MRI: The Next Generation
MRI has come a long way since its inception, and it has yet to cease evolving. New developments continue to surface, bringing with them changes in radiology practice patterns and opportunities to bolster revenues by attracting new patient populations. The evolution of multichannel coils ranks among the most significant developments. Systems with 32 and 48 channels are considered standard; 120- and 128-channel systems, as well as wireless coils and specialty surface coils, are emerging. Robert Lenkinski, PhD, is vice chair of radiology and director of radiology research, experimental radiology, and the 3T MR Spectroscopy Program at Beth Israel Deaconess Medical Center, Boston, Massachusetts. He says, “Multichannel coils, which bring improved signals and greater reliability, are having an enormous impact on MRI in general. They enable parallel imaging, higher-quality scans in a shorter time period (by an average of a factor of four), and broader coverage of body parts with very good signal strength.” Reduced scan time, stemming from access to additional channels, reduces motion-related problems and increases patient compliance. This, in turn, is spurring physicians to make MRI their study choice for a wider range of patients, including pediatric, elderly, and severely debilitated individuals. Richard Semelka, MD, says, “It is much easier now to perform MRI studies on patients who, in the past, could not or would not remain still long enough to complete the process, or presented some other kind of challenge.” Semelka is professor, director of MRI services, vice chair of clinical research, and vice chair for quality and safety of the Department of Radiology at the University of North Carolina Hospitals in Chapel Hill. He continues, “The less time patients must remain in the machine, the easier to perform MRI studies on them, and the greater the likelihood of compliance.” Lawrence Wald, PhD, of the Department of Radiology’s Martinos Center for Biomedical Imaging at Massachusetts General Hospital (MGH) in Boston, is director of the MGH NMR core. He adds that new multichannel phased-array coil systems also permit users to execute different, more exotic imaging procedures. He notes that while such procedures were previously impractical because of the amount of time needed to encode captured images, large coil arrays and multiple channels remove that factor from the equation. “For example, with a higher number of coils, spectroscopic imaging that once took three hours has a 16-fold acceleration, into the 10-minute range,” Wald states. “Capturing 0.75-mm isotropic, high-resolution 3D images of the brain can take 15 minutes, rather than many times that. The more channels, the more MRI will become a replacement for CT, because users are able to get patients in and out in comparable time, with heightened image sensitivity, and without compromising image quality.” See Figures 1, 2, and 3. Diffusion Imaging Migrates The advent of multichannel coil systems and parallel imaging, coupled with such hardware and technical advances as rapid echoplanar imaging and larger-amplitude gradient systems, has also paved the way for the use of diffusion-weighted imaging, not only for evaluating intracranial abnormalities, but for imaging other parts of the body. Figure 1. A 96-channel brain-array prototype constructed at Massachusetts General Hospital in Boston; the detector uses very small loop detectors placed on a close-fitting helmet. The high density of small detectors improves image sensitivity and increases the ability of parallel-imaging techniques to decrease scan time. Bachir Taouli, MD, director of body MRI at Mount Sinai Medical Center, New York, New York, says, “The combination of all of these factors helps to overcome many limitations that previously precluded the application of diffusion-weighted imaging to areas beyond the brain.” For example, prior to recent hardware and technical advances, inherent limitations in signal-to-noise ratio (SNR) and image resolution tended to compound artifacts when patients were evaluated in diffusion-weighted imaging mode. Figure 2. Section from a T1-weighted whole-brain volume acquired with a 32-channel brain-array prototype constructed at Massachusetts General Hospital in Boston; in this case, the scan time was conventional (8 minutes), but at a voxel volume that was four times smaller. In-plane resolution is 0.4 mm and partition thickness is 1.5 mm. Moreover, the larger fields of view necessary for abdominal imaging generally accentuate many of the artifacts inherent in diffusion-weighted imaging, as well as in single-shot echoplanar imaging. Now, however, the rapid imaging fostered by multichannel coil systems (along with shorter echo time, reduced echo-train length, and more rapid k-space filling) made possible by advanced technology has reduced artifacts and rendered diffusion-weighted imaging a more viable option. Radiologists are beginning to use diffusion-weighted imaging in abdominal and pelvic oncologic MRI studies, with single-shot echoplanar imaging employed in assessing lymphadenopathy and renal, hepatic, ovarian, and peritoneal masses, as well as prostatic, colorectal, uterine, and pulmonary tumors. Taouli notes particularly positive results