The iris, the concentric ring-shaped colored portion of the eye that encloses the dark disc of th... more The iris, the concentric ring-shaped colored portion of the eye that encloses the dark disc of the pupil, possesses rich and distinct features unique for each individual. Although most current identification and authentication systems use more traditional biometric features, e.g. fingerprints and handwritten signatures, the iris-based recognition has proven to be far more accurate and therefore provides a good alternative resource. The main algorithmic components that are usually involved in any personal identification system that utilizes the human iris are: image data acquisition, iris localization and segmentation, feature extraction and decision making. These scientific methods that enable this advanced biometric technology is a good example of how cross-disciplinary scientific methods-such as artificial intelligence or AI, image processing, as well as practical software design tools such as graphical-user-interface-design, can come together to provide an integrated solution to a challenging security problem. In our computer science department, we have a need to demonstrate this computer-based technology to our new graduate and undergraduate students as a means to enhance their learning experience and motivate them to utilize their computer knowledge and skills in solving real-life problems. Aside from the scientific algorithm involved, the project addresses several practical considerations involved in the design and development of irisbased recognition systems such as system functionalities as well as the flexibility and the acceptability for the user interface. Two computer science students were assigned the task of designing, building and implementing a prototype using software development tools and scientific computing methods. In this paper, we provide general guidelines while highlighting some of the important issues involved in the design of biometric systems exploring the specific case of iris-based recognition.
Biomedical informatics has been defined as the discipline concerned with the systematic processin... more Biomedical informatics has been defined as the discipline concerned with the systematic processing of data, information and knowledge in medicine and health care. It is an interdisciplinary field based on computer science, information science, the cognitive and decision sciences, telecommunications as well as other fields. Because of the rich and diverse nature of medical information, it has created a fertile ground for innovations and applied research particularly from the prospective of computer science and information technology. Although medical informatics has been recognized as a standalone science, few colleges and universities with computer science programs have acknowledged medical informatics as a viable application and have recognized the importance of incorporating medical informatics courses into their curriculum. Also, there has been no unified approach as to how topics in medical informatics should be integrated into the curriculum. In this paper, we address the need to have a structured paradigm for embedding medical informatics courses in computer science programs in a way that allows students to achieve a multitude of knowledge and a high-level of proficiency which will ultimately enable them to apply their problems-solving skills in health and medicine. We define three main areas from which these courses are derived. First, methodologies for processing data, information and knowledge in medicine and health care are essential for analysis and management of clinical data in research and medical practice especially in administrative and clinical decision supports. Second, understanding the clinical workflow of a typical hospital information system is also crucial for the design of electronic medical record applications and hospital network topology. Finally, biometric applications invite computer scientists to fully employ their knowledge in artificial intelligence and security systems. We believe that our work may contribute as future guidelines for incorporating medical informatics as an optional track in computer science programs.
The objective of this study was to devise an interactive tool to assist in cryoablation therapy t... more The objective of this study was to devise an interactive tool to assist in cryoablation therapy through computer modeling, simulation, and visualization. CryoSim, a software package, accepts a set of acquired and processed three-dimensional ultrasound images, then models heat diffusion (formation of the iceball) based on numerical approximation of the heat equation and knowledge of the thermal properties of the
The objective of this study was to devise an interactive tool to assist in cryoablation therapy t... more The objective of this study was to devise an interactive tool to assist in cryoablation therapy through computer modeling, simulation, and visualization. CryoSim, a software package, accepts a set of acquired and processed three-dimensional ultrasound images, then models heat diffusion (formation of the iceball) based on numerical approximation of the heat equation and knowledge of the thermal properties of the underlying tissues. Results of cryoexperiments were found to be significantly similar to those generated by CryoSim. Therefore, CryoSim provides a viable technique for predicting the outcome of cryosurgery, and establishes a platform for future automation of cryosurgery.
