Drobo - The answer to our storage needs?

Drobo may be the next big thing in file storage.

As a research scientist I have seen my data storage requirements grow and I have been constantly looking for a product that can grow with my needs.

Most recently I have relied on large single disk hard drives as my main storage but with the introduction of data robotics Drobo I am considering that as a viable replacement to my stack of 500GB Lacie drives.

I am not one with enough experience or patience to set up a server or raid device. So when I saw the Data Robotics booth at Photoshop World Boston in 2007 I thought this may be an answer to my prayers directed at the data storage gods.

I was intrigued by three factors. The first is the ability to protect data on the fly using multiple drives in a single piece of hardware without raid configuration. The second is the expansion capability to add drives so the system can grow as hard drive prices decreases and storage needs increase. And third the relative low cost of a unit, inexpensive enough for easy acquisition for lab and personal use.

After multiple encounters with Data Robotics booth staff and explaining to them about the unique needs of Biomedical professionals they sent me an evaluation unit to test out.

I received it about two weeks ago and have used it intensely over that time in the storage of images and other media files as well as backing up my primary production system, an Intel Mac Book Pro.

This is the current picture of the Drobo in use.

Once the box from Data Robotics’ arrived. The unpacking went as expected and I was ready to add SATA hard drives to the unit.

The cover is held in place with a magnetic latch and can easily be removed. Adding the drives was the easiest installation I have seen. The drives slide in ands snap into unseen receptacles fitting power and cables automatically. There are four open drive bays that accept full and half height drives.

Half height 120GB drive in bottom bay.

I won’t bore you with all the details. The rest of the set up went smoothly following the included instructions. I installed the Drobo Dashboard to initialize the Drobo and gain control over options. The Dashboard allows visualization of the status of the unit beyond the informative lights on the front of the Drobo.

The most important factors of any storage device is the ability to hold precious data, “my precious, precious.’

In this regard the Drobo is doing wonderfully. Currently I have approx 650GB of usable data storage with three drives 750, 400, and 320GB. This unbalanced drive size has left 320GB of space reserved for expansion. The kind way to telling that there is some usable space. Using the Drobolator, at www.drobo.com/drobolator, it is easy to configure and optimize drive sizes for the maximum amount of protected storage

Current drive set up in the Drobo and the overall memory situation.

Using the Drobolator I determined that to expand my storage capacity my best next drive purchase should be a 750GB drive (currently ~$180).

Drobolator with optimized hard drive configuration.

My next step will be putting the Drobo through its paces, developing workflows to use the increased storage capacity and confidence in its data protection abilities.

So far it seems to fulfill my in data storage needs. Now lets see how it adds value to my work.

Pros:
1) Provides data protection without the need of Raid configuration
2) Can use SATA drives already on hand
3) Can add additional drives at any time
4) Simple to set up and use
5) I’m am better able to sleep

Improvement Considerations:
1) Correct storage space reported on the Macintosh (Info displays 2TB with Mac operation system)
2) Higher speed transfer (The current Drobo uses USB 2.0 port, why not add additional options)
3) On/off switch ( The unit powers down when the computer is turned off but sometimes I may just want to have the unit off without unplugging the power


Data Robotics Drobo Storage Robot Multiple 4-Drive Enclosure ~ $480 online

Oct 30 - DICOM Workflow Presentation

I have just finished watching the eSeminar titled DICOM workflow for publishing with CS3 Extended.

It seems that one of the slides is a holdover from my eSeminar I gave for Adobe to cover Photoshop CS3 Extended's use as a scientific researcher.

I have been preparing in depth training for lynda.com as well as a book for using Photoshop in biomedical research.

To learn more you can read my articles in the past few issues of Photoshop user.

If I can be of help please contact me at ericjwexlerATyahoo.com.

The other side

As of May 9th 2007 I am in the world of the W9's. Opportunities and need crossed paths where I stood and led me in somewhat different direction. With my involvement on Adobe's Biomedical Image Advisory Group I saw the need of training other scientists in the use of Photoshop for communication and research.

Many perceive Photoshop as a difficult program and the multitudes are intimidated not only by its complexity but by imaging in general. Someday there will be a fourth R taught in school. Along with reading, writing, and rithmatic should be really good imaging.

How many of us are fearful of a word processor. Working with language is as complex as working with images. The power of words can match the power of images. Verbal communication can be eloquent and beautiful or fragmented and disturbing. But many years of training and effort gives us comfort to concentrate more on the content of our words than the controls in the software aiding us in the act of communication. If everyone had as much training in working with visual arts and technology Photoshop would seem less imposing.

All that being said, our society increasingly relies on the visual presentation of information. To take the time to learn and apply imaging resources is an endeavor that will be rewarded by communication your intended message efficiently and effectively.

Then a leap

All good things

From 1995 to 2001 I conducted my own experiments and supported the other researchers at DuPont Pharmaceuticals. In 2001 Bristol Myers Squibb purchased the company and in the span of one year closed down the research site I worked at.

My wife and I relocated to the Boston area where she was offered a nice research position. I relocated and was rehired by Bristol-Myers Squibb at the Medical Imaging site in Billerica, MA.

