Learning Objective

1B2. Articulate characteristics of the information and communication technologies (e.g., infrastructure,servers,web services,storage) that support the healthcare environment.



Software is the term for the set of instructions that a computer uses to do some task. Writing the instructions for software (also called ‘programs’ or simply ‘code’) is called programming. Nearly everything that a digital device does is performed by executing software. For example:

  • Software converts mouse movements to cursor movements on your screen.
  • Software runs the browser that you are reading this on.
  • Web app software tells the browser software to print the words on the page.
  • Software runs Google Search

The following  is software called HTML used to make web pages.  This code makes a box pop up:

<button onclick="myFunction()">Whoomp</button>

function myFunction() {
var txt;
var r = confirm("There it is.");

Try it:

Some people are surprised by how often software code is controlling the everyday things they use.  If your bank card has a chip on it, there is a little software program in there.  Sometimes  the power button on the back of the computer ( a rocker switch) is a physical switch with no software. You flip the switch and power shuts off. However, the power button on the front of most computers is software driven. You press the button, and it decides if it is the right time to turn off or on based on its software instructions.  All of the kitchen appliances in our family home are now heavily computerized.

I'm not saying it's wrong, but software is even running some fancy toilets.

Does it make a difference if something is software controlled (i.e. digital) or not software controlled (i.e. analog)?  It certainly does make a difference and often that difference is very positive,  but sometimes there are trade-offs.  Your car would not be nearly as fuel efficient if a computer were not precisely controlling the amount of fuel, air, timing, temperature, and just about everything about it.  But that benefit comes with a cost: more complexity, more expense.  Similarly, the software in bed controls, medication pumps, and blood glucose meters all add some capability not possible without computing.

The engineers and programmers  who make digital devices and software know of many different types of software. However, for the rest of us there are only two main types of software: Operating Systems (aka “system software”) and Applications.

Operating Systems

Operating systems are used by the digital device to 1) create the user interface you are accustomed to, 2) coordinate all of the parts of the device such as storage, keyboards, cameras, etc. 3) Keep the computer secure from people and other software, and 4) ensure maintenance is done for efficient and safe use. Some people may not distinguish between the device and the operating system, when there is only one OS for that device. For example, the operating system on iOS is a required part of the Apple device. The navigation system in your car may also be ‘built in’ in a sense and not changeable. Mac computers are designed to only use macOS as an operating system. However, personal computers(e.g. non-mac laptops and desktops), can use different operating systems such as Windows 10 OS, Linux, Chromium OS, etc. depending on the users preference.

Windows and Mac operating systems

Every electronic device has an operating system. In a healthcare environment, medication pumps, Xray machines, security cameras, phones and pagers, and computer displays, all have operating systems. IT services and computer centers have many different specialized operating systems that need to be maintained by system administrators.


This amazing device has an operating system. I think it is the precursor to the A.I.pocket spouse.

Apps, -short for Applications, are a form of software program that users interact with. The operating system helps the Apps run correctly by giving the app a standardized way to use device resources. The most common “enterprise” apps (as opposed to personal apps) found in healthcare delivery organizations will be office productivity such as document editors and Email, and security apps such as antivirus. Health IT systems such as EHRs and radiology systems are all apps. Connected device “Smart Speakers” such as Amazon Alexa and Google Home have apps that perform various functions and can interact using voice. Websites are apps that run on computers web servers. You can click here to see the computer code that makes this page. If it doesn’t show up, try “view source ” in your browser menu.

This post-apocalyptic third-person-eater game is 1/95384th the size of Facebook in lines of code.

Programmers may write thousands of lines of source code for small simple programs and millions for larger more complicated programs. As an example, PacMan video game can be written in 650 lines of code. The average iPhone app is 50,000 lines of code. In comparison, Syphilis DNA has 1.1 million base pairs, which would be 16,500 lines long if written. Not to be outdone by Syphilis, Facebook uses 62 million lines of code to waste 41 minutes per day. Per American. Not to outdone by Facebook, Google‘s code ( core search+ google apps+ web services) tops out at more than 2 Billion lines of code.



