From the April 2022 issue of HealthCare Business News magazine
By Hila Goldman-Aslan
Shifting perception of AI within the medical imaging world
AI-based software has become prevalent in all healthcare imaging modalities. Reducing the subjectivity and variability associated with human interpretation, AI standardizes the image analysis process to support clinical decision-making in various healthcare environments.
Ad Statistics
Times Displayed: 19605
Times Visited: 366 Stay up to date with the latest training to fix, troubleshoot, and maintain your critical care devices. GE HealthCare offers multiple training formats to empower teams and expand knowledge, saving you time and money
Although previously hesitant, healthcare professionals around the globe have become increasingly accepting of AI-based imaging software. With the outbreak of COVID-19, hospitals experienced extreme resource shortages, creating an even greater urgency to address the rising need for quicker ultrasound analysis solutions.
The power of AI in medical imaging is that it serves the needs of healthcare teams, automating workflow. Increasing effectiveness and efficiency, AI saves physician analysis time, allowing for objective ongoing patient evaluation and follow-up.
AI is a learning software solution, trained to support different medical environments to provide analysis for numerous applications. Regarding cardiac ultrasounds in the echo lab and point-of-care environments, AI-based software uses the same technology for patient image evaluation and follow-up, limiting confusion and subjectivity among scans.
Meeting cardiac ultrasound imaging challenges
Automating strain analysis for echo patients across the board
AI-based tools can make the analysis process smarter, objective and more consistent, allowing medical workers to tend to larger scales of data, saving time, reducing error and allowing for more accurate data reporting. AI makes advanced analysis accessible to more patients, resulting in improved patient care.
Although continuously evolving, the majority of capturing and evaluating ultrasound images is still performed manually. Making the analysis difficult and error-prone, ultrasound analysis is highly dependent on the user’s experience.
At the echo lab, strain measurement is a key indicator for monitoring subclinical left ventricle (LV) dysfunction; crucial in cases of cardiotoxicity, chemotherapy patients, and follow-up after coronary events and presurgery for aortic stenosis valve replacement. When performed manually, strain is a time-consuming analysis that requires medical professionals to manually select the optimal clips out of 60-80 images, manually adjusting the strain measurements. Time consuming and cumbersome, strain is often not performed as needed and only for a selected number of patients. However, by automating the image selection and measurement processes, healthcare professionals receive the relevant ultrasound scans with the already analyzed strain measurements directly on their viewer for all patients in the echo lab.