using diffusion-weighted imaging in assessing patients with liver disease. Figure 3. A T2-weighted image acquired with a 96-channel brain-array prototype constructed at Massachusetts General Hospital in Boston; in this case, the highly parallel array was used to speed up the acquistion fourfold and reduce distortion artifacts. Each slice was acquired in 150 milliseconds, allowing a whole-brain acquisition in nine seconds. In-plane resolution is 1 mm and slice thickness is 2 mm. “The next-generation hardware and coil systems mean the diffusion-weighted option can be applied to liver imaging with improved image quality,” he reports. “Diffusion-weighted imaging allows qualitative and quantitative assessment of tissue diffusivity without the use of gadolinium chelates, which makes it a highly attractive technique, particularly in patients with severe renal dysfunction.” Some radiologists now simultaneously apply diffusion-weighted and conventional abdominal and pelvic MRI for better detection of many types of primary and metastatic tumors. “There seems to be improved sensitivity for tumors here because they are more conspicuous with diffusion-weighted imaging than with conventional T1, T2, and gadolinium-enhanced imaging,” Taouli says. “When diffusion-weighted imaging comes into play, background tissues are somewhat suppressed; most forms of tumor show restricted water diffusion, so tumor conspicuity becomes moderate to marked. Just as important, adding breath-hold diffusion-weighted imaging to routine abdominal MRI works much better at detecting additional tumor sites in oncology patients, compared with conventional MRI.” Diffusion-weighted imaging is also beginning to have clinical applications on the whole-body front when employed with background body signal suppression. Cancer staging represents a case in point because primary tumors and distant metastases alike show restricted diffusion, and suppression of background tissues permits small metastases to be seen easily. Evaluating the responses of primary tumors and metastases to chemotherapy or radiotherapy constitutes another such application, as does the evaluation of lymphadenopathy in patients whose metastases are predominantly nodal (Figure 4). Figure 4. Transverse fat-suppressed, breath-hold, single-shot echoplanar diffusion images of a 43–year-old woman with breast cancer treated with chemotherapy, obtained using b values of 0 and 500 s/mm2, with corresponding apparent diffusion coefficient (ADC) map and post-contrast image; the higher–signal-intensity necrotic center of the tumor on the b 0 image shows greater signal attenuation on the b 500 image with higher ADC (asterisk), compared with the cellular enhancing rim, which has restricted diffusion and lower ADC (arrows). Images courtesy of Bachir Taouli, MD, Mount Sinai Medical Center, New York, New York. What’s more, physicians are reportedly incorporating whole-body diffusion-weighted imaging into their surveillance protocols for lymphoma and leukemia; background tissue suppression, Taouli notes, allows lymph nodes to be shown with very high tumor-background contrast, and even small nodes are easily seen. In addition, the volumetric display of the whole-body diffusion-weighted dataset is more accurate than routine planar anatomic MRI in portraying nodal tumor distribution. Field-strength Advances The development of MRI equipment with higher field strength is also affecting the use of MRI technology and the variety of case types where it is applied. While machines operating at field strengths of 1T and 1.5T are still very much in use, manufacturers are releasing 3T equipment intended not for research settings, as has traditionally been the case, but for clinical ones. The heightened availability of coils and pads to increase signal and counter artifacts is pushing the envelope as well, as are such innovations as computer-aided detection and protocols for specific 3T MRI studies (for example, spectroscopy for prostate evaluation). “3T for all clinical applications is becoming very much a reality,” based partially on its general benefits, Semelka says. The SNR produced by 3T MRI is up to four times better than that generated by 1.5T MRI. Resolution with 3T imaging is also superior; together, improved SNR and higher spatial resolution lead to improved image clarity and diagnostic utility. “With 3T MRI, we can also use decreased scan times to reduce data artifacts related to motion,” Semelka notes. “This means we can perform more body MRI studies (as well as studies on children, geriatric patients, and anyone who can’t control movement for more than a very short duration) in a compressed interval while preserving image quality.” Lenkinski concurs, adding that the SNR of a 3T body coil is about the same as that of a 1.5T phased-array coil. “Adding a surface coil gives much more SNR headroom, which allows faster image acquisition and patient throughput or higher image resolution while revealing fine anatomic details and physiologic parameters,” he says. Other general advantages also are encouraging physicians to use 3T in clinical settings. For instance, 3T MRI technology permits radiologists to discover subtle abnormalities not seen using 1.5T imaging. Wald cites the example of a 3T MRI scan performed