Abstract
Our aim is to demonstrate a new semi-automated method for measuring periorgan hematoma due to abdominal organ trauma. The new methods' performance was evaluated on randomly selected 31 CT scans from 31 patients (10 females, 21 males, median age:39,5) with traumatic intraabdominal periorgan hematoma. Twenty seven patients had perisplenic, eleven had perirenal and two had perihepatic hematomas. Only seven patients had more than one periorgan hematoma. A new semi-automated segmentation tool ManSeg 2.1 was used by a radiologist for measuring hematoma volumes and the results were compared with conventional method by another blinded radiologist. Statistical analyses were performed by using Student's t test. P <0.05 denoted statistical significance. Sensitivity, specificity, and positive and negative predictive values of the system was managed and compared to that of two radiologists. Measurement of hematoma volume was feasible in all cases. All hematomas were large and measured indirectly. Measurement of periorgan intraabdominal hematoma volume with semiautomated segmentation tool ManSeg 2.1 present us a reliable option comparing with conventional method.
Translated title of the contribution | Semi-automated measurement of intraabdominal periorgan hematoma with a new method in patients with blunt abdominal trauma |
---|---|
Original language | Turkish |
Title of host publication | 2015 Medical Technologies National Conference, TIPTEKNO 2015 |
Publisher | Institute of Electrical and Electronics Engineers Inc. |
ISBN (Print) | 9781467377652 |
DOIs | |
State | Published - Jan 6 2016 |
Externally published | Yes |
Event | Medical Technologies National Conference, TIPTEKNO 2015 - Bodrum, Turkey Duration: Oct 15 2015 → Oct 18 2015 |
Other
Other | |
---|---|
Country/Territory | Turkey |
City | Bodrum |
Period | 10/15/15 → 10/18/15 |
Keywords
- hematoma volume
- semi-automated
- trauma
ASJC Scopus subject areas
- Biotechnology
- Artificial Intelligence
- Computer Science Applications
- Signal Processing
- Biomedical Engineering
- Health Informatics