In addition to technology and algorithm development, many clinical investigators work with the CT Clinical Innovation Center to evaluate emerging clinical applications and translate them into patient care.
Here are some of the CT features and applications that have been introduced into clinical practice and that are being studied to evaluate their performance in a clinical environment.
Automatic selection of the optimal tube potential
Images acquired at 80 kilovolt (kV) with automatic tube potential selection software (left) and at 120 kV not using this optimization tool (right). Images show chronic loculated ascites and fallopian tube adenocarcinoma. Side-by-side evaluation shows that the image quality of the 80 kV scan is equivalent to the 120 kV scan despite the dose reduction.
Description: The tube potential used to acquire CT data has a strong impact on the resulting image quality and the radiation dose that the patient receives.
A generalized approach was recently developed to automatically select the optimal tube potential in order to tailor the exam to both the specific patient and the specific clinical indication. Based on a similar strategy, Siemens Healthcare has implemented a software tool that may allow users to routinely incorporate individualized selection of the optimal tube potential in daily practice.
The performance of the automatic tube potential selection tool implemented by Siemens Healthcare in a clinical setting was assessed. In 101 CT angiography exams and 90 contrast-enhanced abdominopelvic CT exams, the mean dose reduction relative to the reference 120 kV protocol was 25 percent for CT angiography and 13 percent for abdominopelvic without sacrificing diagnostic image quality.
Significance to patient care: Radiation dose can be substantially reduced with an optimal tube potential without sacrificing diagnostic image quality. An automatic strategy was developed to make the selection of optimal tube potential a quantitative and efficient process.
Outcomes: Individualized selection of the most dose-efficient tube potential was successfully implemented on CT scanners.
Mineral stone characterization
Surface-rendered images of calcium oxalate monohydrate and calcium oxalate dihydrate stones calculated from clinical CT images demonstrating the differences in surface texture between the two stone types.
Description: Surface-rendered images of calcium oxalate monohydrate and calcium oxalate dihydrate stones calculated from clinical CT images demonstrated the differences in surface texture between the two stone types. Quantitation of the surface texture is able to differentiate between these two stone types.
In a preliminary study, a curvature-based shape descriptor was developed to accurately differentiate between calcium oxalate monohydrate kidney stones and calcium oxalate dihydrate kidney stones using images acquired from clinical CT scanners.
Significance to patient care: One type of kidney stone is easily broken using shockwave lithotripsy and the other type is very difficult to break in that manner.
Outcomes: It may be possible to avoid shock wave lithotripsy, a procedure that can cause kidney injury.
Affiliated grants: National Institute of Health (DK83007; Lieske, PI and DK59933; Williams, PI)
Adaptive nonlocal means (NLM) noise reduction
Close-up of an original CT slice of the abdomen (left) and the image after denoising algorithm (right). Note the improvement in conspicuity of hepatic metastases and the appearance of liver parenchyma.
Description: A novel denoising method based on the state-of-the-art nonlocal means filter was developed. The filter strength was adaptive to local noise, which was estimated analytically from CT images.
This image-based adaptive nonlocal means denoising method is independent of scanner platform and can process the images in a time frame consistent with the clinical workflow and therefore plays an important role in an overall strategy for reducing radiation dose.
Significance to patient care: Improvement of image quality through noise reduction allows for a reduction in radiation dose to patients during a CT scan.
Outcomes: State-of-the-art denoising algorithms may allow the radiation dose to be reduced by about 50 percent in many clinical CT exams without compromising diagnostic performance.
Pancreatic color mapping
Detection of pancreatic necrosis in the early phase of acute pancreatitis using subtraction color mapping.
Description: By subtracting contrast CT scan from noncontrast CT scan, an enhancement map of the pancreas can be obtained from which pancreatic necrosis can be easily visualized by looking for nonenhanced areas.
Detection of pancreatic necrosis at an early phase of acute pancreatitis is shown using subtraction color mapping. On the left, an axial contrast-enhanced CT image of the pancreas obtained on the day of admission shows a homogeneous low-density area in the head and body of the pancreas. In the center, a subtraction color map created from the same CT clearly shows nonenhancement (color-coded with purple) in the head and body of the pancreas, indicating the presence of pancreatic necrosis. Note viable enhancing tissue in the posterior portion of pancreatic head (color-coded with blue). Right, a CT obtained eight days later confirms the development of pancreatic necrosis.
Significance to patient care: The scan can appreciate the amount of pancreatic necrosis in patients to help steer treatment.
Outcomes: There was increased visualization of pancreatic necrosis.
Evaluation of silicone breast implants
A region of silicone extending outside the medial boundary of the implant is clearly seen on the right (arrow). In the gray-scale image (left), the silicone within the implant is the same brightness as muscle or glandular tissue.
Description: In this dual-energy CT exam of a patient with breast implants, a region of silicone extending outside the medial boundary of the implant is clearly seen using color-coding of high atomic number elements, such as silicone and calcium (arrow). If the leaked silicone were not surrounded by adipose tissue, it would be extremely difficult to differentiate from surrounding glandular tissue or the chest wall, which is the same brightness.
Significance to patient care: The diagnostic evaluation of the breast after augmentation remains a challenge in several respects. The most frequent and relevant diagnostic question in this context is implant integrity. This technique, which color-codes high atomic number elements such as silicone and calcium in red (right), increases the ability to detect loss of integrity in silicone implants.
