A 10-year study of imaging volume demands on a PACS archive at the Medical University of South Carolina shows how significant a factor multislice CT has become in overall image data storage.
A 10-year study of imaging volume demands on a PACS archive at the Medical University of South Carolina shows how significant a factor multislice CT has become in overall image data storage.
At the beginning of the study CR was the major source of demand for the archive. With the advent of multislice CT in the late 90s, and especially the introduction of 64-slice and dual-source CT more recently, overall imaging volumes have begun to soar, and CT has become responsible for about half of the data volume, according to Eugene Mah, MSc, a researcher at MUSC.
Today, the facility routinely sees CT exams that generate nearly 10,000 slices of data, Mah said.
The study traced archive utilization by month and modality back to the introduction of MUSC's PACS in 1996. During that time the amount of data produced each month went from 48 GB to almost 1.2 TB, increasing by a factor of 25.
For most of the period examined, CR made up the bulk of the storage requirement. But beginning in 1999, CR's role began to decline and the role of CT began to increase. Today, CT storage accounts for 47.3%, or 557 GB/month. Next is CR at 18.4%, followed by MR and digital mammography, at 16.7% each.
Among the facility's CT scanners are two 64-slice units and one dual-source unit. The dual-source scanner generates even more data than the 64-slice units, Mah said.
Although the growth in CT data has been exponential, Mah said he expects it to level off as throughput capacity is reached and protocols normalized.
In the meantime, administrators who want to project archive needs can do so with some accuracy, Mah said.
They should:
CR, CT, MR, and digital mammography are the four modalities that will raise the greatest concern, he said.
Deep Learning Detection of Mammography Abnormalities: What a New Study Reveals
June 19th 2023In multiple mammography datasets with the original radiologist-detected abnormality removed, deep learning detection of breast cancer had an average area under the curve (AUC) of 87 percent and an accuracy rate of 83 percent, according to research presented at the recent Society for Imaging Informatics in Medicine (SIIM) conference.
Detecting Intracranial Hemorrhages: Can an Emerging AI Advance Have an Impact?
June 9th 2022A 3D whole brain convolutional neural network could provide enhanced sensitivity and specificity for diagnosing intracranial hemorrhages on computed tomography, according to new research presented at the Society for Imaging Informatics in Medicine (SIIM) conference in Kissimmee, Fla.