A. Afework, M. D. Beynon, F. Bustamante, A. Demarzo, R. Ferreira, R. Miller, M. Silberman, J. Saltz, A. Sussman, and H. Tsang. Digital dynamic telepathology - the Virtual Microscope. In AMIA98. American Medical Informatics Association, November 1998. Also available as University of Maryland Technical Report CS-TR-3892 and UMIACS-TR98 -23.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....transmission within the compression algorithm. Said and Pearlman proposed in [4] an image multiresolution transform that is suited for both lossless and lossy compression. Their lossless compression ratios are among the best in the literature. Afework et al. developed a Virtual Microscope system [1, 3], an integrated computer hardware and software system, to stimulate the high power light microscope. The server program runs on parallel computers with very high processing capability, which makes their system unrealistic for routine image manipulating. Wang et al. developed a virtual microscope ....

A. Afework, M. Beynon, F. Bustamante, A. Demarzo, R. Ferreira, R. Miller, M. Silberman, J. Saltz, A. Sussman, and H. Tsang, "Digital Dynamic Telepathology - The Virtual Microscope," Proceedings of the 1998.

....successful operation of cluster applications. This is because even a single high performance application running on a cluster typically exhibits highly dynamic computational behavior. Moreover, most applications do not run in isolation: they conduct I O, require real time data from remote sensors[3], access largescale remote data contained in digital libraries or share les across the computational grid, support scienti c collaboration by remote visualization of their data[18, 19] and interact with other computations via the Grid [1, 2] Unless run time monitoring is used to determine ....

Asmara Afework, Michael Benyon, Fabian E. Bustamante, Angelo DeMarzo, Renato Ferreira, Rovert Miller, Mark Silberman, Joel Saltz, Alan Sussman. "Digital Dynamic Telepathology - the Virtual Microscope", In Proc. of the 1998 AMIA Annual Fall Symposium, August, 1998.

....successful operation of cluster applications. This is because even a single high performance application running on a cluster typically exhibits highly dynamic computational behavior. Moreover, most applications do not run in isolation: they conduct I O, require real time data from remote sensors[3], access largescale remote data contained in digital libraries or share les across the computational grid, support scienti c collaboration by remote visualization of their data[21, 22] and interact with other computations via the Grid [1, 2] Unless run time monitoring is used to determine ....

Asmara Afework, Michael Benyon, Fabian E. Bustamante, Angelo DeMarzo, Renato Ferreira, Rovert Miller, Mark Silberman, Joel Saltz, Alan Sussman. "Digital Dynamic Telepathology - the Virtual Microscope", In Proc. of the

....support for common operations such as memory management, data retrieval, and scheduling of processing across a distributed memory parallel machine. Examples of data intensive applications implemented with ADR include Titan [11, 12, 35] for 5 satellite data processing, the Virtual Microscope [1, 18] for visualization and analysis of microscopy data, and coupling of multiple simulations for water contamination studies [27] Customization in ADR is achieved through C class inheritance. That is, for each of the customizable services, ADR provides a set of C base classes with virtual ....

Asmara Afework, Michael D. Beynon, Fabian Bustamante, Angelo Demarzo, Renato Ferreira, Robert Miller, Mark Silberman, Joel Saltz, Alan Sussman, and Hubert Tsang. Digital dynamic telepathology - the Virtual Microscope. In Proceedings of the

.... by (1) maintaining intermediate data structures generated by queries, 2) caching input data in memory, and (3) providing support for multi threaded execution (i.e. each query is executed as a thread) We describe initial experimental results using an application, called the Virtual Microscope [2], for browsing digitized microscopy images. 2 Query Client B Client A Query Result Query Server (Query Planning Query Execution) Manager Data Store Disk Farm . Tape Storage Page Space Manager Thread Thread Index Manager Data Source Data Source Data Source Query Result ....

....indices defined on the datasets. The query thread interacts with the index manager to access index data structures and search for data that intersect with the query. 5 Figure 2: The Virtual Microscope client. 3 Analysis of Microscopy Data: The Virtual Microscope The Virtual Microscope (VM) [2] provides a realistic digital emulation of a high power light microscope. Figure 2 shows the client interface for VM. The raw data for such a system can be captured by digitally scanning collections of full microscope slides under high power. A slide can contain multiple focal planes. The size of ....

A. Afework, M. D. Beynon, F. Bustamante, A. Demarzo, R. Ferreira, R. Miller, M. Silberman, J. Saltz, A. Sussman, and H. Tsang. Digital dynamic telepathology - the Virtual Microscope. In Proceedings of the 1998 AMIA Annual Fall Symposium. American Medical Informatics Association, Nov. 1998.

....of the original image. This approach is problematic for several reasons: it does not allow students to focus on an arbitrary region of the image, it involves a significant amount of manual effort to build image libraries and it wastes a lot of disk space. Researchers from University of Maryland [2] have recently developed a virtual microscope system. It is an integrated computer hardware and software system. Using massively parallel computers, the system is shown to be able to serve many users at the same time. However, the algorithm used in the system is not wavelet based. Therefore, the ....

A Afework, MD Beynon, F Bustamante, S Cho, A Demarzo, R Ferreira, R Miller, M Silberman, J Saltz, A Sussman, H Tsang, Digital dynamic telepathology--the virtual microscope, Proc AMIA Symp, pp. 912-6, 1998.

.... that employ large scale scientific datasets, including applications that explore, compare, and visualize results generated by large scale simulations [8] visualize and generate data products from global coverage satellite data [4] and visualize and analyze digitized microscopy images [1]. Such applications often use only a subset of all the data available in both the input and output datasets. References to data items are described by a range query, namely a multi dimensional bounding box in the underlying multi dimensional attribute space of the dataset(s) Only the data items ....