18th International Conference on Pattern Recognition (ICPR'06), 2006
An effective and accurate identification of human individuals from their iris features is largely... more An effective and accurate identification of human individuals from their iris features is largely dependent on proper segmentation of the iris and the pupil features from camera images. Most modern segmentation schemes exploit the circular geometry of the iris to fit a circle or an ellipse to an edge map of the iris. In this paper, we present a new
Lasers in Surgery: Advanced Characterization, Therapeutics, and Systems X, 2000
ABSTRACT Photodynamic therapy (PDT) is an emerging minimally invasive treatment that can be emplo... more ABSTRACT Photodynamic therapy (PDT) is an emerging minimally invasive treatment that can be employed in many human diseases including prostate cancer. This treatment of human prostate cancer depends on the localization of a drug (photosensitizer) into the prostate. The photosensitizer is activated by high- energy laser light and the active drug destroys cancerous tissue. The success of PDT depends on precise placement of light diffusers in the prostate. Since the prostate is irregular in shape, with different dimensions, a transurethral light delivery that is circular in distribution cannot be used in most cases of carcinoma of the prostate. Sources of light and their spatial distribution must be tailored to each individual patient. More uniform, therapeutic light distribution can be achieved by interstitial light irradiation. In this case, the light is delivered by diffusers placed within the substance of the prostate parallel to the urethra at a distance optimized to deliver adequate levels of light and to create the desired photodynamic effect. For this reason, we are developing a computer program that can calculate the distribution of energy depending on the number of light sources placed in the prostate, their position in the gland, the dimension of the prostate, and the attenuation coefficient. A patient's three-dimensional prostate model is built based on ultrasound images. Then the program is being designated to predict the best set of parameters and position of light diffusers in space, displays them in graphical form or in numerical form. The program is amenable for interfacing with robotic treatment systems.
Traditional segmentation methods cannot provide satisfying results for extraction of prostate gla... more Traditional segmentation methods cannot provide satisfying results for extraction of prostate gland from Transrectal Ultrasound (TRUS) images because of the presence of strong speckle noise and shadow artifacts. Most ultrasound image segmentation techniques that adopt model-based approach such as active contour are considered semi-automatic because they require initial seeds or contours to be manually identified. In this paper, we propose a method for automatic segmentation of prostate using feature-based self organizing map (SOM). Median filtering and top hat transform are first applied to remove speckle noise. A technique is developed to remove ultrasound-specific speckles using texturebased thresholding. An SOM algorithm is employed to identify prostate pixels taking spatial information, gray-level as well as texture information to form its input vector. The clustered image is then processed to produce a fully connected prostate contour. A number of experiments comparing extracted contours with manually-delineated contours validated the performance of our method.
2008 3rd International Symposium on Communications, Control and Signal Processing, 2008
Automatic segmentation of prostate boundaries from Transrectal Ultrasound (TRUS) images is an imp... more Automatic segmentation of prostate boundaries from Transrectal Ultrasound (TRUS) images is an important task that continues to challenge minimally-invasive surgical procedures. Several model-based and feature-based approaches have been proposed for segmentation of the prostate. In this paper, a new edge-based segmentation scheme that is based on feature phase symmetry and path cost minimization is introduced. First, the images are preprocessed with noise reduction filters. Dominant prostate edges are then detected at multiple orientations using the principle of phase symmetry, a frequency-based method that is invariant to global contrast variation. The reconstructed feature or edge map combined with an orientation map is used in a path cost minimization scheme. This scheme seeks to find the best closed contour that best match a set of edgebased rules. A number of experiments comparing the extracted contours with manually-delineated contours validated the performance of our method.
2007 International Joint Conference on Neural Networks, 2007
Abstract Prostate segmentation has been a challenging task that has long hindered the progress of... more Abstract Prostate segmentation has been a challenging task that has long hindered the progress of ultrasound-guided procedures of the prostate. The focus of this paper is on the problem of feature extraction, classification and labeling of prostate tissues and non-...
