Sunday, August 26, 2012

Solid State Drives for a “Solid” Computing Experience

It does not take much effort to hear an all too common occurrence in hospitals that “the computers here are so slow.” But why are they slow? Every upgrade cycle means lots of new machines – more expensive machines with more RAM, faster processors and bigger hard drives. That should be enough, right? Or so, many people think, because there is always one component that seems to be neglected - the hard drive. And there is one technology that still does not receive enough attention in the medical realm - the solid state drive (otherwise known as an “SSD” and sometimes referred to as “flash memory” – although it is very different than a typical USB flash drive). We feel that not enough people realize that their computers are “held back” by their hard drives. Because traditional hard disk drives (HDD) contain a spinning disk, they are relatively slow at transferring data. Even with hard drives advertising data transfer at 100 megabytes per second (MB/s), the real-world performance, which involves writing lots of small files (also known as the ‘4 kilobyte (KB) random read/write time’) is notoriously slow – less than 1 MB/s! Furthermore, file fragmentation slows PCs with HDDs considerably; PCs become virtually useless during the defragmentation process that needs to be performed periodically. On the contrary, SSDs’ performance is not impacted by fragmentation, therefore there is no need to perform defragmentation. Various review sites such as have compared the performance of systems running on computers with HDDs to similarly-configured models with SSDs, like this example using PCMark Vantage. The results are staggering, considering that the PCs configuration and the utilized software in this comparison were identical – the only difference being the hard disk drive swapped for a solid state drive. Aside from benchmark tests, what about real world day-to-day performance? From our exprience, it is much faster and smoother working with an SSD, which contains only memory chips and no moving parts. Have you ever waited for minutes for your computer to reboot? That time can be less than 15 seconds, or perhaps faster than Usain Bolt’s 100-meter dash in the 2012 Olympics. Does Microsoft Word, Final Cut Pro or Adobe Photoshop take too long to load? Application load times are shorter when using an SSD (as in this report with Adobe Photoshop), and whether you are using Microsoft Windows, Mac OS X or Linux, the overall boost in operating system performance becomes immediately apparent, particularly when multitasking. Furthermore, many notebooks now feature “instant resume,” allowing a computer to wake up immediately from sleep. This is not a feature restricted just to the new ThinkPads or MacBooks, but is also possible with many notebooks or even desktops – both new and old – when using an SSD, since the SSDs do not need to “spin up” like HDDs to begin reading data. The choice of operating system also affects the time required for systems to resume from sleep; for instance, Windows 7 on a notebook generally allows it to resume more quickly than Windows XP, but there are other hardware and software variables that can affect this. Of note, there are several arguments against the usefulness of SSDs in place of hard disk drives; one being that “not all SSDs are created equal”. As stated earlier, an important performance determinant is not the advertised sequential maximum read and write speeds, but rather, the random input and output operations per second (IOPS), particularly for small files. Solid state disks with slow random read/write speeds may perform no better (or in some cases, worse than) traditional hard drives. This comparison demonstrates an example of the wide spectrum of performance between various SSD models. Furthermore, long term reliability of SSDs may vary dramatically between manufacturers; while Intel, Crucial Memory and Samsung have very low reported failure rates, users have reported higher failure rates with select models from other companies. The most notable disadvantage of SSDs is increased cost-to-space ratio, particularly when used as server storage. Given the vast amount of healthcare data (and in particular for pathology, large images and virtual slide files), SSDs might still not be economically feasible for large-scale storage on a tight budget. However, the market is changing, and the price gap between SSDs and HDDs is narrowing. Five years ago, a paltry 32 GB SSD would cost over $200, whereas one could buy over a terabyte of hard drive space for that amount. Currently, for consumers, a 256 GB version of Samsung’s highly-praised 830 series SSD can be had for under $200. Pair this with a $400 notebook, and one could have a machine with markedly faster day-to-day performance than a $1,200 machine with twice as much RAM and a beefier processor. Moreover, SSDs are becoming much cheaper than, say, buying a new computer or tablet. And there are additional benefits that may make them more cost effective in the long run, such as lower energy usage, greater durability due to the lack of moving parts, and lower heat dissipation (important for server storage). In healthcare, the decision to use older hardware with SSD technology during the next upgrade cycle has probably never been considered. But, given how well technology from last year can perform with current generation SSDs, such a decision may become crucial in saving both time and money. Other options include using an SSD for the operating system and programs, coupled with a separate hard disk drive for data, to obtain “the best of both worlds” of speed and storage space. Are you reporting the word “slow” too many times to your PC support team? If so, then perhaps now is the time for another paradigm shift in healthcare IT, to understand the benefits that SSD technology can provide in a field already fraught with time constraints. As prices continue to drop, we hope that this technology will be supported by future healthcare professionals. By Milon Amin & Ioan Cucoranu