....counts of the strategies when input data elements are uniformly distributed in the attribute space of the output dataset, restricting the output dataset to be a regular d dimensional array. We validate these cost models with queries for synthetic datasets and for several driving applications [1, 4, 8]. 2 2 Overview of ADR In this section we briefly describe three strategies for processing range queries in ADR. First we briefly describe how datasets are stored in ADR, and outline the main phases of query execution in ADR. More detailed descriptions of these strategies and of ADR in general ....

[Article contains additional citation context not shown here]

A. Afework, M. D. Beynon, F. Bustamante, A. Demarzo, R. Ferreira, R. Miller, M. Silberman, J. Saltz, A. Sussman, and H. Tsang. Digital dynamic telepathology - the Virtual Microscope. In Proceedings of the 1998 AMIA Annual Fall Symposium. American Medical Informatics Association, Nov. 1998.

.... that employ large scale scientific datasets, including applications that explore, compare, and visualize results generated by large scale simulations [8] visualize and generate data products from global coverage satellite data [4] and visualize and analyze digitized microscopy images [1]. Such applications often use only a subset of all the data available in both the input and output datasets. References to data items are described by a range query, namely a multi dimensional bounding box in the underlying multi dimensional attribute space of the dataset(s) Only the data items ....

....counts of the strategies when input data elements are uniformly distributed in the attribute space of the output dataset, restricting the output dataset to be a regular d dimensional array. We validate these cost models with queries for synthetic datasets and for several driving applications [1, 4, 8]. 2 Overview of ADR In this section we briefly describe three strategies for processing range queries in ADR. First we briefly describe how datasets are stored in ADR, and outline the main phases of query execution in ADR. More detailed descriptions of these strategies and of ADR in general can ....

[Article contains additional citation context not shown here]

A. Afework, M. D. Beynon, F. Bustamante, A. Demarzo, R. Ferreira, R. Miller, M. Silberman, J. Saltz, A. Sussman, and H. Tsang. Digital dynamic telepathology - the Virtual Microscope. In Proceedings of the 1998 AMIA Annual Fall Symposium. American Medical Informatics Association, Nov. 1998.

....and engineering. Moreover, an increasing number of applications make use of very large multi dimensional datasets. Examples of such datasets include raw and processed sensor data from satellites [12] output from hydrodynamics and chemical transport simulations [10] and archives of medical images [1]. Many applications that make use of multi dimensional datasets have several important characteristics. Both the input and the output are often disk resident datasets. Applications may use only a subset of all the data available in input and output datasets. Access to data items is described by a ....

.... results stored in the accumulator are post processed to produce final results (steps 9 11) Typical examples of application classes that make use of multi dimensional scientific datasets are satellite data processing applications [15, 5] the Virtual Microscope and analysis of microscopy data [1], and simulation systems for water contamination studies [10] Due to limited space, we briefly describe the satellite data processing application here. In satellite data processing, earth scientists study the earth by processing remotely sensed data continuously acquired from satellite based ....

A. Afework, M. D. Beynon, F. Bustamante, A. Demarzo, R. Ferreira, R. Miller, M. Silberman, J. Saltz, A. Sussman, and H. Tsang. Digital dynamic telepathology - the Virtual Microscope. In Proceedings of the 1998 AMIA Annual Fall Symposium. American Medical Informatics Association, Nov. 1998.

No context found.

A. Afework, M. D. Beynon, F. Bustamante, A. Demarzo, R. Ferreira, R. Miller, M. Silberman, J. Saltz, A. Sussman, and H. Tsang. Digital dynamic telepathology - the Virtual Microscope. In AMIA98. American Medical Informatics Association, November 1998. Also available as University of Maryland Technical Report CS-TR-3892 and UMIACS-TR98 -23.

No context found.

A. Afework, M. D. Beynon, F. Bustamante, A. Demarzo, R. Ferreira, R. Miller, M. Silberman, J. Saltz, A. Sussman, and H. Tsang. Digital dynamic telepathology - the Virtual Microscope. In Proceedings of the 1998.

No context found.

A. Afework, M. D. Beynon, F. Bustamante, A. Demarzo, R. Ferreira, R. Miller, M. Silberman, J. Saltz, A. Sussman, and H. Tsang. Digital dynamic telepathology - the Virtual Microscope. In Proc. of the 1998.

No context found.

A. Afework, M. D. Beynon, F. Bustamante, A. Demarzo, R. Ferreira, R. Miller, M. Silberman, J. Saltz, A. Sussman, and H. Tsang. Digital dynamic telepathology - the Virtual Microscope. In Proceedings of the 1998.

....inquiries into the slide image database. This paper describes the design and implementation of a complete software system, called the Virtual Microscope (VM) that implements a realistic digital emulation of a high power light microscope, through a client server hardware and software architecture [2], 17] The client software runs on an end user s PC or workstation, providing a graphical user interface (GUI) for viewing slides, while the database software for storing, retrieving and processing the microscope image data runs on a parallel computer or on a cluster of workstations at a ....

....calls, resulting in about 85 slower execution than the original custom VM server implementation. The current implementation eliminates the extra function calls and achieves much better response times. The current ADR implementation of the VM server is only 6. 6 slower than the original VM server [2], 17] We also examine the effect on performance of partitioning a VM dataset into data chunks, and look at the scalability of the ADR implementation, when the number of clients and the number of processors are varied. We compare the performance of the ADR implementation of the VM server ....

