols are coded independently and then grouped in a single DICOM-compliant file. The proposed method permits the use of different codecs, identifies and transmit data corresponding to a particular substance in the compressed domain without decoding the volume(s), and allows the computation of the 3D morphometry without needing to store or transmit the whole image. The proposed approach reduces the transmitted data in more than 90% when the 3D morphometry evaluation is performed in high density and low density bone. This work can be easily extended to other imaging modalities and applications that work with the Hounsfield scale. (C) 2015 Elsevier Ltd. All rights reserved.

This thesis proposes different models for a variety of applications, such as semantic segmentation, in-the-wild face recognition, microscopy cell counting and detection, standardized re-orientation of 3D ultrasound fetal brain and Magnetic Resonance (MR) cardiac video segmentation. Our approach is to employ the large-scale machine learning models, in particular deep neural networks. Expert knowledge is either mathematically modelled as a differentiable hidden layer in the Artificial Neural Networks, or we tried to break the complex tasks into several small and easy-to-solve tasks. Multi-scale contextual information plays an important role in pixel-wise predic- tion, e.g. semantic segmentation. To capture the spatial contextual information, we present a new block for learning receptive field adaptively by within-layer recurrence. While interleaving with the convolutional layers, receptive fields are effectively enlarged, reaching across the entire feature map or image. The new block can be initialized as identity and inserted into any pre-trained networks, therefore taking benefit from the "pre-train and fine-tuning" paradigm. Current face recognition systems are mostly driven by the success of image classification, where the models are trained to by identity classification. We propose a multi-column deep comparator networks for face recognition. The architecture takes two sets (each contains an arbitrary number of faces) of images or frames as inputs, facial part-based (e.g. eyes, noses) representations of each set are pooled out, dynamically calibrated based on the quality of input images, and further compared with local "experts" in a pairwise way. Unlike the computer vision applications, collecting data and annotation is usually more expensive in biomedical image analysis. Therefore, the models that can be trained with fewer data and weaker annotations are of great importance. We approach the microscopy cell counting and detection based on density estimation, where only central dot annotations are needed. The proposed fully convolutional regression networks are first trained on a synthetic dataset of cell nuclei, later fine-tuned and shown to generalize to real data. In 3D fetal ultrasound neurosonography, establishing a coordinate system over the fetal brain serves as a precursor for subsequent tasks, e.g. localization of anatomical landmarks, extraction of standard clinical planes for biometric assessment of fetal growth, etc. To align brain volumes into a common reference coordinate system, we decompose the complex transformation into several simple ones, which can be easily tackled with Convolutional Neural Networks. The model is therefore designed to leverage the closely related tasks by sharing low-level features, and the task-specific predictions are then combined to reproduce the transformation matrix as the desired output. Finally, we address the problem of MR cardiac video analysis, in which we are interested in assisting clinical diagnosis based on the fine-grained segmentation. To facilitate segmentation, we present one end-to-end trainable model that achieves multi-view structure detection, alignment (standardized re-orientation), and fine- grained segmentation simultaneously. This is motivated by the fact that the CNNs in essence is not rotation equivariance or invariance, therefore, adding the pre-alignment into the end-to-end trainable pipeline can effectively decrease the complexity of segmentation for later stages of the model.

Tissue microarrays (TMAs) are a high-throughput technique that facilitates the survey of very large numbers of tumours, important both in clinical and research applications. However, the assessment of stained TMA sections is laborious and still needs to be carried manually, constituting a bottleneck in the pathologist?s work-flow. This process is also prone to perceptual errors and observer variability.Thus, there is strong motivation for the development of automated quantitative analysis of TMA image data. The analysis of breast TMA sections subjected to nuclear immunostaining begins with the classification of each spot as to the maintype of tissue that it contains, namely tumour, normal, stroma, or fat. Tumour and normal spots are then assigned a so-called quickscore composed of a pair or integer values, the first reflecting the proportion of epithelial nuclei that are stained, and the second reflecting the strength of staining of those nuclei. In this work, an approach was developed to analyse breast TMA spots subjectedto progesterone receptor immunohistochemistry. Spots were classified into their four main types through a method that combined a bag of features approachand classifiers based on either multi-layer perceptrons or latent Dirichlet allocation models. A classification accuracy of 74.6 % was achieved. Tumour and normal spots were scored via an approach that involved the computation of global features formalising the quickscore values used by pathologists, and the use of Gaussian processes for ordinal regression to predict actual quickscores based on global features. Mean absolute errors of 0.888 and 0.779 were achieved in the prediction of the first and second quickscore values, respectively. By setting thresholds on prediction confidence, it was possible to classify and score fractions of spots with substantially higher accuracies and lower mean absolute errors. Amethod for the segmentation of TMA spots into regions of different types was also investigated, to explore the generative nature of latent Dirichlet allocation models.

