Page Menu

Programmable Mouse Design

Programmable Mouse: Quality Improvement Project

Andrew Chen, MD

BACKGROUND

Dictaphone Keyboard Mouse Daily radiology practice centers around 3 devices: the Dictaphone, the keyboard, and the mouse. In our department, these have not changed in the last 15 years, and most PACS computers have the original devices.

Moreover, increasing case volumes and number of images per case contribute to reader fatigue and increase risk for medical errors.

PURPOSE

Identify commercially available devices to increase radiologists’ productivity and reduce reader fatigue as part of a practice quality improvement (PQI) project.

METHODOLOGY

As part of an ongoing project over the past 6 years, more than 20 input devices were purchased to replace the traditional hardware of the keyboard, mouse, and microphone. These devices included 3D mice, wired and wireless mice, tablets and touch pads, and programmable boards produced by companies such as 3Dconnexion®, Logitech®, Razer®, Wacom®, and IBM®.

Essential features for the device

Compatible with PACS system (Windows platform)
Commercially available
Simple to use
Easily programmable

Products tested

Chosen device	Why Logitech® G700/G700s was chosen
For the pilot study, we aimed to test the efficacy of at least 1 device We chose the Logitech G700/G700s for the initial project	Has previously identified essential features PLUS: On-board memory (software installation not required on the PAC systems) Low cost Flywheel and toggle scroll (for large image stacks) Programmable for each radiologist’s needs 3-5 days of battery life with single charge based on radiologist’s typical usage

Programming

The mouse was programmed to perform several common tasks.
A sample of the programmable options is shown in image below.
Learn how to program the mouse. Download PDF

STUDY DESIGN

Participants

Radiologists from the following divisions:

Neuroradiology
Pediatric
Emergency
Cardiothoracic
Abdominal
MSK

Phases

Phase 1: Safety

Early adopter participants bought and programmed their own mice
No safety concerns were identified

Phase 2: Testing

Participants were assigned 3 tasks using either a traditional mouse or the Logitech mouse:
- Scrolling through an image stack
- Spine labeling
- Switching images
Tasks were demonstrated to be statistically faster with the Logitech mouse

Andrew Chen, MD - Programmable Mouse - Scroll Wheel

Andrew Chen, MD - Programmable Mouse - Switching Images

Andrew Chen, MD - Programmable Mouse - Spine Labeling

Phase 3: Productivity

Volunteer radiology attendings were evaluated on productivity as measured by RVUs
- RVUs were assessed on 10 separate days before participants were introduced to the new mouse and on 10 days after they had become accustomed to it.
- To reduce bias, volunteers were informed before the study that their productivity would be assessed but the methodology was not disclosed.
- 2 hr blocks were assessed on each of those 10 days for RVU productivity
  - The 2 hr block was selected based on maximal productivity during each test day.
  - During this block, attendings dictated cases themselves, not through a resident.
  - RVUs rose 11-25% per volunteer that successfully learned the mouse

Survey

Volunteers then completed a survey about their experience:

100% of subjects would recommend the mouse to others
75% of participants think the mouse improved their diagnostic ability.
100% would be willing to participate in future hardware trials.

Of note, 5 of 17 attendings failed to incorporate the mouse into their practice and withdrew from the project.

DISCUSSION

Hardware devices utilized in academic radiology practice have minimally improved

Radiologists need ways to increase their efficiency as image volumes increase.
Few studies have assessed hardware updates as a way to improve productivity. Mammography is one subspecialty that has taken lead.

This study shows that upgrading technology is a simple, low cost way to increase productivity among diagnostic radiologists.

In a standard workday, an 11-25% increase in productivity translates into ~1-2 hours to do more research or teaching, read more cases, or focus more on diagnostic portion of the case and less on the technical side.
Among the functions programmed into buttons, participants particularly enjoyed:
- Initiate dictation
- Ruler
- Toggle scroll: Allows user to toggle between traditional scroll wheel and fly-wheel mode for large image stacks (e.g. CTA with 1000+ images).

Issues raised during the study

In Phase 3 of the study, 5 of 17 attendings failed to incorporate the mouse.
Most common factor hindering adoption of the mouse: Steep learning curve.
- Inertia: “This is how it’s always been done.”
- Relative lack of incentive to increase productivity in academic radiology.
The failure rate was more evident in the senior faculty.
Perhaps considering the Logitech mouse an industrial mouse, not a gaming mouse may increase the willingness of some to incorporate it in practice.

Study design limitations

Participants could not be blinded to the mouse they were using or to the cases
- Ideal study design: Test each radiologist with the older mouse; repeat same cases with the newer mouse.
- This is not realistic: Participants would already know the results from the initial radiologic cases.
Control and experimental arms were not possible because no two radiologists read at the same pace.
Daily/seasonal fluctuations in case volume
- This prevented testing a specific time period (days-months) with and without the new mouse.
- To avoid the issue of seasonal volume fluctuation, ten 2-hour blocks were chosen at random to assess productivity for each radiologist
  - Radiologists were assigned to their respective subspecialty responsibility (e.g. the chest radiologist read chest x-rays).
  - We reviewed each of those 10 days for a 2 hr block in which the radiologist was most productive (measured in RVUs).
  - We recognize that RVUs may not be the ideal measurement of productivity, but we utilize this to maintain consistency.
  - In an academic environment with inherent interruptions from conferences, consultations, and teaching responsibilities, we chose a 2 hr block assuming this is a reasonably reproducible time period where the academic radiologist is reading with little interruption.

This pilot shows that for little cost, a commercially available product can significantly increase radiologists’ productivity, reducing fatigue, improving quality, and potentially reducing medical error.

Trainees are highly encouraged to explore new hardware devices as in this project. We believe this is a practical skill that will inevitably find its place in daily radiology practice. Please email me at Andrew.Chen@umassmemorial.org for comments or questions.

Andrew Chen, MD has no relevant financial or non-financial relationships to disclose.