[Article contains additional citation context not shown here]

A. Afework, M. D. Beynon, F. Bustamante, A. Demarzo, R. Ferreira, R. Miller, M. Silberman, J. Saltz, A. Sussman, and H. Tsang. Digital dynamic telepathology - the Virtual Microscope. In Proceedings of the 1998.

....so that it can be used to completely or partially satisfy a new query. We call the use of such projection operations active caching. With this mechanism, the system has a better chance to exploit reuse than with conventional caching. Based on our experience with Kronos and other applications [2, 11, 14, 27], we have identified four kinds of projection primitives based on the type of reuse they can leverage: dimensional overlap, composable reduction operations, invertible functions, and inductive functions. Dimensional (Spatio temporal) Overlap Primitives: In applications dealing with range ....

A. Afework, M. D. Beynon, F. Bustamante, A. Demarzo, R. Ferreira, R. Miller, M. Silberman, J. Saltz, A. Sussman, and H. Tsang. Digital dynamic telepathology - the Virtual Microscope. In AMIA98. American Medical Informatics Association, November 1998. Also available as University of Maryland Technical Report CS-TR-3892 and UMIACS-TR-9823.

....which will make it fully integrated into the Grid ecosystem. 6 Applications Wenow briefly describe the two applications used as case studies for this paper. A more detailed description of these applications can be found in [6] 6. 1 Analysis of Microscopy Data The Virtual Microscope (VM) [2] is an application designed to support interactive viewing and processing of digitized images of tissue specimens. The raw data for such a system can be captured by digitally scanning collections of full microscope slides at high resolution. A VM query describes a 2 dimensional region in a slide, ....

A. Afework, M. D. Beynon, F. Bustamante, A. Demarzo, R. Ferreira, R. Miller, M. Silberman, J. Saltz, A. Sussman, and H. Tsang. Digital dynamic telepathology - the Virtual Microscope. In AMIA98. American Medical Informatics Association, November 1998. Also available as University of Maryland Technical Report CS-TR-3892 and UMIACS-TR98 -23.

....Section 5, we present and discuss the experimental results we have obtained. Related research is compared to our work in Section 6. Finally, conclusions and future extensions to our current work are given in Section 7. 2 System Architecture We have implemented the Virtual Microscope application [2] using a generic multiple query aware middleware for data analysis applications. This middleware infrastructure, which consists of a multithreaded query server engine, was previously described in [3] but we provide a description of the middleware here in order to help with the presentation of the ....

....base classes defined in our core infrastructure. 3 Analysis of Microscopy Data with The Virtual Microscope Before we describe the multi query scheduling model, we present the implementation of a microscopy visualization application using our middleware. The Virtual Microscope (VM) application [2] implements a realistic digital emulation of a high power light microscope. Figure 2(a) displays the VM client GUI. VM can not only emulate the behavior of a physical microscope, including continuously moving the stage and changing magnification, but also provides functionality that is impossible ....

Asmara Afework, Michael D. Beynon, Fabian Bustamante, Angelo Demarzo, Renato Ferreira, Robert Miller, Mark Silberman, Joel Saltz, Alan Sussman, and Hubert Tsang. Digital dynamic telepathology - the Virtual Microscope. In AMIA98. American Medical Informatics Association, Nov 1998.

....becomes inadequate to hold new computed intermediate results. 5 Applications We now briefly describe the two applications used as case studies for this paper. A more detailed description of these applications can be found in [5] 5. 1 Analysis of Microscopy Data: The Virtual Microscope (VM) [2] is an application designed to support interactive viewing and processing of digitized images of tissue specimens. The raw data for such a system can be captured by digitally scanning collections of full microscope slides at high resolution. A VM query describes a 2 dimensional region in a slide, ....

A. Afework, M.D. Beynon, F. Bustamante, A. Demarzo, R. Ferreira, R. Miller, M. Silberman, J. Saltz, A. Sussman, and H. Tsang. Digital dynamic telepathology - the Virtual Microscope. In AM1A98. American Medical Informatics Association, November 1998. Also available as University of Maryland Technical Report CS-TR-3892 and UMIACS-TR- 98-23.

....of science and engineering. An increasing number of applications make use of very large multidimensional datasets. Examples of such datasets include raw and processed sensor data from satellites [23] output from hydrodynamics and chemical transport simulations [19] and archives of medical images [1]. We are developing a compiler which processes data intensive applications written in a dialect of Java and compiles them for efficient execution on cluster of workstations or distributed memory machines [2, 3, 14] Our chosen dialect of Java includes data parallel extensions for specifying ....

....have spatial coordizates associated with them. For example, the pixels in the satellite data processing application have latitude and longitude values associated with them [23] The pixels in a multi resolution virtual microscope image have the x and y coordinates of the image associated with them [1]. Moreover, the actual layout of the data is not regular in terms of the spatial coordinates. Second, the application processes only a subset of the available data, on the basis of spatial coordinates. For example, in the satellite data processing application, only the pixels within a bounding box ....

[Article contains additional citation context not shown here]

Asmara Afework, Michael D. Beynon, Fabian Bustamante, Angelo Demarzo, Renato Ferreira, Robert Miller, Mark Silberman, Joel Saltz, Alan Sussman, and Hubert Tsang. Digital dynamic telepathology - the Virtual Microscope. In Proceedings of the 1998 AMIA Annual Fall Symposium. American MedicM Informatics Association, November 1998.

....datasets. Large scientific datasets arise in many fields. Examples include datasets from long running simulations of time dependent phenomena that periodically generate snapshots of their state [12, 16, 24, 35] archives of raw and processed remote sensing data [23, 26] archives of medical images [1, 36], and gene and protein databases [27, 29] One example of a data analysis applications is water contamination studies. Environmental scientists study the water quality of bays and estuaries using long running hydrodynamics and chemical transport simulations [16] The fluid flow data generated by ....

