Data Set for Medical Imaging: Omurga and Pelvis Anatomy Auto Diagnostic AI Model

An important step was taken in artificial intelligence and medical imaging. The researchers published a specially designed data set to recognize complex variations in the spine anatomy: CTSpinoPelvic1K. This data set accommodates detailed anominations of the pelvis region, including ribs and femura.

The main problem address of the project is:

• Anatomical confusion in the passage area ofLumbosacral: The number of vertebra is made from C2, but the lumbar region will not be C2 when viewed alone and the count becomes clear to errors
• Lumbosacral transition variations (LSTV): Four ribs free, L1 omuru ribs can be interpreted as a shape or sheetrum melted L5 that carries ribs—this creates trouble when putting uncertainty diagnosis -Multi organ coverage: The data set is the first comprehensive collection of spine, pelvis, ribs and femur head in the same coordinate frame -1000 CT image: A wide set of anominations including Abdominopelvic CT scans, provide enough data for the training of machine learning models
• Open access: This resource offered to the entire research community creates the opportunity to develop independent verification and algorithm

This data set will play a key role in developing automated diagnostic systems especially for patients with anatomical variation. Reducing spinal counting errors will significantly increase surgical planning and diagnostic accuracy. Presenting as open source creates a valuable opportunity for the algorithm development and testing of researchers in different institutions.