Thomas Fevens

Classification of Breast Cancer Malignancy Using Cytological Images of Fine Needle Aspiration Biopsies

Classification of Breast Cancer Malignancy Using Cytological Images of Fine Needle Aspiration Biopsies According to the World Health Organization (WHO), breast cancer (BC) is one of the most deadly cancers diagnosed among middle-aged women. Precise diagnosis and prognosis are crucial to reduce the high death rate. In this paper we present a framework for automatic malignancy grading of fine needle aspiration biopsy tissue. The malignancy grade is one of the most important factors taken into consideration during the prediction of cancer behavior after the treatment. Our framework is based on a classification using Support Vector Machines (SVM). The SVMs presented here are able to assign a malignancy grade based on preextracted features with the accuracy up to 94.24%. We also show that SVMs performed best out of four tested classifiers.

Absorbing Boundary Conditions for the Schrödinger Equation

A large number of differential equation problems which admit traveling waves are usually defined on very large or infinite domains. To numerically solve these problems on smaller subdomains of the original domain, artificial boundary conditions must be defined for these subdomains. One type of artificial boundary condition which can minimize the size of such subdomains is the absorbing boundary condition. The imposition of absorbing boundary conditions is a technique used to reduce the necessary spatial domain when numerically solving partial differential equations that admit traveling waves. Such absorbing boundary conditions have been extensively studied in the context of hyperbolic wave equations. In this paper, general absorbing boundary conditions will be developed for the Schrodinger equation with one spatial dimension, using group velocity considerations. Previously published absorbing boundary conditions will be shown to reduce to special cases of this absorbing boundary condition. The well-posedness of the initial boundary value problem of the absorbing boundary condition, coupled to the interior Schrodinger equation, will also be discussed. Extension of the general absorbing boundary condition to higher spatial dimensions will be demonstrated. Numerical simulations using initial single Gaussian, double Gaussian, and a narrow Gaussian pulse distributions will be given, with comparision to exact solutions, to demonstrate the reflectivity properties of various orders of the absorbing boundary condition.

Computer-Aided Breast Cancer Diagnosis Based on the Analysis of Cytological Images of Fine Needle Biopsies

The effectiveness of the treatment of breast cancer depends on its timely detection. An early step in the diagnosis is the cytological examination of breast material obtained directly from the tumor. This work reports on advances in computer-aided breast cancer diagnosis based on the analysis of cytological images of fine needle biopsies to characterize these biopsies as either benign or malignant. Instead of relying on the accurate segmentation of cell nuclei, the nuclei are estimated by circles using the circular Hough transform. The resulting circles are then filtered to keep only high-quality estimations for further analysis by a support vector machine which classifies detected circles as correct or incorrect on the basis of texture features and the percentage of nuclei pixels according to a nuclei mask obtained using Otsus thresholding method. A set of 25 features of the nuclei is used in the classification of the biopsies by four different classifiers. The complete diagnostic procedure was tested on 737 microscopic images of fine needle biopsies obtained from patients and achieved 98.51% effectiveness. The results presented in this paper demonstrate that a computerized medical diagnosis system based on our method would be effective, providing valuable, accurate diagnostic information.

Algorithms for Computing Geometric Measures of Melodic Similarity

Greg Aloupis, Thomas Fevens, Stefan Langerman, Tomomi Matsui, Antonio Mesa, Yurai Nunez, David Rappaport, and Godfried Toussaint *School of Computer Science, McGill University 3480 University Street Montreal, Quebec, Canada H3A 2A7 {athens,godfried}@cs.mcgill.ca †Department of Computer Science and Software Engineering, Concordia University 1455 de Maisonneuve Boulevard West Montreal, Quebec, Canada H3G 1M8 fevens@cs.concordia.ca ‡ Chercheur Qualifie du FNRS, Departement d’Informatique, Universite Libre de Bruxelles CP212 Boulevard du Triomphe, 1050 Bruxelles, Belgium Stefan.Langerman@ulb.ac.be §Department of Mathematical Informatics Graduate School of Information Science and Technology, University of Tokyo 7-3-1 Hongo, Bunkyo-ku, Tokyo, 113-8656 Japan tomomi@misojiro.t.u-tokyo.ac.jp ¶Departamento Ciencias de la Computacion Facultad de Matematica y Computacion, Universidad de La Habana San Lazaro y L, Vedado 10400 Ciudad de La Habana, Cuba tonymesa@matcom.uh.cu, yurainr@yahoo.com **School of Computing, Queen’s University Kingston, Ontario, Canada K7L 3N6 daver@cs.queensu.ca Algorithms for Computing Geometric Measures of Melodic Similarity

Semi-automatic computer aided lesion detection in dental X-rays using variational level set

A semi-automatic lesion detection framework is proposed to detect areas of lesions from periapical dental X-rays using level set method. In this framework, first, a new proposed competitive coupled level set method is used to segment the image into three pathologically meaningful regions using two coupled level set functions. Tailored for the dental clinical setting, a two-stage clinical segmentation acceleration scheme is used. The method uses a trained support vector machine (SVM) classifier to provide an initial contour for two coupled level sets. Then, based on the segmentation results, an analysis scheme is applied. Firstly, the scheme builds an uncertainty map from which those areas with radiolucent will be automatically emphasized by a proposed color emphasis scheme. Those radiolucent in the teeth or jaw usually suggested possible lesions. Secondly, the scheme employs a method based on the average intensity profile to isolate the teeth and locate two types of lesions: periapical lesion (PL) and bifurcation lesion (BL). Experimental results show that our proposed segmentation method is able to segment the image into pathological meaningful regions for further analysis; our proposed framework is able to automatically provide direct visual cues for the lesion detection; and when given the orientation of the teeth, it is able to automatically locate the PL and BL with a seriousness level marked for further dental diagnosis. When used in the clinical setting, the framework enables dentist to improve interpretation and to focus their attention on critical areas.

