Our Faculty, Staff & Students

Contact Us

Professor and Department Head

Dr. Monty Escabi
Phone: (860) 486-0063
Email: monty.escabi@uconn.edu

Staff

Lisa Ephraim
Undergraduate Academic Advisor
Phone: (860) 486-0163
E-mail: lisae@engr.uconn.edu

Jennifer Seyford
Financial Assistant II
Phone: (860) 486-0116
E-mail: jennifer.seyford@uconn.edu

Birgit Sawstrom
Admin Service Assistant III
Phone: (860) 486-5838
E-mail: birgit.sawstrom@uconn.edu

Sowmya Ramesh
Financial Assistant I
Phone: (860) 486-7139
E-mail: sowmya.ramesh@uconn.edu

 

Main Office Address

Biomedical Engineering Department
A.B. Bronwell Building, Room 217
260 Glenbrook Road, Unit 3247
University of Connecticut
Storrs, CT 06269-3247
Phone: (860) 486-5838
Fax: (860) 486-2500

Dr. Yong-Jun Shin Receives Microsoft Azure Research Award

Prof. Yong-Jun Shin, a BME core faculty member, has received the Microsoft Azure Research Award. 

“Cloud-enabled Smart Microscopes for Biological Research”

Project Description

Biological imaging enabled by advances in microscopy and software (image processing, etc.) has played an important role in biological research. µManager is an open-source, cross-platform desktop application, to control a wide variety of motorized microscopes, scientific cameras, stages, illuminators, and other microscope accessories [1, 2]. Since its development in 2005, µManager has grown to support a wide range of microscopy hardware and is now used by thousands of researchers around the world. Combined with ImageJ [3], a public domain, Java-based image processing program developed at the National Institutes of Health, the application provides a mature graphical user interface and offers open programming interfaces (Java, Python, and MATLAB) to facilitate plugins and scripts, which can be used to automate various microscope and imaging operations (Figure 1).

aziotf1

However, this automation may cause unexpected errors and issues that are difficult to fix for biologists. This project will develop a cloud-enabled smart microscope solution that attempts to solve the problem. It will connect µManager applications in multiple microscopes, which provide data from device manufacturer information to real-time operation status, to Microsoft Azure Machine Learning (Azure ML). Biologists will be able to capture the data, transmit it into the cloud and combine it into a single dashboard that serves up two basic types of data: alarms that indicate an immediate issue, and events, which are stored and used for management. The solution will provide them with instant diagnostic capabilities and rich, real-time data visualization. With Azure ML service, biologists will have a view into microscope operations and maintenance on a continual basis. The system contains an intelligent feedback mechanism: Data from microscopes is fed into dynamic predictive ML models, which iteratively update datasets through seamless integration with Azure. Smart microscopes will help biologists making decisions when solving problems encountered during the operations. The server application on Azure will communicate with Azure ML using Web API and the client applications will communicate with the server application using SignalR, which allows real-time bi-directional communication and broadcasting (Figure 2) [4, 5].  

Resource Requirement

In May 2015, the PI, Prof. Rajasekaran (UCONN Computer Science and Engineering), and Prof. Ramprasad (UCONN Material Science and Engineering) jointly received $1.4 M for building a cloud-enabled HPC infrastructure (hybrid Azure) on UCONN campus, which will be used for this project.  1-year 2 ML Seat Subscription, ML Studio usage, and ML API usage will also be required.

chart2

Figure 2. Service-oriented architecture for cloud-enabled smart microscopes.

 

References

  1. mManager https://www.micro-manager.org/
  2. Arthur D Edelstein, Mark A Tsuchida, Nenad Amodaj, Henry Pinkard, Ronald D Vale, and Nico Stuurman (2014), Advanced methods of microscope control using μManager software.Journal of Biological Methods 2014 1(2):e11 doi:10.14440/jbm.2014.36
  3. Schneider, C. A.; Rasband, W. S. & Eliceiri, K. W. (2012), “NIH Image to ImageJ: 25 years of image analysis”,Nature methods 2012 9(7): 671-675, PMID 22930834
  4. SignalR Javascript Client https://github.com/SignalR/SignalR/wiki/SignalR-JS-Client
  5. SignalR Java Client https://github.com/SignalR/java-client

 

About Microsoft Azure:

Microsoft Research solicits proposals for projects that use Microsoft Azure in research and welcomes research proposals from any branch of scholarly activity. Selected proposals are awarded large allocations of Microsoft Azure storage and compute resources for a period of one year. For more information:

http://research.microsoft.com/en-US/projects/azure/awards.aspx