Is AI better than humans at segmenting feline intestines?

VRU 64(1): 131-139

Author

Ryan Appleby
May 17, 2023

This study evaluates inter- and intraobserver repeatability and agreement in measuring intestinal wall layer thicknesses and segmentation of transverse sections of small intestines in ultrasound images of cats.

Results: Segmentations of small intestines have higher interobserver agreement than measurements of intestinal wall thicknesses.

Conclusion: Automated machine learning approaches could be used in ultrasound images of feline intestines to classify normal from abnormal conditions, such as IBD or lymphoma, in lieu of traditional thickness measurements.

Ultrasound image for cat with SCEL, illustrating examples of segmentations. Technical parameters were transverse sections obtained via B-mode from a digital linear array at 75 Hz, resulting in an 8-bit image with dimensions of 768×1024 pixels (see Supporting Information 1 and 2 for details). The left panel displays the original image with the transverse section of interest to segment shown by the yellow arrow, the bottom panels display four different segmentations from one observer (solid cyan, green, blue, and yellow lines), and the right panel displays the four segments superimposed with the average segmentation, shown in red dashed line

How did we do?

Login or Subscribe to participate in polls.

Disclaimer: The summary generated in this email was created by an AI large language model. Therefore errors may occur. Reading the article is the best way to understand the scholarly work. The figure presented here remains the property of the publisher or author and subject to the applicable copyright agreement. It is reproduced here as an educational work. If you have any questions or concerns about the work presented here, reply to this email.