By Greg Freiherr, Editor, gfreiherr@cmp.comA short time back, my mother gave me an old photo album. There I was, an infant in one picture, on my grandmother's lap, looking as
By Greg Freiherr, Editor, gfreiherr@cmp.com
A short time back, my mother gave me an old photo album. There I was, an infant in one picture, on my grandmother's lap, looking as though I'd figured out a secret to the universe. Any hopes of intellectual greatness at an early age, however, were dashed a couple pages later by a picture of me trying to balance a plastic bowling pin on my head.
These were moments I thought I'd lost. Finding them got me thinking about what might be lost and found in the future of medical imaging. For much of the first few decades of PACS, I heard concerns about the longevity of medical images and other patient data. Will they last long enough? Will they be corrupted by time? Now, looking at this photo album I thought was lost, I began to wonder if there might be another side of this. What if these images and other data last longer than we want?
The technology sector is only beginning to recover from the boom and consequent bust of the dot-coms that went under. The first few years after the crash were characterized by plummeting sales of servers and archives whose prices turned to mush in the wake of a secondary market chock full of hardly used equipment. I can only guess how much and what kinds of data were contained on those systems. The anecdotal stories about personal information falling into the hands of people who buy used PCs is chilling enough. Bank records, stock trades, correspondence-all written onto magnetic media. What will prevent this from happening in radiology?
Direct attached storage solutions are the least of our worries. These connect hard drives to servers in much the same way hard drives are connected to CPUs in PCs and laptops. The data transfer chain is pretty easy to follow. But the trail gets tougher when using an NAS (network area storage), which shares data from a single archive with multiple servers. Its gets tougher still with SANs (storage area networks), which require a secondary network infrastructure to connect storage devices. And it's downright mind-boggling with hierarchical storage management, which moves data among multiple storage subsystems.
The march of progress makes computing and archival solutions obsolete within months, leading to their replacement with newer hardware. I'd like to know who will be in charge of tracking down and extinguishing every bit and byte. Because I've got a feeling pushing the delete button will not be enough.
Can AI Enhance PET/MRI Assessment for Extraprostatic Tumor Extension in Patients with PCa?
December 17th 2024The use of an adjunctive machine learning model led to 17 and 21 percent improvements in the AUC and sensitivity rate, respectively, for PET/MRI in diagnosing extraprostatic tumor extension in patients with primary prostate cancer.
Can Radiomics Bolster Low-Dose CT Prognostic Assessment for High-Risk Lung Adenocarcinoma?
December 16th 2024A CT-based radiomic model offered over 10 percent higher specificity and positive predictive value for high-risk lung adenocarcinoma in comparison to a radiographic model, according to external validation testing in a recent study.