A database is special kind of application that manages data for other applications. For example, a physician practice may have a Cerner Millennium EHR that stores all of the patient data in an Oracle Database version 19c. Specially skilled IT staff who maintain databases are called Database Administrators, or DBAs.

In case you are wondering,  DBAs  do realize that “database administrator”  is two words, -not three. However “DA” was already taken by another job they didn’t want to mess with.

There are a relatively small number of database vendors and products, and DBAs typically specialize in one vendor product. Therefore, efficiency would suggest that an IT department should consolidate on one database vendor product. Despite being ideal, this is rarely possible. Application vendors such as your EHR may require a different database than your human resources IT system requires, and since databases can be very specialized in one function such as warehousing and analytics, the best one (for you) for analytics may be a different vendor or product than the best one for EHR/operational apps.


While software is information, Hardware is the physical part of an electronic device. It should not require much practice to tell the difference between software and hardware: If it is a physical thing (at least somewhere), then it is hardware.


When there is confusion between the two, it is about storage media. Storage media is hardware that holds software.

The metaphor that generations of computer scientists want me to use: Paper is the storage medium, and the words on the paper are software.

USB ‘thumb’ drives and Blu Ray disks are common storage media that is portable. Non-portable storage media can be either internal to your device (Hard drives, RAM) or external to your device( Network Attached Storage NAS) -also known as portable cloud storage, and various backup devices.


The CPU, Central Processing Unit, or simply Processor, is the brain of the computer and controls all of the computers operations. The CPU interprets and executes the instructions given by a program. A CPUs ‘thinking’ is done by processing units called “cores”, that work independently of one another. More cores usually means more work can be done simultaneously. CPUs have a clock speed that determines how fast the cores can make calculations. For example, a CPU with a clock speed of 3.8 GHz can perform 3,800,000,000 clock cycles per second. The CPU does a little bit of work each clock cycle, and all those little bits add up fast. Faster clock speeds usually means work can be done more quickly, but not always. CPUs of different types are not always comparable using clock speed and cores. Some chip architectures are simply more efficient at doing certain types of work per clock cycle than others. When deciding on a CPU, use an objective comparison metric that is consistent with how you will use it (video processing, numeric calculations, etc.) CPU Boss is one of several sites that offer head to head comparisons of processor performance.

This thing is a CPU

Storage and Memory

If you ever hope to know more about computers than your grandparents, you should know there is an important distinction between storage and memory. Both memory and storage hold data. When it comes to electronic devices: “storage” is the hardware technology where data is saved. “Memory” is hardware technology where data is temporarily kept while being used by the processor. You might think of memory as the size of the device’s conscious. A phone or computer with 8GB of memory means that phone has room to hold 8GB of stuff open at the same time.

Think of storage as a library to save data long term. In order to use data from storage, the computer has to go get it from the library and put it into memory before it can be used.

Memory hardware (such as RAM) is often faster and more expensive than storage hardware(Such as disk based hard drives) . While it is possible for devices to use enough memory to forgo storage altogether, the high cost makes it impractical. Therefore, devices often use larger amounts of storage compared to memory. Storage is measured in bytes. Gigabytes (GB) are 1,000 bytes and Terabytes are 1,000,000,000,000 bytes. High definition movies take up about 5GB of your storage. However, you only need about 1 GB of memory to watch it.

Common Storage Devices

Hard drives store lots of data
  • Hard Drive Disk (HDD)
  • Solid State Drive (SSD)
  • Tape
  • Compact Disc (CD)
  • DVD and Blu-ray Discs
  • USB Flash Drive
  • Secure Digital Card (SD Card)
SD Card


Superfast computer CPUs needs fast access to lots of memory.  Most storage media technologies are relatively slow, but less expensive than memory technologies.  Typically, the faster the memory the more expensive it is.  That’s why computers use a tiered architecture for memory:  The fastest memory is used in small quantities close to the CPU, and slower memory is used in larger quantities a little farther away.