The quantitative assessment of lymph node size plays an important role in treatment of diseases like cancer. In current clinical practice, lymph nodes are analyzed manually based on very rough measures of long and/or short axis length, which is error prone. In this paper we present a graph-based lymph node segmentation method to enable the computer-aided three-dimensional (3D) assessment of lymph node size. Our method has been validated on 111 cases of enlarged lymph nodes imaged with X-ray computed tomography (CT). For unsigned surface positioning error, Hausdorff distance and Dice coefficient, the mean was around 0.5 mm, under 3.26 mm and above 0.77 respectively. On average, 5.3 seconds were required by our algorithm for the segmentation of a lymph node.

<p>In the last several years, there has been an increased focus on connecting image processing and artificial intelligence. Especially in the field of medical image diagnostics the benefits for such integration is apparent. In this paper we present use of the Common Object Request Broker Architecture (CORBA), as the mean for connecting existing systems for image processing and artificial intelligence. To visualize this, we will use CORBA for connecting Dynamic Imager and JavaCreek. Dynamic Imager is an image processing software, that is especially suitable for setting up and test customized sequences of image processing operations. JavaCreek is an artificial intelligence software based on the Case-Based Reasoning (CBR) theory.</p><p>After connecting the two systems with CORBA, we proceed develop the specific image processing methods for data gathering, and a knowledge base for diagnosis in the artificial intelligence system. The image processing methods and the knowledge base are produced for one special knowledge domain, for visualizing how the proposed system can help in medical image diagnostics.</p><p>The task we use to visualize our approach, is detecting malignancy in breast tumors, from magnetic resonance (MR) images taken over time as contrast agents is injected. This is from a reasonably new method for deciding if a tumor is malignant or benign. All image processing methods and the knowledge base is produced to let the two systems cooperate to find and diagnose tumors.</p><p>The image processing methods, the knowledge model, and the selected software with CORBA connection, was the basis for our system implementation. The implementation was tested with data gathered during the development of the clinical method for determining if a tumor is malignant, from the MR images. In all 127 patient cases was available, where 77 has malignant tumors in the gathered images. The results was then compared with diagnosis methods based on manual detection, and on other image processing methods. Although the found results were promising, there was also found several areas for future work.</p>

The implementation of data mining techniques in the medical area has generated great interest because of its potential for more efficient, economic and robust performance when compared to physicians. In this paper, we focus on the implementation of Multiple-Instance Learning (MIL) in the area of medical image mining, particularly to hard exudates detection in retinal images from diabetic patients. Our proposed approach deals with the highly noisy images that are common in the medical area, improving the detection specificity while keeping the sensitivity as high as possible. We have also investigated the effect of feature selection on system performance. We describe how we implement the idea of MIL on the problem of retinal image mining, discuss the issues that are characteristic of retinal images as well as issues common to other medical image mining problems, and report the results of initial experiments. / Singapore-MIT Alliance (SMA)

In the last several years, there has been an increased focus on connecting image processing and artificial intelligence. Especially in the field of medical image diagnostics the benefits for such integration is apparent. In this paper we present use of the Common Object Request Broker Architecture (CORBA), as the mean for connecting existing systems for image processing and artificial intelligence. To visualize this, we will use CORBA for connecting Dynamic Imager and JavaCreek. Dynamic Imager is an image processing software, that is especially suitable for setting up and test customized sequences of image processing operations. JavaCreek is an artificial intelligence software based on the Case-Based Reasoning (CBR) theory. After connecting the two systems with CORBA, we proceed develop the specific image processing methods for data gathering, and a knowledge base for diagnosis in the artificial intelligence system. The image processing methods and the knowledge base are produced for one special knowledge domain, for visualizing how the proposed system can help in medical image diagnostics. The task we use to visualize our approach, is detecting malignancy in breast tumors, from magnetic resonance (MR) images taken over time as contrast agents is injected. This is from a reasonably new method for deciding if a tumor is malignant or benign. All image processing methods and the knowledge base is produced to let the two systems cooperate to find and diagnose tumors. The image processing methods, the knowledge model, and the selected software with CORBA connection, was the basis for our system implementation. The implementation was tested with data gathered during the development of the clinical method for determining if a tumor is malignant, from the MR images. In all 127 patient cases was available, where 77 has malignant tumors in the gathered images. The results was then compared with diagnosis methods based on manual detection, and on other image processing methods. Although the found results were promising, there was also found several areas for future work.

none