ABSTRACT Two-dimensional (2D) ultrasound imaging is commonly used for diagnosis in a variety of m... more ABSTRACT Two-dimensional (2D) ultrasound imaging is commonly used for diagnosis in a variety of medical fields. However, there are several drawbacks of conventional 2D-ultrasound imaging. These include prostate or transducer movement that produces sets of different images that are difficult to interpret. Also during patient's reexamination correspondence between sets of images before reexamination and after is difficult to establish. This can be described as a problem of correlation between two sets of images: the first created before distortion or examination, the second one after. We propose a method to register 2D ultrasound volumes based on external markers introduced in the prostate. The metal balls are inserted in the prostate at three distinct locations in the prostate. These appear as bright dots in the ultrasound field, serve as reference points, are then outlined through a user-interactive program from two sets of images. Then, the computer program rotates and translates till they match respectively, and displays the mapped points with their corresponding location. Based on this idea we developed an image-guided system for PDT that require high-precision placement of implants. In the planning stage, the system performs an automatic acquisition of 2D transrectal ultrasound images that will ultimately be used to construct the treatment plan. At the time of the therapy, new sets of ultrasound images are acquired and a match is established between the virtual world and the patient's real world with the aid of manually introduced markers and image matching algorithms.
2006 IEEE International Conference on Electro/Information Technology, 2006
In this paper we present a new method for iris texture recognition for the purpose of human ident... more In this paper we present a new method for iris texture recognition for the purpose of human identification using statistical analysis of gray-level distribution. Many studies have been aimed at extracting iris features that are unique to every individual. While many have been successful, most requires complex filtering and processing. Our proposed method is based on a simple estimate of
Automatic segmentation of prostate boundaries from transrectal ultrasound (TRUS) images still pos... more Automatic segmentation of prostate boundaries from transrectal ultrasound (TRUS) images still poses significant challenge in minimally-invasive surgical procedures. The presence of strong speckle noise and shadow artifacts limits the effectiveness of classical segmentation schemes. Several model-based and feature-approaches have been proposed for segmentation of the prostate. In this paper, we propose a new energy-based method for segmentation of ultrasound prostate images using active contour modeling guided by dot-pattern textural energy map. First, impulsive noise and speckles are reduced with median filtering and top-hat transform. Prostate features are then extracted from the filtered images using non-linear dot-pattern select operator. An elastic template shape model that incorporates a priori knowledge of the average geometric shape of the prostate boundaries as well as the energy derived from the dot-pattern feature image are utilized to search for the optimal prostate contour. A number of experiments comparing the extracted contours with manually-delineated contours validated the performance of our method.
The iris, the concentric ring-shaped colored portion of the eye that encloses the dark disc of th... more The iris, the concentric ring-shaped colored portion of the eye that encloses the dark disc of the pupil, possesses rich and distinct features unique for each individual. Although most current identification and authentication systems use more traditional biometric features, e.g. fingerprints and handwritten signatures, the iris-based recognition has proven to be far more accurate and therefore provides a good alternative resource. The main algorithmic components that are usually involved in any personal identification system that utilizes the human iris are: image data acquisition, iris localization and segmentation, feature extraction and decision making. These scientific methods that enable this advanced biometric technology is a good example of how cross-disciplinary scientific methods-such as artificial intelligence or AI, image processing, as well as practical software design tools such as graphical-user-interface-design, can come together to provide an integrated solution to a challenging security problem. In our computer science department, we have a need to demonstrate this computer-based technology to our new graduate and undergraduate students as a means to enhance their learning experience and motivate them to utilize their computer knowledge and skills in solving real-life problems. Aside from the scientific algorithm involved, the project addresses several practical considerations involved in the design and development of irisbased recognition systems such as system functionalities as well as the flexibility and the acceptability for the user interface. Two computer science students were assigned the task of designing, building and implementing a prototype using software development tools and scientific computing methods. In this paper, we provide general guidelines while highlighting some of the important issues involved in the design of biometric systems exploring the specific case of iris-based recognition.
Biomedical informatics has been defined as the discipline concerned with the systematic processin... more Biomedical informatics has been defined as the discipline concerned with the systematic processing of data, information and knowledge in medicine and health care. It is an interdisciplinary field based on computer science, information science, the cognitive and decision sciences, telecommunications as well as other fields. Because of the rich and diverse nature of medical information, it has created a fertile ground for innovations and applied research particularly from the prospective of computer science and information technology. Although medical informatics has been recognized as a standalone science, few colleges and universities with computer science programs have acknowledged medical informatics as a viable application and have recognized the importance of incorporating medical informatics courses into their curriculum. Also, there has been no unified approach as to how topics in medical informatics should be integrated into the curriculum. In this paper, we address the need to have a structured paradigm for embedding medical informatics courses in computer science programs in a way that allows students to achieve a multitude of knowledge and a high-level of proficiency which will ultimately enable them to apply their problems-solving skills in health and medicine. We define three main areas from which these courses are derived. First, methodologies for processing data, information and knowledge in medicine and health care are essential for analysis and management of clinical data in research and medical practice especially in administrative and clinical decision supports. Second, understanding the clinical workflow of a typical hospital information system is also crucial for the design of electronic medical record applications and hospital network topology. Finally, biometric applications invite computer scientists to fully employ their knowledge in artificial intelligence and security systems. We believe that our work may contribute as future guidelines for incorporating medical informatics as an optional track in computer science programs.
The objective of this study was to devise an interactive tool to assist in cryoablation therapy t... more The objective of this study was to devise an interactive tool to assist in cryoablation therapy through computer modeling, simulation, and visualization. CryoSim, a software package, accepts a set of acquired and processed three-dimensional ultrasound images, then models heat diffusion (formation of the iceball) based on numerical approximation of the heat equation and knowledge of the thermal properties of the
The objective of this study was to devise an interactive tool to assist in cryoablation therapy t... more The objective of this study was to devise an interactive tool to assist in cryoablation therapy through computer modeling, simulation, and visualization. CryoSim, a software package, accepts a set of acquired and processed three-dimensional ultrasound images, then models heat diffusion (formation of the iceball) based on numerical approximation of the heat equation and knowledge of the thermal properties of the underlying tissues. Results of cryoexperiments were found to be significantly similar to those generated by CryoSim. Therefore, CryoSim provides a viable technique for predicting the outcome of cryosurgery, and establishes a platform for future automation of cryosurgery.
18th International Conference on Pattern Recognition (ICPR'06), 2006
An effective and accurate identification of human individuals from their iris features is largely... more An effective and accurate identification of human individuals from their iris features is largely dependent on proper segmentation of the iris and the pupil features from camera images. Most modern segmentation schemes exploit the circular geometry of the iris to fit a circle or an ellipse to an edge map of the iris. In this paper, we present a new
Lasers in Surgery: Advanced Characterization, Therapeutics, and Systems X, 2000
ABSTRACT Photodynamic therapy (PDT) is an emerging minimally invasive treatment that can be emplo... more ABSTRACT Photodynamic therapy (PDT) is an emerging minimally invasive treatment that can be employed in many human diseases including prostate cancer. This treatment of human prostate cancer depends on the localization of a drug (photosensitizer) into the prostate. The photosensitizer is activated by high- energy laser light and the active drug destroys cancerous tissue. The success of PDT depends on precise placement of light diffusers in the prostate. Since the prostate is irregular in shape, with different dimensions, a transurethral light delivery that is circular in distribution cannot be used in most cases of carcinoma of the prostate. Sources of light and their spatial distribution must be tailored to each individual patient. More uniform, therapeutic light distribution can be achieved by interstitial light irradiation. In this case, the light is delivered by diffusers placed within the substance of the prostate parallel to the urethra at a distance optimized to deliver adequate levels of light and to create the desired photodynamic effect. For this reason, we are developing a computer program that can calculate the distribution of energy depending on the number of light sources placed in the prostate, their position in the gland, the dimension of the prostate, and the attenuation coefficient. A patient's three-dimensional prostate model is built based on ultrasound images. Then the program is being designated to predict the best set of parameters and position of light diffusers in space, displays them in graphical form or in numerical form. The program is amenable for interfacing with robotic treatment systems.
Traditional segmentation methods cannot provide satisfying results for extraction of prostate gla... more Traditional segmentation methods cannot provide satisfying results for extraction of prostate gland from Transrectal Ultrasound (TRUS) images because of the presence of strong speckle noise and shadow artifacts. Most ultrasound image segmentation techniques that adopt model-based approach such as active contour are considered semi-automatic because they require initial seeds or contours to be manually identified. In this paper, we propose a method for automatic segmentation of prostate using feature-based self organizing map (SOM). Median filtering and top hat transform are first applied to remove speckle noise. A technique is developed to remove ultrasound-specific speckles using texturebased thresholding. An SOM algorithm is employed to identify prostate pixels taking spatial information, gray-level as well as texture information to form its input vector. The clustered image is then processed to produce a fully connected prostate contour. A number of experiments comparing extracted contours with manually-delineated contours validated the performance of our method.
2008 3rd International Symposium on Communications, Control and Signal Processing, 2008
Automatic segmentation of prostate boundaries from Transrectal Ultrasound (TRUS) images is an imp... more Automatic segmentation of prostate boundaries from Transrectal Ultrasound (TRUS) images is an important task that continues to challenge minimally-invasive surgical procedures. Several model-based and feature-based approaches have been proposed for segmentation of the prostate. In this paper, a new edge-based segmentation scheme that is based on feature phase symmetry and path cost minimization is introduced. First, the images are preprocessed with noise reduction filters. Dominant prostate edges are then detected at multiple orientations using the principle of phase symmetry, a frequency-based method that is invariant to global contrast variation. The reconstructed feature or edge map combined with an orientation map is used in a path cost minimization scheme. This scheme seeks to find the best closed contour that best match a set of edgebased rules. A number of experiments comparing the extracted contours with manually-delineated contours validated the performance of our method.
2007 International Joint Conference on Neural Networks, 2007
Abstract Prostate segmentation has been a challenging task that has long hindered the progress of... more Abstract Prostate segmentation has been a challenging task that has long hindered the progress of ultrasound-guided procedures of the prostate. The focus of this paper is on the problem of feature extraction, classification and labeling of prostate tissues and non-...
ABSTRACT Two-dimensional (2D) ultrasound imaging is commonly used for diagnosis in a variety of m... more ABSTRACT Two-dimensional (2D) ultrasound imaging is commonly used for diagnosis in a variety of medical fields. However, there are several drawbacks of conventional 2D-ultrasound imaging. These include prostate or transducer movement that produces sets of different images that are difficult to interpret. Also during patient's reexamination correspondence between sets of images before reexamination and after is difficult to establish. This can be described as a problem of correlation between two sets of images: the first created before distortion or examination, the second one after. We propose a method to register 2D ultrasound volumes based on external markers introduced in the prostate. The metal balls are inserted in the prostate at three distinct locations in the prostate. These appear as bright dots in the ultrasound field, serve as reference points, are then outlined through a user-interactive program from two sets of images. Then, the computer program rotates and translates till they match respectively, and displays the mapped points with their corresponding location. Based on this idea we developed an image-guided system for PDT that require high-precision placement of implants. In the planning stage, the system performs an automatic acquisition of 2D transrectal ultrasound images that will ultimately be used to construct the treatment plan. At the time of the therapy, new sets of ultrasound images are acquired and a match is established between the virtual world and the patient's real world with the aid of manually introduced markers and image matching algorithms.
2006 IEEE International Conference on Electro/Information Technology, 2006
In this paper we present a new method for iris texture recognition for the purpose of human ident... more In this paper we present a new method for iris texture recognition for the purpose of human identification using statistical analysis of gray-level distribution. Many studies have been aimed at extracting iris features that are unique to every individual. While many have been successful, most requires complex filtering and processing. Our proposed method is based on a simple estimate of
Automatic segmentation of prostate boundaries from transrectal ultrasound (TRUS) images still pos... more Automatic segmentation of prostate boundaries from transrectal ultrasound (TRUS) images still poses significant challenge in minimally-invasive surgical procedures. The presence of strong speckle noise and shadow artifacts limits the effectiveness of classical segmentation schemes. Several model-based and feature-approaches have been proposed for segmentation of the prostate. In this paper, we propose a new energy-based method for segmentation of ultrasound prostate images using active contour modeling guided by dot-pattern textural energy map. First, impulsive noise and speckles are reduced with median filtering and top-hat transform. Prostate features are then extracted from the filtered images using non-linear dot-pattern select operator. An elastic template shape model that incorporates a priori knowledge of the average geometric shape of the prostate boundaries as well as the energy derived from the dot-pattern feature image are utilized to search for the optimal prostate contour. A number of experiments comparing the extracted contours with manually-delineated contours validated the performance of our method.
Uploads
Papers by Amjad Zaim