....sensors. A typical analysis processes satellite data for several days to a year and generates one or more composite images of the area under study [14] We now briefly describe the two applications used as case studies for this paper. Analysis of Microscopy Data: The Virtual Microscope (VM) [1] is an application designed to support interactive viewing and processing of digitized images of tissue specimens. VM provides a realistic digital emulation of a high power light microscope. The raw data for such a system can be captured by digitally scanning collections of full microscope slides ....

A. Afework, M. D. Beynon, F. Bustamante, A. Demarzo, R. Ferreira, R. Miller, M. Silberman, J. Saltz, A. Sussman, and H. Tsang. Digital dynamic telepathology - the Virtual Microscope. In AMIA98. American Medical Informatics Association, November 1998. Also available as University of Maryland Technical Report CS-TR-3892 and UMIACS-TR98 -23.

....of large datasets in a wide range of scientific application areas. We have used ADR to develop applications in diverse fields, including coupling of multiple scientific simulation codes [10] analysis and processing of satellite datasets [13] analysis and visualization of microscopy data [1], and volume rendering and iso surface rendering to support visualization of very large datasets [9] The image analysis framework discussed in this paper customizes ADR s parallel back end (PBE) and front end (FE) The implementation also provides a tool for loading and cataloging ....

A. Afework, M. D. Beynon, F. Bustamante, A. Demarzo, R. Ferreira, R. Miller, M. Silberman, J. Saltz, A. Sussman, and H. Tsang. Digital dynamic telepathology - the Virtual Microscope. In Proceedings of the

....with points in a multidimensional attribute space) is an important component of science and engineering. Examples of these datasets include raw and processed sensor data from satellites [27] output from hydrodynamics and chemical transport simulations [23] and archives of medical images[1]. These datasets are also very large, for example, in medical imaging, the size of a single digitized composite slide image at high power from a light microscope is over 7GB (uncompressed) and a single large hospital can process thousands of slides per day. Applications that make use of ....

.... object oriented parallel systems like Titanium [36] HPC [5] and Concurrent Aggregates [11] 6 Interface Reducinterface f Any object of any class implementing this interface is a reduction variable g public class VMPixel f char colors[3] void Initialize( f colors[0] 0 ; colors[1] = 0 ; colors[2] 0 ; g Aggregation Function void Accum(VMPixel Apixel, int avgf) f colors[0] Apixel.colors[0] avgf ; colors[1] Apixel.colors[1] avgf ; colors[2] Apixel.colors[2] avgf ; g g public class VMPixelOut extends VMPixel implements Reducinterface; public class ....

[Article contains additional citation context not shown here]

Asmara Afework, Michael D. Beynon, Fabian Bustamante, Angelo Demarzo, Renato Ferreira, Robert Miller, Mark Silberman, Joel Saltz, Alan Sussman, and Hubert Tsang. Digital dynamic telepathology - the Virtual Microscope. In Proceedings of the

....tools for discovery, allocation, and management of distributed resources. As a result, long running, large scale simulations [20,46,56,58] are producing unprecedented amounts of data. In addition, advanced sensors attached to instruments, such as earth orbiting satellites and medical instruments [3,62], are generating very large datasets that must be made available to a wider audience. Looking at available technology, disk space has become plentiful and relatively inexpensive. Using off the shelf components, it is currently possible to build a disk based storage cluster with about 1 Terabyte ....

....collections of very large multi dimensional datasets. Examples of such applications include satellite data processing [24,27,62] full scale water contamination studies and surface subsurface petroleum reservoir simulations [44,66] visualization and processing of digitized microscopy images [3], visualization of largescale data [5,8,29,42,61] and data mining [4,7,34,68] Although the datasets used for analysis and the data products generated by applications that manipulate those datasets may differ in many ways, a close look at many dataintensive applications [17,21,31,42,44] reveals ....

[Article contains additional citation context not shown here]

A. Afework, M. D. Beynon, F. Bustamante, A. Demarzo, R. Ferreira, R. Miller, M. Silberman, J. Saltz, A. Sussman, and H. Tsang. Digital dynamic telepathology - the Virtual Microscope. In Proceedings of the

....distributed processing power, when there are fewer queries than the number of processors. 3 We are in the process of implementing different versions of the send method to minimize communication overheads. 5 4 Example Application: The Virtual Microscope The Virtual Microscope (VM) application [1] implements a realistic digital emulation of a high power light microscope. VM can be used in a training environment, where a group of fellows or students may examine and manipulate the same set of slides. In such a setting, the data server has to process multiple queries simultaneously. The ....

A. Afework, M. D. Beynon, F. Bustamante, A. Demarzo, R. Ferreira, R. Miller, M. Silberman, J. Saltz, A. Sussman, and H. Tsang. Digital dynamic telepathology - the Virtual Microscope. In AMIA98. American Medical Informatics Association, Nov 1998.

....for our satellite data processing application Titan consists of over 1. 4TB of data [11] For our Virtual Microscope application, one focal plane of a single slide requires over 7GB (uncompressed) at high power, and a hospital such as Johns Hopkins produces hundreds of thousands of slides per year [1]. Similarly, the computation for one ten day composite Titan query for the entire world takes about 100 seconds per processor on the Maryland sixteen node IBM SP2. Efficient processing for these data intensive applications clearly requires use of multiple processors, with an associated very large ....

....Java and will also be applicable to suitable extensions of other languages, such as C or C . 3 Interface Reducinterface f Any object of any class implementing this interface is a reduction variable g public class VMPixel f char[ colors; void Initialize( f colors[0] 0 ; colors[1] = 0 ; colors[2] 0 ; g void Accum(VMPixel Apixel, int avgf) f colors[0] Apixel.colors[0] avgf ; colors[1] Apixel.colors[1] avgf ; colors[2] Apixel.colors[2] avgf ; g g public class VMPixelOut extends VMPixel implements Reducinterface; public class VMScope f static int Xdimen ....

[Article contains additional citation context not shown here]

Asmara Afework, Michael D. Beynon, Fabian Bustamante, Angelo Demarzo, Renato Ferreira, Robert Miller, Mark Silberman, Joel Saltz, Alan Sussman, and Hubert Tsang. Digital dynamic telepathology - the Virtual Microscope. In Proceedings of the 1998 AMIA Annual Fall Symposium. American Medical Informatics Association, November 1998.

....datasets and an execution environment for expensive user defined functions. This would allow many scientific applications to minimize programming complexity and take advantage of the built in features of an ORDBMS, while still achieving good performance. Our study of a large set of applications [1, 2, 8, 12] indicates that they often share several characteristics. First, the data is usually multi dimensional, meaning that the data elements are associated with points in some multidimensional attribute space. Also, the processing usually follows a well defined sequence of basic steps consisting of (1) ....

....developer defines an ADR dataset in SQL:1999, assume we are interested in building an application that serves large images to client programs, where a query is specified by a bounding box into the image data and a subsampling factor, as for the U. Maryland Johns Hopkins Hospital Virtual Microscope [2] project. The most straightforward way to define such a dataset is to use a simple extension to SQL:1999 type declarations, by which an application developer can create a multi dimensional array of objects of a given type, for example: CREATE ADRSET imaget OF TYPE pixelt (x, y) This declaration ....

A. Afework, M. D. Beynon, F. Bustamante, A. Demarzo, R. Ferreira, R. Miller, M. Silberman, J. Saltz, A. Sussman, and H. Tsang. Digital dynamic telepathology - the Virtual Microscope. In Proceedings of the 1998 AMIA Annual Fall Symposium. American Medical Informatics Association, Nov. 1998.

....many fields of science and engineering. Examples of such large datasets include collections of raw and processed sensor data from satellites [21] output of long running simulations of time dependent phenomena that periodically generate snapshots of their state [18] and archives of medical images [2]. A large fraction of the applications that make use of such datasets share several important characteristics. First, the dataset is usually multi dimensional which means that it represents attributes associated with points in some multi This researchwas supportedby NSF grant ACR 9982087awarded ....

....is also highly stylized. It basically consists of retrieving the input elements for the region of interest, mapping these input points into output points and aggregating the values of input that map to the same output. One very simple example of such application is the Virtual Microscope [2], 13] Large quantities of microscope slides are kept digitized at high resolution in some storage device. Users want to browse some slides looking for interesting features on the images. The browsing happens in low resolution and the users switch to higher resolutions as they believe they see an ....

[Article contains additional citation context not shown here]

Asmara Afework, Michael D. Beynon, Fabian Bustamante, Angelo Demarzo, Renato Ferreira, Robert Miller, Mark Silberman, Joel Saltz, Alan Sussman, and Hubert Tsang. Digital dynamic telepathology - the Virtual Microscope. In Proceedings of the 1998 AMIA Annual Fall Symposium. American Medical Informatics Association, November 1998.

....of science and engineering. An increasing number of applications make use of very large multidimensional datasets. Examples of such datasets include raw and processed sensor data from satellites [23] output from hydrodynamics and chemical transport simulations [19] and archives of medical images [1]. We are developing a compiler which processes data intensive applications written in a dialect of Java and compiles them for efficient execution on cluster of workstations or distributed memory machines [2, 3, 14] Our chosen dialect of Java includes data parallel extensions for specifying ....

....have spatial coordinates associated with them. For example, the pixels in the satellite data processing application have latitude and longitude values associated with them [23] The pixels in a multi resolution virtual microscope image have the x and y coordinates of the image associated with them [1]. Moreover, the actual layout of the data is not regular in terms of the spatial coordinates. Second, the application processes only a subset of the available data, on the basis of spatial coordinates. For example, in the satellite data processing application, only the pixels within a bounding ....

[Article contains additional citation context not shown here]

Asmara Afework, Michael D. Beynon, Fabian Bustamante, Angelo Demarzo, Renato Ferreira, Robert Miller, Mark Silberman, Joel Saltz, Alan Sussman, and Hubert Tsang. Digital dynamic telepathology - the Virtual Microscope. In Proceedings of the 1998 AMIA Annual Fall Symposium. American Medical Informatics Association, November 1998.

....and engineering. An increasing number of applications makeuseof very large multidimensional datasets. Examples of such datasets include raw and processed sensor data from satellites [31] output from hydrodynamics and chemical transport simulations [25] and archives of medical images [2]. This work was supported by NSF grantACR 9982087, NSF CAREER award ACI 9733520, and NSF grant CCR 9808522. Many of these domains involve complex and sparse datasets. For example, the dataset captured by a satellite can be viewed as a sparse three dimensional array, where time, latitude, and ....

....systems like Titanium [38] HPC [7] and Concurrent Aggregates [13, 33] and are not unique to our approach. These constructs are: Interface Reducinterface f f Any object of a class implementing g f this interfaceisareduction variable g g public class pixel f short bands[5] # short geo[2] # g class block f short time # pixel bands[204 204] # pixel getData(Point[2] p) f f Search for the (lat, long) on geodata g f Return the pixel if it exists g f elsereturn null g g g f Low level Data Layout g class SatOrigData f block[1d] data # void SatOrigData(RectDomain[1] ....