There I was able to reacquire a majority of the equipment I used at the DuPont site. I then set up a histology lab and a shadow of my digital darkroom.

Over the next 4 years I researched vulnerable plaque, PET perfusion imaging agent, and a variety of other imaging agents, mainly cardiovascular. I applied my knowledge and expanded into new technical areas like autoradiography, cellular kinetics, and special histology stains. All the histology efforts on the site were conducted by myself and my part time intern, Be Luu. Her attention to detail and concentration on the tasks at hand produced samples demonstrating morphology in cryosections better than what I have seen in paraffin sections produced by contract labs.

Again I conducted my own research and supported both discovery and development with my microscopy and imaging skills. Under my supervisor - Padmaja Yalamanchili I was given the opportunity to learn and grow not only in my imaging and histology skills but also in cell biology, tissue based assays, and radioligand based research.

The discovery biology group was small (< 20) and everyone needed to be able to conduct a wide variety of work. The Discovery Biology group put together and managed by Simon Robinson was a dedicated and talented group of people doing extraordinary work in both scientifically and productively with very limited resources. I was proud to be part of that team.

Yet another step

Photoshop works well with others

During the late 90's two of my main projects combined imaging abilities with scientific and technical skills.

The evaluation of cell cycle inhibitors for the oncology group relying on the evaluation of tumors from the xenograft model. With the digital darkroom established multitudes of tumor sections were analyzed by me in my lab. My lab partners handled the in vivo portion of the experiments which took quite a but of there efforts. I would be present at the take down and collect and gross the tumors. Then the sections (no thicker than a nickel to allow for proper fixation) were processed onto paraffin blocks. The samples were sectioned at 5 microns and stained with hematoxylin & eosin for morphological examination. For the analysis of the triad of tumor growth - Proliferation, Angiogenesis, and Apoptosis. Sections were stained using immunohistochemistry of BrdU and Factor VIII , and the Apotag in situ kit.

The image acquisition and analysis of MCID M2 system worked well with the image compositions and output features of Photoshop. I was able to effectively and efficiently communicate my findings by creating composites of the different stained images of the serial sections of the tumors.

The details can be found in these two papers -

Novel small molecule alpha v integrin antagonists: comparative anti-cancer efficacy with known angiogenesis inhibitors.
Anticancer Res. 1999 Mar-Apr;19(2A):959-68.

Tumor biology: use of tiled images in conjunction with measurements of cellular proliferation and death in response to drug treatments.
Clin Cancer Res. 2000 Aug;6(8):3361-70.

The other project was to determine a compound potential for prevented infarct damage due to a stroke. In this case I developed and validated the images analysis methodology and gave feedback to those conducting the in vivo portion of the work. Here, Image Pro Plus 4.1 from Media Cybernetics was invaluable for successfully completing the research. I was able to determine a more objective method to measure the infarcted ares of the brain sections. Later I published my work and presented it in front of a neurobiology section at Experimental Biology in Orlando, 2000.

My publication, An objective procedure for ischemic area evaluation of the stroke intraluminal thread model in the mouse and rat. J Neurosci Methods. 2002 Jan 15;113(1):51-8, was published after I was laid off (details next entry).

So, I didn't have access to my own article unless I paid $30 to the publisher. I sure I will rant about the situation in scientific publishing later.

With Photoshop CS3 Extended I could conduct the same image processing and analysis that I did with Image Pro Plus. I hope to share that full workflow when I have it down on paper.

One more step

Imaging Support Grows with Opportunities

I feel I must continue to bring readers up to speed on my development of imaging skills. Mainly to encourage those who have an opportunity to see an unfilled niche and develop the resources to fill it. At this point in the mid 90's I was getting my M.B.A. and everything could be summed up in two words "create value."

My development of the Digital Darkroom capabilities expanded with brining in a flatbed scanner to digitize pictures previously made and the acquisition of additional imaging software, Adobe Photoshop Version 4 and Image Pro Plus from Media Cybernetics. The integration of image acquisition, processing, analysis, and output provided a complete package to aid myself and other scientists to conduct research.

As part of the general pharmacology group led by Andrew Slee and my immediate supervisor Janet Kerr I provided services to the different therapeutic groups of Dupont Merck. Two groups I concentrated on helping were oncology and cardiovascular disease. But over time I worked expanded to helping those working on inflammation, CNS, and anti-microbial therapies.




Creating Calibration Standards using Photoshop CS3 Extended

Creating calibrations to match the number of pixels with a know distance is easy with Photoshop CS3 Extended. A image that is parallel to the image sensor plane with a known distance is best to use. Most of my work concerns scanned materials or images from a microscope.

Open the image with the known distance

Pull down the Analysis Menu and select Set measurement scale > custom
This opens the measurement scale dialog window.

Move your cursor to the image. Photoshop has kindly and automatically given you the ruler tool.

Place one end on a edge of your known distance then move the other edge of the ruler tool to the other end of the known distance. Make sure that the point is located in the same way. If your first point is on the left side of a black line then the ending point should also be on the left side. To constrain the tool to 45 degree increments hold down the shift key.

With the ruler distance matching the known distance Photoshop reports the number of pixels in the length.

Fill in the actual known distance in the logical length field. If you know that the length you drew was 4 centimeters fill in the number 4.

Logical units is to set the name of the units of measurement. In this field enter in centimeter or an abbreviation like cm.

Finally save the calibration file you just made by clicking on the save preset button and giving a name to your calibration.

This calibration in now accessible under the set measurement scale call out to be applied on any other image you open in Photoshop.


Click image for better view.

The step after

Development of my Imaging Capabilities

Once the new imaging system was set up I started to develop my knowledge of digital imaging for science. I was surrounded by Phd's and peers with the need to incorporate imaging and image analysis techniques into their research.

The beginning of the digital dark room support at DuPont Merck Pharmaceuticals was an off shoot of my microscopy and histology skills. In order for me to convey what I saw through the scope I had relied on 35mm camera backs and the training I had at ATCC microscopy course taught by Mortimor Abramowitz.

The cost, time, and effort of film based images was the price my company had to pay until I went digital. The payback was immediate and rewarding. The epifluorescence work I was in the middle of conducting went from a turn around time of days and then sometime having to redo an entire experiment to being able to precisely visualize the optimum exposure and focus.




With turnaround time becoming instantaneous and the new ability to work with images directly on a monitor possibilities of using this resource to advance our pharmaceutical research grew.






Photoshop CS3 Extended

The new analysis menu item is only accessible if your have the Extended version of Photoshop CS3. Under this menu one can find calibration, data point selection, basic analysis tools, and the ability to automatically place a calibrated scale bar annotation in an image. Associated with the eyedropper tool are two analysis tools, the ruler tool and a new count tool.

Many researchers use Photoshop to optimize images for communication. Some, including myself, will incorporate Photoshop in their workflow to process images prior to analysis in dedicated image analysis software. My favorite is Image Pro Plus from Media Cybernetics. Now Photoshop analysis abilities will not supplant a dedicated program it can, for simpler analysis, process the entire workflow.

The next step

Let's get to business. I have used Photoshop in one version or another since 1995. Initially it was just to open images, tweak them for presentations or publications, then print them out. At first, my work was never more complex than adding annotations.



During this time I was researching proliferation of cancer cells in the mouse xenograft model. In order to investigate properly I was given the opportunity to build a imaging system. After thorough investigation I decided on a Olympus AX70 with full manual controls coupled to Imaging Research's MCID M2 turnkey system. My acquisition camera was a Sony DXC970 video camera(640x480). The strength of the system was the ability to create tiled images - full field of view of my specimens with high resolution. To properly convey my images Photoshop became a necessary tool to work with these large images. (up to 200mb in 1995)






The need preceded the learning. I started to see Photoshop as more than a communication tool. Its features and abilities would combine to provide additional research potential.







Photoshop CS3 Extended
Photoshop CS3 Extended now supports opening DICOM images without additional plug-ins. There are multiple ways to import images that are saved in this medical image file standard.

There are at least five ways to open DICOM files with PSP CS3 Extended:

1) Open individual images through the DICOM import.
2) Open a selected group of DICOM images via DICOM import.
3) Open the selected group as an image stack by selecting just the first of a set of DICOM images bringing them directly in as an animation.
4) Load multiple DICOM images script under the File>scripts.
5) Convert DICOM into JPEG and then bring them in as JPEG's (losing data information in the lower bit depth and lossy compression.

Every journey begins with a first step

Every blogger has a first post. This is mine. Over the next few entries I will cover some of the basics:


1) Who am I
2) What is the purpose of my blog
3) Why should you even care

The first personal goal is to contribute useful material everyday. The second is to improve my communication skills.

For my readers my goal is to transfer the knowledge of imaging in pharmaceutical and biological research. To give a leg up on understanding and using imaging as a tool to conduct successful research.

We all stand on the "shoulders of giants." I am no exception. Much of what I know was given to me by respected individuals freely sharing their knowledge and wisdom. I would like to be a conduit so we can see clearly over the crowds. Some the these shoulders I would like to thank are: Edward Tufte, John Russ, Chris Russ, Mortimer Abramowitz, and Ashley Manning Still to name but a few.

Sometimes the topic will be technical - usually concerning imaging in biomedicine. Other times it will be more philosophical look at life.

This blog will also support efforts to increase the proper use of Photoshop by biomedical clinicians and researchers.

My credentials briefly are:

Career
20 years Biological research in Chemical and Pharmaceutical industry

Education
B.A. in Biology (1988)
M.B.A. (1998)

Passion
Computer technology - First computer 1983 Atari 400, and many Macs and PCs since then
"computers are a tool not a religion"

Current Affiliation
Adobe's Biomedical Image Advisory Group - http://www.adobe.com/products/photoshop/photoshopextended/medical/
There you will find more information about the use of Photoshop in Biomedicine as well as individual success stories and useful white papers.

I welcome you to learn, apply, and teach the best practices you find in this blog.

Take and give care,

Eric J. Wexler