Local algorithms for dominating and connected dominating sets of unit disk graphs with location aware nodes

Many protocols in ad-hoc networks use dominating and connected dominating sets, for example for broadcasting and routing. For large ad hoc networks the construction of such sets should be local in the sense that each node of the network should make decisions based only on the information obtained from nodes located a constant number of hops from it. In this paper we use the location awareness of the network, i.e. the knowledge of position of nodes in the plane to provide local, constant approximation, deterministic algorithms for the construction of dominating and connected dominating sets of a Unit Disk Graph (UDG). The size of the constructed set, in the case of the dominating set, is shown to be 5 times the optimal, while for the connected dominating set 7.453 + Ɛ the optimal, for any arbitrarily small Ɛ > 0. These are to our knowledge the first local algorithms whose time complexities and approximation bounds are independent of the size of the network.

An automatic variational level set segmentation framework for computer aided dental X-rays analysis in clinical environments

An automatic variational level set segmentation framework for Computer Aided Dental X-rays Analysis (CADXA) in clinical environments is proposed. Designed for clinical environments, the segmentation contains two stages: a training stage and a segmentation stage. During the training stage, first, manually chosen representative images are segmented using hierarchical level set region detection. Then the window based feature extraction followed by principal component analysis (PCA) is applied and results are used to train a support vector machine (SVM) classifier. During the segmentation stage, dental X-rays are classified first by the trained SVM. The classifier provides initial contours which are close to correct boundaries for three coupled level sets driven by a proposed pathologically variational modeling which greatly accelerates the level set segmentation. Based on the segmentation results and uncertainty maps that are built based on a proposed uncertainty measurement, a computer aided analysis scheme is applied. The experimental results show that the proposed method is able to provide an automatic pathological segmentation which naturally segments those problem areas. Based on the segmentation results, the analysis scheme is able to provide indications of possible problem areas of bone loss and decay to the dentists. As well, the experimental results show that the proposed segmentation framework is able to speed up the level set segmentation in clinical environments.

High delivery rate position-based routing algorithms for 3D ad hoc networks

Position-based routing algorithms use the geographic position of the nodes in a network to make the forwarding decisions. Recent research in this field primarily addresses such routing algorithms in two dimensional (2D) space. However, in real applications, nodes may be distributed in three dimensional (3D) environments. In this paper, we propose several randomized position-based routing algorithms and their combination with restricted directional flooding-based algorithms for routing in 3D environments. The first group of algorithms AB3D are extensions of previous randomized routing algorithms from 2D space to 3D space. The second group ABLAR chooses m neighbors according to a space-partition heuristic and forwards the message to all these nodes. The third group T-ABLAR-T uses progress-based routing until a local minimum is reached. The algorithm then switches to ABLAR for one step after which the algorithm switches back to the progress-based algorithm again. The fourth group AB3D-ABLAR uses an algorithm from the AB3D group until a threshold is passed in terms of number of hops. The algorithm then switches to an ABLAR algorithm. The algorithms are evaluated and compared with current routing algorithms. The simulation results on unit disk graphs (UDG) show a significant improvement in delivery rate (up to 99%) and a large reduction of the traffic.

Power-aware semi-beaconless 3D georouting algorithms using adjustable transmission ranges for wireless ad hoc and sensor networks

Due to the limited lifetime of the nodes in ad hoc network, energy efficiency needs to be an important design consideration in any routing algorithm for ad hoc and sensor networks. In most of the existing position-based routing algorithms the nodes use the maximum transmission power to discover neighbors, which may cause excessive power consumption. This paper presents several localized power-aware 3D position-based routing algorithms that increase the lifetime of a network by maximizing the average lifetime of its nodes. New algorithms are semi-beaconless, using for neighbor discovery an optimal transmission range (OR) for control packets, and, if needed, maximal transmission range (MR) during routing process, and using adjusted transmission radius for message transmission. PAGR algorithm selects neighbor closest to destination among those within OR if any exists providing progress, or otherwise among those within MR. If greedy progress is not possible, PAGR:CFace(1) variant resorts to face routing on projected network in coordinate plane until recovery is possible, at which point PAGR algorithm resumes. We evaluate our algorithms and compare their power savings with the current power-aware routing algorithms. The simulation results show a significant improvement in the overall network lifetime.

Power-Aware 3D Position-based Routing Algorithms for Ad Hoc Networks

A crucial problem in ad hoc networks is finding an efficient and correct route between a source and a destination; however for many networks, a more important problem is providing an energy efficient route because of, for example, the limited battery life of the wireless nodes. Most previous routing protocols make the routing decision without taking into account the energy budget of the nodes. In addition, when using a fixed transmission power, nodes may waste power by transmitting with more power than is needed for correct reception. In position- based routing algorithms, the nodes use the geographical position of the nodes to make the routing decisions. In this paper we present several localized power-aware 3D position-based routing algorithms that increase the life-time of the network by maximizing the life time of the nodes. These new algorithms use the idea of replacing the constant transmission power of the node with an adjusted transmission power during two stages - first a lower power while discovering the neighboring nodes, and, if needed, a second higher transmission power during the routing process. We evaluate our algorithms and compare their power savings with the current power-aware routing algorithms. The simulation results show a significant improvement in the energy saving (up to 50%).