[Article contains additional citation context not shown here]

Asmara Afework, Michael D. Beynon, Fabian Bustamante, Angelo Demarzo, Renato Ferreira, Robert Miller, Mark Silberman, Joel Saltz, Alan Sussman, and Hubert Tsang. Digital dynamic telepathology - the Virtual Microscope. In Proceedings of the 1998 AMIA Annual Fall Symposium. American Medical Informatics Association, November 1998.

....simulations is to better understand the problem at hand. Understanding can be achieved by analyzing the data generated by simulations and experiments. With the help of more powerful computers and advanced sensors, long running, large scale simulations [8, 29, 40] and experimental measurements [1, 34, 42] are generating collections of very large datasets. The availability of low cost systems built from networks of highperformance commodity computers and high capacity, commodity disks has greatly enhanced a scientist s ability to store large scale scientific data. For instance, a PC cluster with ....

....of applications. We have used ADR to develop applications in diverse fields, including coupling of simulation codes [25] analysis and processing of satellite datasets [12, 38] volume shape analysis from multi perspective video sequences [7] and analysis and visualization of microscopy data [1]. In this section, we briefly describe some of these applications. 4.1 Satellite Data Processing Earth scientists study the earth by processing remotely sensed data continuously acquired from satellitebased sensors, since a significant amount of earth science research is devoted to developing ....

[Article contains additional citation context not shown here]

A. Afework, M. D. Beynon, F. Bustamante, A. Demarzo, R. Ferreira, R. Miller, M. Silberman, J. Saltz, A. Sussman, and H. Tsang. Digital dynamic telepathology - the Virtual Microscope. In Proceedings of the 1998 AMIA Annual Fall Symposium. American Medical Informatics Association, Nov. 1998.

.... that employ large scale scientific datasets, including applications that explore, compare, and visualize results generated by large scale simulations [15] visualize and generate data products from global coverage satellite data [7] and visualize and analyze digitized microscopy images [1]. Such # This research was supported by the National Science Foundation under Grant #ACI 9619020 (UC Subcontract # 10152408) and the Office of Naval Research under Grant #N6600197C8534. 1 O # Output Dataset, I Input Dataset ( Initialization ) 1. foreach o e in O do 2. read o e 3. a e # ....

....of the strategies when input data elements are uniformly distributed in the attribute space of the output dataset, restricting the output dataset to be a regular d dimensional array. We present an experimental evaluation of these models for synthetic datasets and for several driving applications [1, 7, 15]. 2 Query Execution Strategies In this section we briefly describe three strategies for processing range queries in ADR. First we briefly describe how datasets are stored in ADR, and outline the main phases of query execution in ADR.More detailed descriptions of these strategies and of ADR in ....

[Article contains additional citation context not shown here]

A. Afework, M. D. Beynon, F. Bustamante, A. Demarzo, R. Ferreira, R. Miller, M. Silberman, J. Saltz, A. Sussman, and H. Tsang. Digital dynamic telepathology - the Virtual Microscope. In Proceedings of the 1998 AMIA Annual Fall Symposium. American Medical Informatics Association, Nov. 1998.

....and engineering. Moreover, an increasing number of applications make use of very large multi dimensional datasets. Examples of such datasets include raw and processed sensor data from satellites [23] output from hydrodynamics and chemical transport simulations [19] and archives of medical images [2]. For example, a dataset of coarse grained satellite data (with 4.4 km pixels) covering the whole earth surface and captured over a relatively short period of time (10 days) is about 4.1GB; a finer grained version (1.1 km per pixel) contains about 65 GB of sensor data. In medical imaging, the ....

.... results stored in the accumulator are post processed to produce final results (steps 9 11) Some typical examples of applications that make use of multi dimensional scientific datasets are satellite data processing applications [1, 8, 30] the Virtual Microscope and analysis of microscopy data [2, 14], and simulation systems for water contamination studies [19] In satellite data processing, for example, earth scientists study the earth by processing remotely sensed data continuously acquired from satellite based sensors. Each sensor reading is associated with a position (longitude and ....

[Article contains additional citation context not shown here]

A. Afework, M. D. Beynon, F. Bustamante, A. Demarzo, R. Ferreira, R. Miller, M. Silberman, J. Saltz, A. Sussman, and H. Tsang. Digital dynamic telepathology - the Virtual Microscope. In Proceedings of the 1998 AMIA Annual Fall Symposium. American Medical Informatics Association, Nov. 1998.

....the projection that best suits her needs. Sensor values are pre processed to correct the effects of various distortions, such as instrument drift, atmospheric distortion and topographic effects, before they are used. 2. 2 Virtual Microscope and Analysis of Microscopy Data The Virtual Microscope [5, 16] provides a realistic digital emulation of a high power light microscope. The raw data for such a system can be captured by digitally scanning collections of full microscope slides under high power. The digitized images from a slide are effectively a threedimensional dataset, since each slide can ....

A. Afework, M. D. Beynon, F. Bustamante, A. Demarzo, R. Ferreira, R. Miller, M. Silberman, J. Saltz, A. Sussman, and H. Tsang. Digital dynamic telepathology - the Virtual Microscope. In Proceedings of the 1998 AMIA Annual Fall Symposium. American Medical Informatics Association, Nov. 1998.

....to copying and loss of locality. Furthermore, every application developer has to implement support for managing and scheduling the processing. Over the past three years, we have been working with several scientific research groups to understand the processing requirements for such applications [1, 2, 6, 16]. Our study of a large set of applications indicates that the processing for such datasets is often highly stylized and shares several important characteristics. Usually, both the input dataset as well as the result being computed are multi dimensional. The basic processing step usually consists ....

