Handheld 3D Ophthalmic Ultrasound Tomography System

Abstract

Purpose : Scarcity of ophthalmic ultrasonographers necessitate a standardized ophthalmic ultrasound imaging technique independent of user training. 3D ophthalmic ultrasound can offer a clinically practical imaging method to evaluate gross anatomy and pathology. We demonstrate our handheld 3D printed device and algorithm to generate 3-dimensional (3D) tomographic models of the posterior pole.

Methods : A novel handheld device was designed and 3D printed to mechanically move a portable Butterfly iQ ultrasound in an arc across a human globe. The probe head was set as the pivot point where the device moved along an angular sweep of 24 degrees. Sequential ultrasound b-scan images were extracted from a recorded video at a rate of 30 frames-per-second for a total of 0.67 degrees per ultrasound image slice. A gaussian filter was used to maximize image contrast and then the data between ultrasound image slices was interpolated and integrated using our algorithm developed in Matlab to create a solid body 3D model. The 3D models were segmented and exported for qualitative analysis using ITK-SNAP – open-source 3D medical imaging software.

Results : The 3D printed handheld ultrasound rotational device safely and reliably acquired standardized ophthalmic ultrasound images. Our developed algorithm was able to create a 3D model of the globe, which could delineate the boundaries between vitreous, retina-choroid complex, and optic nerve.

Conclusions : We demonstrated our 3D printed handheld ophthalmic ultrasound tomography acquisition device and modeling algorithm. We were able to safely and effectively acquire sequential ultrasound b-scans of the retina-choroid complex and assembled them into a 3D model.

This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.

 

Figure 1. A) Completely assembled handheld 3D ultrasound acquisition device. The 3D printed device rotates a Butterfly iQ ultrasound probe across a human subject’s eye and records the images using Butterfly Inc’s mobile application on iPad or smartphone. B) Demonstration of the handheld acquisition device over a human subject’s eye during ultrasound imaging.

 

Figure 2. A) Raw ultrasound b-scan of a human subject eye before image processing. B) Reconstructed 3D model using a developed algorithm to process acquired ultrasound b-scan images. Data between b-scan frames are interpolated and integrate to form a solid body 3D model of the retina-choroid complex.