Patient demand stays too low for speed to matterImage quality, not speed, is the focus of Philips Medical Solutions in the development of its PET equipment."There is a difference in philosophy between us and some of our
Patient demand stays too low for speed to matter
Image quality, not speed, is the focus of Philips Medical Solutions in the development of its PET equipment.
"There is a difference in philosophy between us and some of our competitors," said Horace Hines, Ph.D., chief technical officer for Philips' line of nuclear medicine products. "Some competitors seem to think that speed is the most important attribute. We don't."
The company, which was catapulted into the upper strata of nuclear medicine in December 2000 with the acquisition of ADAC Laboratories (SCAN 1/31/01), has adopted radiology's time-honored mantra of image quality as the number-one priority. Central to that way of thinking is the belief that all patients should get a good study, according to Hines, who expects no argument from competitors on that score. But he goes on to state that PET utilization is so low right now-about three patients per day per installed scanner-that developing technology to reduce scan time solves a problem for a very limited number of sites. Throughput simply is not the primary issue.
"What we need more of are good PET images to grow the industry and the patient volume, so that we will actually need speed several years from now," Hines said. "That is the reason Philips is working on methods that improve image quality."
The company has put its money on a scintillator called GSO (gadolinium oxyorthosilicate), which promotes image quality by allowing more counts over a set time than the previously developed and still widely used BGO (bismuth germanate). Scintillators are at the heart of the issue over PET image quality versus throughput. Scintillator crystals emit a flash of light after being struck by a high-energy photon. This flash is then counted and, with thousands of others, translated into an image. The time needed to acquire the number of counts needed for a diagnostic-quality image, therefore, depends on the efficiency of the crystal-the time needed for the crystal to flash and return to a passive state so another flash can be generated.
The number of counts and the time needed to acquire them become an issue because the balance of these two determines image quality (many counts) or shorter scan times (fewer counts), leading to increased productivity and patient comfort. Philips has chosen image quality over productivity. Hines said that the time saved by shortening the period for data acquisition is typically lost through patient handling.
"The mechanics of getting a patient ready for a scan-going to the waiting room, getting the patient, explaining the procedure, putting the patient on a scanner, then getting the patient off the table and preparing the room for the next patient-takes about 20 minutes," he said. "So what have you gained? You are not going to change the throughput in your department that dramatically."
The trade-off is always there, said Josh Gurewitz, vice president of marketing for Philips' nuclear medicine product line. Gains in speed come at the expense of data.
"If you are looking for small lesions, you want the best quality you can get," he said. "Since the average site is doing fewer than three patients per day, image quality-not how quickly you can get the patient done-should be the primary focus."
In the end, the question revolves around how many counts are enough and how good is good enough. This is a clinical question that is best answered by physicians, not equipment vendors, according to Philips executives. They do not wish to intrude on the clinical practice of PET, Gurewitz said, but rather to underscore the opportunities that currently exist within this practice.
"So long as the time is available, scans should deliver the best quality possible," he said. "This promotes the industry by improving the credibility of PET and helps physicians build the volume that will support the argument that technology needs to go faster."
Can MRI-Based AI Enhance Pre-Op Prediction of Tumor Deposits in Patients with Rectal Cancer?
October 31st 2024For patients with rectal cancer, an emerging nomogram that combines deep learning and clinical factors had greater than 16 percent and 23 percent increases in accuracy and specificity, respectively, for pre-op prediction of tumor deposits in comparison to clinical factors alone.