....reduce the amount of memory available on the back end nodes for system total processing time computation others VM 1.74 0.36 1.38 T2 3.22 1.39 1.86 Table 3. Query processing times (in sec. for VM and T2. query processing. 3. 2 The Virtual Microscope The Virtual Microscope (VM) system [2] provides the ability to access high power, high resolution digital images of entire pathology slides. A VM query specifies a highpower digitized microscopy image, the region of interest and the desired magnification for display, and an output image is computed by subsampling the high power input ....

A. Afework, M. D. Beynon, F. Bustamante, A. Demarzo, R. Ferreira, R. Miller, M. Silberman, J. Saltz, A. Sussman, and H. Tsang. Digital dynamic telepathology - the Virtual Microscope. In Proceedings of the 1998 AMIA Annual Fall Symposium. American Medical Informatics Association, Nov. 1998.

....: 28 3.6. 3 Filters : 28 3 1 Introduction Many scientific applications generate and use datasets consisting of data values associated with a multi dimensional space [5, 1, 7]. Scientific simulations typically generate datasets with at least three spatial dimensions and a temporal dimension. Satellite data and microscopy data generally have two (or more) spatial dimensions and a temporal dimension. Applications frequently need to access spatially defined data subsets ....

.... visualize results generated by multiple very large scale simulations [7] 2) programs that visualize or generate data products from global coverage satellite data [5] and (3) applications that visualize and classify microscopy data and carry out content based queries that return data subsets [1]. Spatial subsets can encompass contiguous regions of space, as for retrieving satellite data covering a particular geographical region. Spatial subsets can also be defined once features of interest are categorized using spatial indices. For instance, subsetting can be carried out to retrieve ....

Asmara Afework, Michael D. Beynon, Fabian Bustamante, Angelo Demarzo, Renato Ferreira, Robert Miller, Mark Silberman, Joel Saltz, Alan Sussman, and Hubert Tsang. Digital dynamic telepathology - the Virtual Microscope. In Proceedings of the 1998 AMIA Annual Fall Symposium. American Medical Informatics Association, November 1998.

....and engineering simulations. In addition, large amounts of data are being gathered by sensors of various sorts, attached to devices such as satellites and microscopes. There are many examples of large useful datasets from simulations [26, 29, 33] sensor data [25, 28] and medical imaging [2] (pathology, MRI, CT scan, etc. The primary goal of generating data through large scale simulations or sensors is to better understand the causes and effects of physical phenomena. Understanding is achieved through running analysis codes on the stored data, or by a more interactive visualization ....

....deterministic volume, which can be used for choosing placement for the entire execution. For the applications we are targeting, such as volume visualization, database decision support, and image processing, these assumptions appear to hold. 5. Application: Image Processing The Virtual Microscope [2] is a query response application that processes multi dimensional image data to satisfy client queries. The dataset contains high power digitized images of microscope slides, which effectively forms a 3D dataset because each slide can contain multiple 2D focal planes at different depths. Images ....

[Article contains additional citation context not shown here]

A. Afework, M. D. Beynon, F. Bustamante, A. Demarzo, R. Ferreira, R. Miller, M. Silberman, J. Saltz, A. Sussman, and H. Tsang. Digital dynamic telepathology - the Virtual Microscope. In Proceedings of the 1998 AMIA Annual Fall Symposium. American Medical Informatics Association, November 1998.

....by the Office of Naval Research under Grant #N66001 97 C 8534. of data are being gathered by sensors of various sorts, attached to devices such as satellites and microscopes. There are many examples of large useful datasets from simulations [26, 29, 33] sensor data [25, 28] and medical imaging [2] (pathology, MRI, CT scan, etc. The primary goal of generating data through large scale simulations or sensors is to better understand the causes and effects of physical phenomena. Understanding is achieved through running analysis codes on the stored data, or by a more interactive visualization ....

....deterministic volume, which can be used for choosing placement for the entire execution. For the applications we are targeting, such as volume visualization, database decision support, and image processing, these assumptions appear to hold. 5. Application: Image Processing The Virtual Microscope [2] is a query response application that processes multi dimensional image data to satisfy client queries. The dataset contains high power digitized images of microscope slides, which effectively forms a 3D dataset because each slide can contain multiple 2D focal planes at different depths. Images ....

[Article contains additional citation context not shown here]

A. Afework, M. D. Beynon, F. Bustamante, A. Demarzo, R. Ferreira, R. Miller, M. Silberman, J. Saltz, A. Sussman, and H. Tsang. Digital dynamic telepathology - the Virtual Microscope. In Proceedings of the 1998 AMIA Annual Fall Symposium. American Medical Informatics Association, November 1998.

....systems across a wide area network. DataCutter provides support for subsetting of datasets through multi dimensional range queries, and application specific aggregation on scientific datasets stored in an archival storage system. We discuss an implementation of the Virtual Microscope application [2] using DataCutter. The Virtual Microscope is representative of data intensive applications that involve browsing and processing large multi dimensional datasets. Other examples include satellite data processing systems [7] and water contamination studies that couple multiple simulators [20] We ....

.... that employ large scale scientific datasets, including applications that explore, compare, and visualize results generated by large scale simulations [20] visualize and generate data products from global coverage satellite data [7] and visualize and analyze digitized microscopy images [2]. Many scientific applications generate and use datasets consisting of data values associated with a multi dimensional space. Scientific simulations typically generate datasets with at least three spatial dimensions and a temporal dimension. Satellite data and microscopy data generally have two ....

[Article contains additional citation context not shown here]

A. Afework, M. D. Beynon, F. Bustamante, A. Demarzo, R. Ferreira, R. Miller, M. Silberman, J. Saltz, A. Sussman, and H. Tsang. Digital dynamic telepathology - the Virtual Microscope. In Proceedings of the 1998 AMIA Annual Fall Symposium. American Medical Informatics Association, Nov. 1998.