Level 1(L1) and Level(2) cache are your computers fastest memory and are located on the CPU itself.  You cannot upgrade or replace L1/L2 cache with replacing the CPU.

Random Access Memory (RAM) is a big workhorse.  RAM is relatively fast, yet affordable, and can be easily upgraded.   Desktop computers may need between 4GB – 32GB of RAM depending on their usage.  Applications that need to open entire files, rather than stream them, typically use more RAM.  For example, streaming (or watching) a video doesn’t require much RAM, but editing a video does.  Web browsers with lots of scripting and pages open, video games, and statistical applications will also consume significant RAM.

Virtual Memory

Macs and Windows PCs both use Virtual Memory.  Virtual memory is actually hard drive storage that is kept in reserve in case RAM is not enough.  Virtual memory is too slow to be a substitute for RAM, but it will keep your computer from crashing when the primary memory (RAM) is at capacity.

I am text block. Click edit button to change this text. Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

[table id=3 /]

For comparison:

  • The human genome as (GACT) text uses 750MB
  • A mammography study (single image) uses 45 MB

Peripheral Devices

Keyboards and Mouse are peripheral devices

A Peripheral Device is any device attached to a computer (or phone) that sends data into the computer, or brings data out of the computer (input devices and output devices) Peripheral devices can sometimes be inside the computer, but are typically external, or outside of the computer. A keyboard is a input peripheral device because it receives key-presses and sends them into the computer. A mouse is another input peripheral device. Printers and display monitors are examples of output peripheral devices because the computer uses them to communicate data outward.

The Otoscope is one of the most common medical peripheral devices

Healthcare peripheral devices are interesting and very important. The first peripheral devices a patient may notice are 1) a scanner for capturing ID and insurance card, 2) credit card reader , 3) touchscreen signature capture , and 4)perhaps a camera for taking the patient portrait. Some front desks may even use biometrics such as fingerprint scanners to identify staff or patients. In the clinic, vital signs devices, EKG, and stress test machines may also be configured as peripherals. There are many devices, including specialty devices such as an ophthalmology diagnostic device, that are peripherals to computers.

Common Peripheral Devices found in healthcare are:

  • Input
    • Keyboard
    • Computer mouse
    • Credit Card Reader
    • Graphic tablet
    • Touchscreen
    • Barcode reader
    • Image scanner
    • Microphone
    • Webcam/Digital Camera
    • Scanner/electronic fax
    • RFID or ID Reader
    • Biometric reader (e.g. palm, fingerprint, iris)
    • Medical device -specific reader
  • Output
    • Computer display
    • Printer
    • Projector
    • Speaker
Haptic devices are types of peripheral devices that add touch sensation to the user experience. For example, a mouse equipped with haptic feedback will allow the user to feel bumps and pressure through the mouse. Laparoscopic and dental training computers use force feedback to better simulate procedures. Haptic feedback is also key component to telesurgery (performing surgery from a distance) .


Networking is how devices communicate

Networking is the technology that allows computers and mobile devices to communicate. A network can be wireless, wired, or some of both. A network can be secure (private) or publicly accessible. A network can be very small (e.g. two devices) and local such as the wireless Bluetooth connection between a phone and a speaker used to play music, or the Internet, which is geographically large large network with millions of devices covering the planet.

No matter the size or structure, all networks use at least one of several rules called communication protocol standards. These standards specify how data should be formatted, addressed, sent, and confirmed delivered. Addressing systems identify the computers and devices, and are unique on any network. For example, the printer on your local network at home may have an internet protocol (IP) address of . When you print that funny cat picture, your computer (or mobile device) sends the picture to that specific address for printing. However, that address is only good within your home network. If you were on another network such as your work network, -or the Verizon mobile phone data network, your computer would be using a different list of addresses for everything and would not be able to see your printer.

Ethernet Cables are fast and secure. -But also awfully cable-ish

Wired Networks

Wired networks are still the fastest and considered the most secure types of networks. Ethernet cables are one type of wired network technology and is rated using different categories : Cat 1, Cat 2, on up. Cat 6 cable can transfer 10 Gb of data per second. Despite being fast, Ethernet cables are unpopular for personal computers and other small devices because they are, -well, because they are there. In fact, most personal devices are not sold with available Ethernet ports. Instead they use wireless network technology.

Wireless Networks

There are various wireless network technologies, with Wi-Fi being the most common family of standards for local wireless networking.

Wifi, also known as IEEE 802.11 standards(not really known) has improved in speed significantly since its release in 1997. Bluetooth is another wireless network protocol that is most often used for device-to-device file transfers and wireless audio.

ZigBee is a network mesh local area network (LAN) protocol. “Mesh” means devices use each other to efficiently pass along messages long distances. Instead of relying on strong antenna signals to reach the Wifi router, ZigBee devices only have to be powerful enough to send a message to the closest device, which then in turn passes it along to the next device. It was originally designed for building automation and control (e.g. thermostats and lighting systems).

3G, 4G, and 5G are versions (i.e. “Generations”) of mobile phone/device networks that use the IMT-2000 standards. 3G was the first high-speed network for phones and other devices using the network, but Tupac and Biggie were the original 2Gs.

5G is huge advance in throughput and may disrupt other industries by bringing broadband data speeds to nearly any device. For example, medical devices could skip the local wifi and transmit securely to the cloud. Or you can be your own personal broadband hotspot for your TV game console, etc. 5G has a ways to go, but is beginning to emerge in select cites and technologies.

Healthcare wireless networks are essential to providing and documenting modern care, however they introduce enormous challenges for support and security. Many networking technologies and standards were designed for consumer use and do not work well within the unusual hospital environment. For example, mobile phones and pager devices are essential tools for hospital staff, but some healthcare equipment can interfere with the signals they support (WiFi , 4G, etc.) MRI machines, linear accelerators, and the shielding materials used in construction can conflict with network signals. Solutions include strategically deploying Wifi access points considering the interference, supporting as many WiFi frequencies and standards as possible as they each have different characteristics, and deploying unusual rebroadcasting technologies that are not affected by the environment such as AM frequency signals and LTE (an older standard) for consumer mobile devices.

At the forefront of concern, is securing wireless networks and devices within healthcare. The use of wireless networks and devices inherently creates more access points that could be exploited by bad guys leading to compliance violations, device interruption, or worse. Therefore, wireless networks require changes to user training, IT security, and proper vendor-technology choices to minimize risk.

Broadband & Internet

Broadband is term for consumer and small business internet access that reflects a ‘faster speed’.  Although the FCC has recently defined broadband as being at least 25 Mbps download and 3 Mbps upload, the term is still loosely used.  Any of the five common consumer internet access technologies (DSL, Cable, Fiber, Fixed Wireless, Satellite) can theoretically reach broadband capabilities, but the actual observed speeds are much lower.

DSL – Uses plain copper telephone lines to carry internet signal.  It is cheap and often available in rural areas where other technologies are not.  However, it is slower in comparison and can be affected by older infrastructure.

Cable– is transmitted across coaxial cable TV lines, is priced compatible to fiber, and can support fast (e.g. 1000Mbps ) plans.   Cable is a legacy technology like DSL, and the basic infrastructure may not be improved.

Fixed Wireless – Fixed wireless is similar in many ways to mobile phone signals, with a few major differences.  Fixed wireless uses different frequencies from mobile phones, shorter distances covered, but much faster data bandwidth.  Fixed wireless is called ‘fixed’ because it uses fixed directional antennae for sender and receiver. Cost is reasonable, but speed can be impacted by terrain (hills, trees, etc.)

Fiber – Fiber cable is the king of network technology.   One tiny thread carries immense bandwidth at the speed of light.  Fiber is expensive, but the consumer market has brought the costs of some plans in line with cable.  Fiber is relatively fragile, so you wouldn’t use it to patch your XBox into the LAN.

