Stylus/tablet user input device for MRI heart wall segmentation: efficiency and ease of use

Area of Science:

• Medical Imaging
• Cardiology
• Image Analysis

Background:

• Cardiac magnetic resonance (CMR) imaging is crucial for cardiovascular assessment.
• Accurate segmentation of cardiac ventricles is essential for quantitative analysis.
• Traditional mouse-based segmentation can be time-consuming and lead to observer fatigue.

Purpose of the Study:

• To compare the efficiency and user experience of a stylus versus a mouse for manual segmentation in CMR.
• To evaluate the impact of different user input devices (UIDs) on segmentation time and observer variability.

Main Methods:

• Four observers independently segmented left and right ventricles in 25 clinical CMR datasets using both stylus and mouse UIDs.
• Segmentation time, intra-observer, and inter-observer variability were recorded and analyzed.
• Agreement between UIDs and observer preference were assessed using intraclass correlation coefficients (ICC), Bland-Altman plots, and questionnaires.

Main Results:

• Stylus-based segmentation was significantly faster than mouse-based segmentation for both left and right ventricles (p<0.001).
• Intra- and inter-observer agreement rates were excellent (ICC≥0.75) for both UIDs.
• Excellent agreement was observed between measurements derived from stylus and mouse segmentation, with observers preferring the stylus.

Conclusions:

• A stylus is a superior user input device for CMR ventricular segmentation, offering significant time savings.
• The stylus provides a more comfortable user experience, reducing fatigue and enhancing efficiency.
• Both stylus and mouse UIDs yield excellent agreement in segmentation accuracy, but the stylus offers practical advantages.