Chapter 6: Remote Image Review

Dr. Ilan Kolkowitz

As described in the preceding chapters, after the launch of an educational ultrasound program, ongoing image review is critical component of any educational program. Remote image review also offers a way for experts to help with clinical decision making from afar. While some image review can be done locally, often, expert educators have returned home thus creating a need for remote image review.  The following section describes considerations and possible solutions to make remote image review feasible.

Image considerations

Before sharing any images, it is important to consider what images are to be shared. This includes thinking about file size (clips versus still images), labeling, and patient information and identifiers (or lack thereof). Most will likely wish to review both still images and clips. File size will depend on the length of the clip. It may be best to set the clip length to 4 to 6 seconds as longer clips become very large and take longer to transfer.

 It is also important to consider patient health information (PHI) and de-identification. It is best practice to maintain HIPAA compliance with remote image review. An easy solution is to link a de-identified ID number to the ultrasound file instead of patient names. All image review should be password-protected. Discussing PHI safeguards with local partners is important to ensure that all institutional guidelines are being met.

Finally, it is important to consider the purpose of image review. Is it for quality feedback for educational purposes or for acute clinical questions. For education review, a system in which images are reviewed in batches once a week or once a month may work However, for acute clinical questions, a mechanism for immediate review is important. The techniques described below are not exhausted, but detail some examples of how ultrasound images can be shared and reviewed. 

Transmitting images

The first of the hurdles in remote image review is how to get the images themselves to those who need to see them. Several of these strategies will depend on the resources of the facility where you are working. In most cases, the biggest limiting factor is the local availability of reliable internet.

Most portable ultrasound machines are capable of uploading scans to a PACS system via WiFi. However, to set this up often requires dedicated IT, as well as a reliable WiFi connection at the site where the ultrasound machine is stored and, of course, a PACS system. Ultrasound file sizes vary, but POCUS studies are generally on the order of magnitude of 10MB per study and so should not take long, even at upload speeds of 1 Mbps.

Several newer ultrasound systems offer cloud-based platforms which can be used to share images and clips. These cloud-based platforms offer easy upload and sharing of images, without additional IT assistance, and can often be uploaded via cellular data as opposed to requiring a dedicated WiFi network (although this may incur added costs to the program or users). It should be noted that there may be concerns about data ownership with some these privately operated cloud platforms. In addition, some systems charge an annual fee for storage on their cloud platform, although scholarships for low-resource setting may be available.  See Chapter 2 for a table with cloud-storage costs.

Although it can be cumbersome, transferring the clip or still image to an external source, and then uploading studies at a more convenient time and location is another method when the cloud or WiFi is not immediately available. Some devices offer direct upload to a computer from a mobile device (phone or tablet). Alternatively, images can be downloaded from the machine onto a USB drive and then uploaded to a desired database at a convenient time. This strategy allows for upload  where and when there is internet readily available and allows you to store the ultrasound machine in a location that does not have WiFi. It also has the benefit of not requiring IT involvement. One major disadvantage of downloading studies onto a USB from the ultrasound machine is that it can take quite a long time per study. Designating a responsible individual or ultrasound champion to be in charge of regularly scheduled “data dumps” from the ultrasound machine onto a USB drive and then uploading the studies onto an image sharing platform is a good solution.

Sending physical USBs with scans via mail is also an option, although this would introduce a significant time lag between the studies being performed and feedback being received, as well as introduce the risk of the USB being lost in the mail. This is the least preferred option.

Sharing feedback

Once uploaded to a computer, one will need a means to share the images and a platform to give feedback. A variety of platforms are available for sharing clips uploaded from a USB drive. Many groups have been creative and incorporated several platforms together (for example, sharing images in Dropbox and then using a google docs form for image interpretation). Some cloud-storage platforms have built-in forms for image interpretation while others do not.

Dropbox offers an easy to use platform for the sharing of any type of file, including video files. However, it does not offer any specific organizational or feedback components tailored to ultrasound quality assurance. In addition, individual files will have to be uploaded and then opened, which may be somewhat clunky. As with all image transfer, personal health information may be at risk for identification as well, and a method to de-identify images should be employed (see below).

SonoClipShare developed by Dr. Ben Smith (sonoclipshare.com) is a free ultrasound image sharing platform that allows for uploading from a computer to a cloud, with a ‘comments’ interface designed for feedback and quality assurance. In addition, it allows images to be easily de-identified and shared in folders which is convenient for education purposes.

What’s App is another tool that is often used for quick feedback about ultrasound exams and relevant patient questions. Although it is not as easy to set up for a ‘data dump’ and large-volume review, it offers a quick way for fast feedback. It is also a nice way for an entire group to review or discuss a case. This may be most relevant for an acute clinically-based question and less for educational feedback. Some groups have combined What’s App with other storage systems, such as Google drive, to create a hybrid model for image review (see article by Fox et al). 

 Google docs is commonly used for image interpretation. Forms can be built that ask operators for their findings and interpretation with space reserved for expert review and feedback as well. A disadvantage of google docs is that it does not link directly to the images and a means for linking image interpretation to image is needed.

Image review

Several groups have had success using remote telemonitored ultrasound (example FaceTime, What’s App, or other video conferencing) which can be used to both interpret ultrasound images and to learners to obtain high quality images. Some devices even offer a built-in video conference program. For example, the Phillips Lumify system has an integrated tele-ultrasound program- Reacts- in which users can have live feedback on ultrasound studies done with the device. All of these methods for doing real-time image review require a robust enough internet or cellular connection for live video. Telemonitored ultrasound also requires careful coordination of time availability, which can be difficult across work schedules and time zones.

Although this section has focused on remote image review as it is most practical, in-person review allows for the hands-on training and real-time  feedback that most of us are most comfortable with from our own training experiences. It is also a great way to maintain project momentum and build relationships. Some strategies for success for in-person image review include: spreading out the responsibility for site visits amongst a group of qualified providers, planning ahead and making time for image-review, and providing a structured feed-back session after any clinical scanning. 

Take aways

  • Successful remote image review requires careful consideration of available resources including WiFi and internet capabilities, timeliness of feedback, and 
  • Consideration should be taken to protect patient health information in all encounters.
  • In-person image review is still extremely valuate and will support the relationship between the sites.

Resources

Create your website with WordPress.com
Get started
%d bloggers like this: