Robotic Ultrasonography for Autonomous Non-Invasive Diagnosis - A Systematic Literature Review

This systematic review was conducted to synthesize and critically appraise the current evidence on the automation of non-invasive diagnostic ultrasonography using robots. Medline, Scopus, EMBASE, Web of Science, and PubMed databases were searched using a defined search strategy. Eighteen studies were included for consideration after reviewing 1,932 studies. Through appraisal, the selected studies were found to be of high quality. An automated ultrasound examination can be divided into two stages. The first stage involves localization of the region of interest and moving the probe close to that region, while the second stage involves manipulating the probe configuration around the region of interest to scan and acquire an optimized image. Although the second stage was fully automated in all the analyzed systems, the first initialization stage requires full or partial assistance from the sonographer in 11 of the systems. Future studies should propose novel methods to perform these procedures without the sonographer's aid. Fourteen studies used a combination of visual and force feedback to control the scanning and optimization. The utilization of ultrasound image feedback for probe manipulation during this process has the potential to improve the accuracy and robustness of the system.