Satellite – Satellite is a internet technology that is available almost everywhere, although terrain, weather, and trees can sometime block the signal.  Satellite is much more expensive, and at much slower speeds compared to other internet technologies.  However, for some remote health clinics, it is the only option.


Universal Broadband coverage in the US

[table id=4 /]

Business Class Internet

If you are familiar with the costs and speeds of residential consumer internet, you may be surprised at business class internet plans.  First off, there are multiple tiers of business class: Small business, -which uses technology similar to residential (Cable, Fiber, etc.) and higher tiers that require higher performance technologies.  Although small business class have plans with speeds comparable with residential plans:  the upload/download speeds are more balanced, and the cost is 2-4 times that of residential plans of the same speed.

Business Class Internet Service Comparison

  • DSL
    • Download Max:  50 Mbps (Theoretical claims up to 500Mbps)
    • Upload Max  8 Mbps
    • Price for Max: $300 per month
    • Uses phone lines, slow upload
  • Cable
    • Download Max: (Usually 500Mbps, some vendors e.g. Xfinity blend fiber network for higher speeds )
    • Upload Max: 50 Mbps
    • Price for Max: $500 per month
    • Susceptible to area congestion
  • Fiber
    • Download Max 1 Gbps
    • UploadMax: 880 Mbps
    • Prices for Max $650 per month
  • Satellite
    • Download max 100 Mbps
    • Upload max 4 Mbps
    • Prices for max $500 per month
    • Latency problems

The higher costs of business class internet service (compared to the same speed residential) is primarily due to two things:  1) Contracts and SLAs(Service Level Agreements) with more flexible terms and more reliability, and 2)Much better support for business class.  When residential internet service goes down, you can almost hear the customer service rep giggling in the background when they ask you to reboot the router.   (if there were any actual customer service reps) Business class support is much more responsive.

How Much Internet Access Do you Need?

The Federal Communications Commission (FCC) recommends a minimum of 4Mbps download per provider for office based providers.  On the other end of the spectrum, the FCC says a small hospital could get by with 100 Mbps download.

These figures, while described as minimum, may seem very low to residential users.  Keep in mind, this is the actual bandwidth needed, not the theoretical max of a plan followed with an asterix * .  Still, bandwidth needs add up quickly as you can see in the following list:

  1. EHR vendors such as eClinicalWorks and AthenaHealth  recommend dedicating about 1Mbps per provider (up AND down) just for the EHR.
  2. Telehealth apps (such as Zoom) also need about 2-3Mbps (up AND down)
  3. Heavy image use (see table below) with significantly increase bandwidth demand.
  4. Device generated data (either patient or provider) has the potential to significantly burden the network.  Examples include patient monitors, Devices such as medication pumps that report data, etc.
  5. To a lesser extent: patient surveys, administrative applications such as billing, email and other communications; health information exchange; and


Image Type Image Size (bits) Images
per Exam
Size of One Exam (MB)
Nuclear medicine 128 × 128 × 16 30–60 1–2
Magnetic resonance imaging 256 × 256 × 12 60 6
Ultrasound (color) 512 × 512 × 24 20–230 16–180
Digital angiography 512 × 512 × 8 15–40 4–10
Digitized electron microscopy 512 × 512 × 8 1 0.26
Digitized color microscopy 512 × 512 × 24 1 0.79
Computed tomography 512 × 512 × 12 40 20
Computed radiograph 2,048 × 2,048 × 12 2 16
Digitized X rays 2,048 × 2,048 × 12 2 16
Digitized mammography 4,096 × 5,625 × 16 4 184


Network Redundancy

Healthcare providers require a fast and reliable network between their computer and the EHR server.  For cloud (aka software as a service) EHRs, this means your internet service must be high quality. No internet service provider can guarantee 100% up-time.  This is why EHR vendors often require TWO different internet access services for each clinic or hospital.  This is called redundancy.  The result is a backup for when the primary internet access goes down.  It also costs twice as much as the already inflated costs of business internet.