on a seizure victim; it showed heterotopic gray matter associated with a developmental problem. A previous 1.5T scan revealed no such abnormalities. Other benefits are easier identification of changes in metabolite peaks, as well as improvements in fat/water-suppression imaging techniques. Moreover, 3T MRI is said to enhance advanced functional MRI sequences, among them diffusion-weighted, diffusion-tensor, and blood oxygen level dependent imaging. Benefits and Body Parts As for more specific, individual applications of 3T MRI in nonresearch environments, Lenkinski cites scanning of the brain as a particular area of interest. “The fact that conventional brain imaging at 3T can be completed in the same exam time and can achieve a higher SNR than at 1.5T is just part of it,” he states. “With an SNR three to four times higher, 3T MRI more precisely localizes areas of activation, permitting accurate mapping of brain function in patients more than 90% of the time.” Figure 5. A T2-weighted axial fast-spin echo image taken of a prostate at 3T using an endorectal coil in combination with an external pelvic array; the high spatial resolution available using this combination allows the visualization of both detailed anatomy and small abnormalities within the gland. Lenkinski adds, “Diffusion-tensor imaging at 3T, unlike diffusion-tensor imaging at lower field strengths, makes it possible to view brainstem structure clearly. Contrast-enhanced studies performed at 3T allow significantly better differentiation between the brain and a tumor, and venography offers a better view of tumor environments than is obtainable at 1.5T.” The improvements offered by 3T and a combination of endo-rectal and surface-array coils are quite marked in the prostate, he adds (Figure 5). Musculoskeletal imaging using 3T MRI is also growing. Radiologists are finding that the improved SNR not only speeds up imaging of the joints, but yields an enhanced view of trabecular detail and improves fat saturation of the joints. A combination of fat saturation and contrast is said to enable users to perform more sensitive studies of inflammation, tumors, and healing. Breast imaging, particularly for women who are at risk for early breast cancer, is among other promising 3T body-imaging applications now making their ways into clinical radiology. Proponents of 3T breast MRI claim that it generates more detailed images (which are crucial for seeing tumor borders and ductal anatomy) and higher resolution, resulting in clearer/sharper images that are vital to the detection both of subtle cancers and of less-subtle cancers at earlier stages. Radiologists also consider 3T images more accurate because they exclude the whites of fatty tissue while leaving cancerous cells visible and are twice as sensitive as 1.5T images when it comes to contrast enhancement. Similarly, manufacturers of 3T scanners report that their equipment scans contiguous 1-mm slices, allowing up to 30% more of the breast to be scanned and reducing the likelihood that lesions could be missed due to skip (the scanning of alternating 1-mm slices instead of all slices). Additional applications of 3T MRI made possible by hardware and software developments include whole-body MRI, dimensional high-resolution cholangiopancreatography, staging of cervical carcinoma and evaluation of its extension beyond the cervical stroma, staging of rectal carcinoma, staging of prostate cancer, and high-resolution MR angiography for evaluation of the renal and mesenteric arteries. For MR spectroscopy in the abdomen and pelvis, 3T imaging with parallel acquisition might be preferable to 1.5T imaging in that the chemical-shift effect is doubled, producing improved spectral resolution of metabolites that are obscure at 1.5T. MRI Versus CT Recently, rising concerns about radiation exposure and the cost of PET/CT have led radiologists to question whether MRI might replace CT as a study mode of choice. There appears to be no clear answer. “With modern MRI advances, the faster speed of data acquisition, and clearer imaging without artifacts, there can certainly be more MRI procedures performed than is the case now,” Semelka says. “For example, there are a tremendous number of CT studies performed on children with seemingly minor injuries to ensure that there has been no undetected injury to the spleen, kidneys, and so forth. The ability to obtain higher-quality images, in less time and without sedation, would be a selling point for using MRI in these cases instead. By contrast, in major trauma cases, I don’t see MRI replacing CT completely any time soon.” Some analysts also believe that because diffusion-weighted imaging does not provide true metabolic information, it will probably be used to complement, rather than entirely replace, PET/CT. “MRI, together with diffusion-weighted imaging, can more realistically supplement PET/CT in cancer imaging, with initial evaluation using PET/CT used to zero in on primary and metastatic tumor sites and follow-up diffusion-weighted imaging used to monitor tumor response to therapy,” Taouli says. Still, he notes, the promise of next-generation MRI is seemingly unlimited. Additional Reading - Wide-open MRI Julie Ritzer Ross is a contributing writer for Radiology Business Journal.