Friday, July 29, 2011

Tipping Points in Digital Pathology

Social consensus through the influence of committed minorities” (DOI: 10.1103/PhysRevE.84.011130) is a fascinating read. In it, Xie et al. of the Rensselaer Polytechnic Institute take the well-known phenomenon of an inflection point in thinking – a tipping point past which a minority opinion will become the majority opinion – and use computational and analytic simulations to try to determine what that point might be. Their conclusion? That when the number of “true believers” reaches 10% of a given population, conversion of the rest of the population becomes inevitable. Below 10%, it would take far longer for such a conversion to take place.

Historically, there is some evidence to back this up: the article cites how the civil rights movement, for instance, only came into full fruition “shortly after the size of the African-American population crossed the 10% mark”. From our own history, we know that Joseph Lister and Ignaz Semmelweis, though they were correct about the important role of hand-washing in patient care, were not able to convince the majority of their peers on this matter. It took the work of Louis Pasteur in confirming the germ theory to sway the majority professional opinion, years later. More recently, when Barry Marshall and Robin Warren first discovered that gastritis and peptic ulcers can be caused by Helicobacter pylori, they were initially met with a great deal of skepticism; it was not until later that their discovery would be generally accepted as medical fact.

This article does have a few limitations – first of all, it utilizes a computational sociology model known as the Naming Game . In Xie et al.’s implementation of this model, “at each simulation time step, a randomly chosen speaker voices a random opinion from his list to a randomly chosen neighbor, designated the listener. If the listener has the spoken opinion in his list, both speaker and listener retain only that opinion, else the listener adds the spoken opinion to his list”. While this model has its advantages – computational simplicity among them – it has some drawbacks:

  • It assumes that all interaction is strictly held in pairs.
  • It does not attempt to place a hierarchy of trustworthiness on the population.
  • It does not take into account the feasibility of an idea being expressed.

Compare this to the real world, in which:

  • A large amount of interaction is not pair-based.
  • Opinions carry different weight depending on who is articulating them.
  • A given population generally has an idea of what is feasible and what is not.

Even with its flaws, this paper does make me ponder on our own field of pathology informatics. As we all know, ask ten different practicing pathologists about how technologies like whole-slide imaging will affect our profession, and you’ll get ten different answers. I just did a totally unscientific survey of residents (n = 7) and attending pathologists (n = 9); 5 of 7 residents believed that WSI is the way pathology will be practiced in the near future, as opposed to 1 in 9 pathologists. I wonder what a survey of a larger population of both trainee and attending pathologists would show.

How close are we to our digital pathology tipping point?

Tuesday, January 18, 2011

Web Presence of State Pathology Societies

Professional organizations like state medical associations (SMA) and state pathology societies (SPS) play an important role in organizing and representing Pathologists at a state and national level. Often started as a small social group, they grow into large organizations that perform advocacy on behalf of pathologists, are called upon to give opinions on medical matters and offer Continuing Medical Education (CME) courses to their members.

Often led by dynamic pathologist-leaders, these organizations serve as an excellent recruitment and talent pool. Needless to say, the success of a particular SPS heavily hinges on its member enrollment and the benefits of its membership. It is felt that young pathologists join SPS as trainees, but later become inactive citing diminishing returns.

Today, SPS have started employing web-based solutions to spread their reach, connect their members online, provide membership and meeting information, publish newsletters and by-laws and provide educational material. Across the USA, the extent of this adoption and online services offered is very heterogenous. We surveyed several SPS listed on the College of American Pathologists website and evaluated each web site.

We found that every state in the USA has a formal SPS, but that only 50 % (25/50) of these societies have a web presence. Of these 25 web sites, 84% (21/25) are hosted on a domain registered in the name of the SPS, while 16% (4/25) were part of a SMA. Of note, 76% (19/25) had a newsletter of which only 44% (11/25) had published their current (2010) edition. We found that 76% (19/25) had information regarding meetings online, of which 72% (18/25) pertained to a recent or future meeting. Furthermore, 60% (15/10) had posted their by-laws and 32% (8/25) allowed online membership enrollment, including online forms for 36% (9/25) societies. Finally, 32% (8/25) allowed for dues to be paid online and only 8% (2/25) had some form of active case archives for member education.

Our conclusion is that SPS have not fully utilized the Internet and Web 2.0. Their web sites often provide outdated and limited information. This is a lost opportunity. Perhaps a national effort is needed to standardize and upgrade SPS websites. Such an effort would likely repay itself in the long term through increased member enrollment, improved benefits to members, and better coordination of advocacy activities across states.

Submitted by: Gaurav Sharma (

Monday, October 18, 2010

Screens for Pathologists: Portrait vs. Landscape

With the adoption of laboratory information systems and digital pathology, pathologists are increasingly reviewing and signing out pathology reports electronically. However, the majority of the pathologists still wish to review a draft or a final copy of their work on paper before sign-out. One of the major reasons cited for this duplication is the apparent ease of readability of a piece of paper compared to a screen, as an electronic page must be scrolled up and down to review all the information before sign-out.

Considering that the final output of a pathology report is an electronic representation that gets displayed in the electronic medical record or printed out on a portrait (taller than wider) orientation, it is surprising to observe that we are using landscape and now widescreen (i.e. wider than tall) monitors to draft and read pathology reports. There are several reasons for this mismatch (historical as well as lack of thought on workstation design) and little research has been done to see if the readability of electronic display can be improved.

For humans, it is easier to track objects moving sideways since our field of vision is more attuned to a horizontal spread, enhanced by an artificial reference or horizon (imagine watching horses galloping in a grassland at sunset). However, humans have historically preferred reading text arranged in small columns that wrap the text to the next line (reading a newspaper with text neatly arranged in short lines and long columns), and this is reflected in the almost universal adoption of a portrait format for forms used for textual information (from stone tablets to the modern A4 page).
In the 1970s, the modern computer monitor was derived from a modified television screen (a landscape display). We are now straddled with a convention wherein we are using an unsuitable display format for reading and writing large amounts of electronic textual information. This mismatch may be attributed to the preferences of early programmers who preferred to have more space for long lines of codes (landscape) and an absence of thoughts on design elements of a textual display device.
As a result, pathologists are now scrolling up and down to read different parts of an electronic pathology report and entering data into text fields without having the benefit of the viewing the complete page. Of course, we have the option of zooming out, but on a landscape monitor this renders text unreadable and just gives the rough outline of the document. This frustration often translates into several mouse-clicks and/or making a hard copy (duplication of work and not desirable in a green environment).
The manufacturers of common e-reading devices (e.g. Amazon Kindle, I-Pad and I-Phone) have adopted portrait as their default mode. It is easier to keep a mobile phone or e-reader upright and read down a column versus rotating it 90 degrees and then read text by moving across the screen. This reproduction of a ‘book’ format was hailed as a milestone in the e-publishing industry, but has been completely missed by the conventional computer users who type and read text on home and office computer screens.
Ironically, this mismatch can easily be addressed by rotating your screen 90 degrees, hitting ‘ctrl’+’alt’, and the right arrow key to adjust the video output. This converts the screen to a more logical portrait view on which a report can be entirely represented, text is easier to read and no visual ‘real estate’ is lost on the sides. A quick 5 minute trial (and allowing for a minute to adjust to new orientation) demonstrates the easy readability of a MS Word document, PDFs of medical literature and e-mails. For the surgical pathologist, it is easier to review the patient name, accession number, clinical history etc. while typing in the final diagnosis and reviewing the gross findings with this portrait display. The potential improvement in reading portrait electronic displays not only makes ergonomic sense, but could translate into some benefit in patient safety.  

Submitted by: Gaurav Sharma (

Friday, October 15, 2010

If LIS companies sold Cars

 A. Deciding What To Buy

One fine day you realize the benefits of owning a car. It seems clear that this car will ‘revolutionize’ the way you commute. Therefore, you decide to call the the nearest car dealer. The car dealer asks for a detailed statement regarding your commuting needs and features that you wish to have in your car. Any omitted feature (eg. a steering wheel or breaks) will later come as an ‘added feature’ at ‘a nominal extra cost’. After several calls and visits from their sales team you finally get to sit down with the dealer's chief designer. By the way, a break light can be painted green and indicator lights can be magenta. It does not seem to matter, since every other car on the road was designed by its own user with the dealer's designer.
B. Signing On The Dotted Line
On the big day your contract is presented to you. Per the contract, this car will have a limited warranty and a requirement to upgrade it with a newer model every 3 years, because the manufacturer may not be willing to support your old legacy car. Also, you need to buy a license to drive “X” number of people at a time in this car. If you want to take your son’s friend (X+1) to a football game, you will have to either apply for an extra user license or make extra trips. Now that you own your car, you need to maintain a team of ‘car specialists’ to support it 24x7x365. You need to pay the dealer a ‘small orientation fee’ to train your people to manage your car. 
C. Driving Your New Car
After several mysterious updates the mechanics living in your backyard carefully assemble your car. A fight breaks out between the the dealer's mechanics (who want to go home) and your own recently hired car specialists  (who need the dealer's mechanics to show them the ropes). Finally, they hand you the keys to your car. Before the mechanics leave, however, a team of middleware managers arrive at your house and promise to fix any of your car's shortcomings, simply by adding some of their own parts. On the day that you decide to test drive your new car, you find out that you can only drive it on the roads maintained by your city’s infrastructure (IT) department. Going to another city will cost you extra at each ‘interface’ for that city en route. Before you decide to buy a GPS, you will need to first check if your car can support this ‘extra hardware’ and if it needs to be ‘custom programmed’. Without that GPS, it looks like you will never know where you are in cyberspace. 
D. In The End
You resign to the fate of owning an expensive car that is slow to drive, limited to a single city, needs a lot of maintenance and is frequently down for repairs. You seldom (if ever) travel over 10 mph, waiting for the day your city infrastructure department upgrades their dirt tracks to asphalt roads. Your budget is perpetually threatened by a nearing upgrade, or the possibility of having to buy a new car because the manufacturer went out of business. You realize that before the car, you took a reliable bus to work. Now you drive (under 10 mph) to work to make enough money to afford a car, a team of your own specialists, and some middleware guys. Perhaps this was not the ‘commuting revolution’ that you had hoped for.
Submitted by: Gaurav Sharma (

Monday, September 20, 2010

Welcome to The Journal of Pathology Informatics (JPI)

The Journal of Pathology Informatics (JPI) is an open access peer-reviewed journal dedicated to the advancement of pathology informatics. This is the official journal of the Association for Pathology Informatics (API). The journal aims to publish broadly about pathology informatics and freely disseminate all articles worldwide. This journal is of interest to pathologists, informaticians, academics, researchers, health IT specialists, information officers, IT staff, vendors, and anyone with an interest in informatics. We encourage submissions from anyone with an interest in the field of pathology informatics. We publish all types of papers related to pathology informatics including original research articles, technical notes, reviews, viewpoints, commentaries, editorials, book reviews, and correspondence to the editors. All submissions are subject to peer review by the well-regarded editorial board and by expert referees in appropriate specialties.

Bibliographic listings:

The journal is indexed with Caspur, DOAJ, EBSCO Publishing’s Electronic Databases, Expanded Academic ASAP, Genamics JournalSeek, Google Scholar, Hinari, Index Copernicus, OpenJGate, PrimoCentral, ProQuest, PubMed, Pubmed Central, SCOLOAR, SIIC databases, Summon by Serial Solutions, Ulrich’s International Periodical Directory