.... that employ large scale scientific datasets, including applications that explore, compare, and visualize results generated by large scale simulations [15] visualize and generate data products from global coverage satellite data [7] and visualize and analyze digitized microscopy images [1]. Such applications often use only a subset of all the data available in both the input and output datasets. References to data items are described by a range query, namely a multi dimensional bounding box in the underlying multi dimensional attribute space of the dataset(s) Only the data items ....

....of the strategies when input data elements are uniformly distributed in the attribute space of the output dataset, restricting the output dataset to be a regular d dimensional array. We present an experimental evaluation of these models for synthetic datasets and for several driving applications [1, 7, 15]. 2 Query Execution Strategies In this section we briefly describe three strategies for processing range queries in ADR. First we briefly describe how datasets are stored in ADR, and outline the main phases of query execution in ADR.More detailed descriptions of these strategies and of ADR in ....

[Article contains additional citation context not shown here]

A. Afework, M. D. Beynon, F. Bustamante, A. Demarzo, R. Ferreira, R. Miller, M. Silberman, J. Saltz, A. Sussman, and H. Tsang. Digital dynamic telepathology - the Virtual Microscope. In Proceedings of the 1998 AMIA Annual Fall Symposium. American Medical Informatics Association, Nov. 1998.

....and engineering. Moreover, an increasing number of applications make use of very large multi dimensional datasets. Examples of such datasets include raw and processed sensor data from satellites [23] output from hydrodynamics and chemical transport simulations [19] and archives of medical images [2]. For example, a dataset of coarse grained satellite data (with 4.4 km pixels) covering the whole earth surface and captured over a relatively short period of time (10 days) is about 4.1GB; a finer grained version (1.1 km per pixel) contains about 65 GB of sensor data. In medical imaging, the ....

.... results stored in the accumulator are post processed to produce final results (steps 9 11) Some typical examples of applications that make use of multi dimensional scientific datasets are satellite data processing applications [1, 8, 30] the Virtual Microscope and analysis of microscopy data [2, 14], and simulation systems for water contamination studies [19] In satellite data processing, for example, earth scientists study the earth by processing remotely sensed data continuously acquired from satellite based sensors. Each sensor reading is associated with a position (longitude and ....

[Article contains additional citation context not shown here]

A. Afework, M. D. Beynon, F. Bustamante, A. Demarzo, R. Ferreira, R. Miller, M. Silberman, J. Saltz, A. Sussman, and H. Tsang. Digital dynamic telepathology - the Virtual Microscope. In Proceedings of the 1998 AMIA Annual Fall Symposium. American Medical Informatics Association, Nov. 1998.

....existing attributes to synthesize new attributes. Since the same physical entity may be described in a complementary manner by different types of datasets, applications may need to generate new datasets by performing joins over pre existing datasets. Our study of a large set of applications [1, 2, 6, 11, 14, 18, 21] indicates that the processing is often highly stylized and shares several important characteristics. The basic processing step usually consists of mapping multi dimensional coordinates of the retrieved data items to the coordinates of the proper output data items, and computing output data items ....

....their values from the database, instead of with the identity element for the aggregation function. Typical examples of applications that make use of multi dimensional scientific datasets are satellite data processing applications [1, 21, 6] the Virtual Microscope and analysis of microscopy data [2, 11], and simulation systems for water contamination studies [13] In satellite data processing, for example, earth scientists study the earth by processing remotely sensed data continuously acquired from satellite based sensors. Each sensor reading is associated with a position (longitude and ....

[Article contains additional citation context not shown here]

A. Afework, M. D. Beynon, F. Bustamante, A. Demarzo, R. Ferreira, R. Miller, M. Silberman, J. Saltz, A. Sussman, and H. Tsang. Digital dynamic telepathology - the Virtual Microscope. In Proceedings of the 1998 AMIA Annual Fall Symposium. American Medical Informatics Association, Nov. 1998.

....call for digitizing tens of thousands of such slides. These images are to be used for telepathology, medical research and pedagogy and require a variety of processing including three dimensional reconstruction of tissue sections, image segmentation, virtual staining and histological image analysis [3]. In this paper, we evaluate architectural alternatives for scaling the processing power with the growth in dataset size. We consider three alternatives: Active Disks [2, 19, 23, 31] see section 2 for a brief review of Active Disks. clusters and shared memory multiprocessors (SMPs) Each of ....

A. Afework, M. Beynon, F. Bustamante, A. Demarzo, R. Ferriera, R. Miller, M. Silberman, J. Saltz, A. Sussman, and H. Tsang. Digital dynamic telepathology -- the virtual microscope. In Proceedings of the AMIA'98 Fall Symposium, 1998. To appear. 2 http://www.emulex.com

No context found.

A. Afework, M. D. Beynon, F. Bustamante, A. Demarzo, R. Ferreira, R. Miller, M. Silberman, J. Saltz, A. Sussman, and H. Tsang. Digital dynamic telepathology - the Virtual Microscope. In Proceedings of the 1998.

No context found.

Afework A, Benyon M, Bustamante F, et al. Digital Dynamic Telepathology -- the Virtual Microscope. Proc AMIA Fall Symposium; 1998 Nov.. AMIA, Hanley and Belfus, pp 912-916.

No context found.

Afework, A., Beynon, M. D., Bustamante, F., Demarzo, R., Ferreira, R., Miller, R., Silberman, M., Saltz, J., Sussman, A., and Tsang, H. Digital dynamic telepathology - the Virtual Microscope. 1998.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC