Biomedical subjects

David J Foran

Publications and source records attributed to David J Foran.

8 recordsLinked to original sources

PubMed2005

Image mining for investigative pathology using optimized feature extraction and data fusion.

In many subspecialties of pathology, the intrinsic complexity of rendering accurate diagnostic decisions is compounded by a lack of definitive criteria for detecting and characterizing diseases and their corresponding histological features. In some cases, there exists a striking disparity between the diagnoses rendered by recognized authorities and those provided by non-experts. We previously reported the development of an Image Guided Decision Support (IGDS) system, which was shown to reliably discriminate among malignant lymphomas and leukemia that are sometimes confused with one another during routine microscopic evaluation. As an extension of those efforts, we report here a web-based intelligent archiving subsystem that can automatically detect, image, and index new cells into distributed ground-truth databases. Systematic experiments showed that through the use of robust texture descriptors and density estimation based fusion the reliability and performance of the governing classifications of the system were improved significantly while simultaneously reducing the dimensionality of the feature space.

Diagnosis, Differential↗

PubMed2005

Unsupervised segmentation based on robust estimation and color active contour models.

One of the most commonly used clinical tests performed today is the routine evaluation of peripheral blood smears. In this paper, we investigate the design, development, and implementation of a robust color gradient vector flow (GVF) active contour model for performing segmentation, using a database of 1791 imaged cells. The algorithms developed for this research operate in Luv color space, and introduce a color gradient and L2E robust estimation into the traditional GVF snake. The accuracy of the new model was compared with the segmentation results using a mean-shift approach, the traditional color GVF snake, and several other commonly used segmentation strategies. The unsupervised robust color snake with L2E robust estimation was shown to provide results which were superior to the other unsupervised approaches, and was comparable with supervised segmentation, as judged by a panel of human experts.

Algorithms↗

PubMed2004

A prototype for unsupervised analysis of tissue microarrays for cancer research and diagnostics.

The tissue microarray (TMA) technique enables researchers to extract small cylinders of tissue from histological sections and arrange them in a matrix configuration on a recipient paraffin block such that hundreds can be analyzed simultaneously. TMA offers several advantages over traditional specimen preparation by maximizing limited tissue resources and providing a highly efficient means for visualizing molecular targets. By enabling researchers to reliably determine the protein expression profile for specific types of cancer, it may be possible to elucidate the mechanism by which healthy tissues are transformed into malignancies. Currently, the primary methods used to evaluate arrays involve the interactive review of TMA samples while they are viewed under a microscope, subjectively evaluated, and scored by a technician. This process is extremely slow, tedious, and prone to error. In order to facilitate large-scale, multi-institutional studies, a more automated and reliable means for analyzing TMAs is needed. We report here a web-based prototype which features automated imaging, registration, and distributed archiving of TMAs in multiuser network environments. The system utilizes a principal color decomposition approach to identify and characterize the predominant staining signatures of specimens in color space. This strategy was shown to be reliable for detecting and quantifying the immunohistochemical expression levels for TMAs.

Adolescent↗

PubMed2003

Dynamic quiz bank: a portable tool set for authoring and managing distributed, Web-based educational programs in radiology.

RATIONALE AND OBJECTIVES: The authors performed this study to evaluate a portable, platform-independent software program that enables users from remote sites to transform raw materials (eg, text, images, video) into Web-ready, interactive tutorials and examinations. MATERIALS AND METHODS: The software program evaluated consists of three modules: a network-based interface developed in the Java programming language, an Oracle 8i relational database with liaison software, and shell scripts developed in the Perl programming language to automate the authoring, maintenance, and updating of content in a dynamic quiz bank (DBQ). Four faculty members, one radiology resident, and two undergraduates majoring in computer science volunteered to create questions for the DQB and to evaluate ease of authoring. RESULTS: Faculty members with various levels of computer proficiency were able to establish a comprehensive DQB of more than 1,000 interactive questions. These radiologists reported the scripts reliable and easy to use. The DQB, offered in a pathology course for 2nd-year medical students, was used by 151 students and may have played a role in improving standardized test scores. Eighty-seven percent (n = 131) of the students believed that the DQB was extremely useful as an educational tool. CONCLUSION: The DQB software program facilitated access to, and authoring and maintenance of, Web-based educational materials developed in the departments of radiology and pathology.

Computer-Assisted Instruction↗

PubMed2003

A network-reactive model for distributed telemicroscopy.

We have developed a network-reactive robotic telemicroscopy system that uses high-speed image processing and quality of service (QoS) feedback to maximize performance. In tests, four magnifications (4x, 10x, 40x, 60x) of a section stained with haematoxylin and eosin were used. Up to 25 clients, distributed across two Internet domains, accessed the server. When given control of the microscope, any of the clients could direct the microscope to change focus and objectives within 3 s and move the stage to any x-y position within 1 s. Transmission rates of 50-200 kbit/s were achieved for video (320 x 240 pixels at 24 bit/pixel). Performance varied according to the complexity of the images being transmitted. The 25 simultaneous users were supported without affecting the operation of the server and the system overhead was relatively small. The server's message-processing time and the network transmission delay amounted to 83 ms. The compression ratio of five previously stored video-sequences ranged from 336:1 to 1213:1. The underlying model for this system could serve a wide range of telemedicine and distance-learning applications.

Humans↗

PubMed2002

Mechanical implications of the domain structure of fiber-forming collagens: comparison of the molecular and fibrillar flexibilities of the alpha1-chains found in types I-III collagen.

Fibrillar collagens store, transmit and dissipate elastic energy during tensile deformation. Results of previous studies suggest that the collagen molecule is made up of alternating rigid and flexible domains, and extension of the flexible domains is associated with elastic energy storage. In this study, we model the flexibility of the alpha1-chains found in types I-III collagen molecules and microfibrils in order to understand the molecular basis of elastic energy storage in collagen fibers by analysing the areas under conformational plots for dipeptide sequences. Results of stereochemical modeling suggest that the collagen triple helix is made up of rigid and flexible domains that alternate with periods that are multiples of three amino acid residues. The relative flexibility of dipeptide sequences found in the flexible regions is about a factor of five higher than that found for the flexibility of the rigid regions, and the flexibility of types II and III collagen molecules appears to be higher than that found for the type I collagen molecule. The different collagen alpha1-chains were compared by correlating the flexibilities. The results suggest that the flexibilities of the alpha1-chains of types I and III collagen are more closely related than the flexibilities of the alpha1-chains in types I and II and II and III collagen. The flexible domains found in the alpha1-chains of types I-III collagen were found to be conserved in the microfibril and had periods of about 15 amino acid residues and multiples thereof. The flexibility profiles of types I and II collagen microfibrils were found to be more highly correlated than those for types I and III and II and III. These results suggest that the domain structure of the alpha1-chains found in types I-III collagen is an efficient means for storage of elastic energy during stretching while preserving the triple helical structure of the overall molecule. It is proposed that all collagens that form fibers are designed to act as storage elements for elastic energy. The function of fibers rich in type I collagen is to store and then transmit this energy while fibers rich in types II and III collagen may store and then reflect elastic energy for dissipation through viscous fibrillar slippage. Impaired elastic energy storage by extracellular matrices may lead to cellular damage and changes in signaling by mechanochemical transduction at the extracellular matrix-cell interface.

Animals↗

PubMed2002

Unsupervised imaging, registration and archiving of tissue microarrays.

Tissue microarray (TMA) technology offers several advantages over traditional methods of specimen preparation by maximizing limited tissue resources and providing the means for visualizing molecular targets. Currently, the primary methods used to evaluate arrays involves the interactive review of TMA samples while they are viewed by microscope and are subjectively evaluated and scored. The process is slow, tedious and prone to error. In order to facilitate large-scale, multi-institutional studies a more automated and reliable means for processing tissue microarrays is needed. We have developed a web-based prototype which features automated imaging, registration and intelligent archiving of tissue microarrays in multi-user, network environments.

Databases, Factual↗

PubMedDate not supplied

Mechanical behavior of vessel wall: a comparative study of aorta, vena cava, and carotid artery.

We have used incremental stress-strain curves to study the mechanical behavior of porcine aorta, carotid artery, and vena cava. Elastic and viscous stress-strain curves are composed of low and high strain regions that are approximately linear. Analysis of the low strain behavior is consistent with previous studies that suggest that the behavior is dominated by the behavior of elastic fibers, and that the collagen and elastic fibers are in parallel networks. At high strain, the behavior is different than that of skin where it is dominated by the behavior of the collagen fibers. The high strain behavior is consistent with a series arrangement of the collagen and smooth muscle; however, the arrangement of smooth muscle and collagen may be different in aorta than in the other vessels studied. It is concluded that the mechanical behavior of the vessel wall differs from the behavior of other extracellular matrices that do not contain smooth muscle. Our results indicate that at least some of the collagen fibrils in the media are in series with smooth muscle cells and this collagen-smooth muscle network is in parallel with parallel networks of collagen and elastic tissue in aorta, carotid artery, and vena cava. It is concluded that the series arrangement of collagen and smooth muscle may be important in mechanochemical transduction in vessel walls and that the exact quantity and arrangement of these components may differ in different vessels.

Animals↗