Outcomes: Dual-energy CT may serve as an alternative technique for patients who need speedy evaluation of silicone breast implants.
Detection of post-traumatic bone marrow lesions in knee
Bone bruising image of a 29-year-old man using virtual noncalcium imaging.
Description: In this case of lateral femoral condyle bone bruising in a 29-year-old man, the coronal image of the right knee (a) shows no significant lesion, while the coronal virtual noncalcium image (b) demonstrates bone bruising (arrow) corresponding to the T2-weighted MR image (c) of the same region. Color-coding of the virtual noncalcium image (d) makes the bone marrow lesion readily visible (arrow).
Significance to patient care: Traumatic bone marrow lesions in patients with acute knee injuries aren't visible on regular CT exams, and MRI is usually the imaging method of choice. However, using dual-energy CT techniques, the calcium signal can be suppressed and post-traumatic bone marrow lesions can be detected.
Outcomes: Dual-energy CT may provide an alternative for patients who can't have MRI safely because of contraindications or when MRI is not available.
Detection of urinary stones on virtual nonenhanced images generated using dual-energy CT
In this dual-energy CT urogram study, the renal calculus (arrows) visible in the true noncontrast scan (top left) is hidden in the contrast-enhanced scan (top right). Using the virtual, noncontrast, dual-energy application, the iodine can be electronically removed, revealing the renal calculus (bottom).
Description: Nonenhanced CT is the reference standard for the detection of urinary stones, which are often obscured by high-attenuating iodinated contrast material in the renal parenchyma or collecting system in contrast-enhanced CT urography. Removing the iodine electronically using a dual-energy CT virtual nonenhanced technique allows depiction of urinary stones submerged in iodine solutions.
Significance to patient care: This application could potentially replace the need for the true noncontrast portion of a CT urogram, thus lowering patient dose.
Outcomes: Detection of urinary stones was possible in contrast-enhanced CT images.
Affiliated grant: National Institute of Health (DK83007; Lieske, PI)
Dual-energy CT can be used to automatically subtract bone signal from CT angiographic studies in order to better visualize vessel anatomy. The bone signal can be quickly turned on (left, top and bottom) or off (right, top and bottom), allowing surgeons to easily view the vascular tree and the bony anatomic landmarks essential for surgical planning.
Description: Dual-energy CT can differentiate calcium and iodine based on the energy dependence of X-ray photons in these two materials. Bone, which is mainly composed of calcium, can then be easily removed while vessels filled with iodine remain.
Significance to patient care: Turning off the bone signal setting on the CT scanner allows surgeons to see vessel anatomy. Turning it on allows surgeons to see bony anatomic landmarks. Both are crucial in surgical planning.
Outcomes: There was appreciation of both bone and systems within bone.
Noninvasive differentiation of urinary stone compositions in patients with nephrolithiasis using dual-source dual-energy CT
Dual-energy CT can be used to differentiate chemical composition of urinary stones. Sample images from cases with different stone types: uric acid (A, color-coded in red), cystine (B, yellow), calcium oxalate (C, green) and apatite (D, blue).
Using commercial software, multiple renal stones in this patient were determined to be of non-UA composition (color-coded in blue). Further analysis suggested the stones to be either calcium oxalate or brushite. Later analysis by micro-CT and infrared spectroscopy confirmed the stones to be calcium oxalate.
Description: Dual-energy CT using a dual-source CT scanner provides in vivo differentiation of uric acid stones with greater than 95 percent accuracy. The second-generation dual-source CT scanner includes an additional tin filter on the high-energy X-ray tube (Sn140kV) to increase the X-ray spectra separation. This, in turn, increases the differences in dual-energy CT number ratio between stone types. Differentiation between four groups of stone types was feasible using the second-generation dual-source CT scanner.
Significance to patient care: Noninvasive identification of stone composition may assist selection of a treatment strategy tailored to the patient's specific stone type. For example, uric acid stones can be treated with medications that alkalinize the urine. Other stone types, such as cystine or brushite, are highly resistant to extracorporeal shock wave lithotripsy, and surgical removal should be considered when spontaneous passage fails.
Outcomes: This method allows individualized treatment planning for patient with nephrolithiasis.
Affiliated grant: National Institute of Health (DK83007; Lieske, PI and DK59933; Williams, PI)
Dual-energy CT can easily distinguish uric acid crystals (gout), which are color-coded green here, from calcium-containing crystals (such as calcium pyrophosphate dihydrate), which are color-coded purple here, similar to bone.
Description: Dual-energy CT can easily distinguish uric acid crystals from calcium-containing crystals that are similar to bone (such as calcium pyrophosphate dehydrate). Studies demonstrated that dual-energy CT performs well in the detection of uric acid crystals in gout and can add information to joint fluid aspiration.
Significance to patient care: Patients may not have to undergo joint fluid aspiration, a painful procedure for those with an acutely inflamed joint. In addition, this tool could provide a diagnostic modality for those who can't or don't undergo joint aspiration.
Outcomes: A separation of the crystals that make up gout help diagnose the severity of the disease in bony anatomy.
Additional focus areas
Other focus areas in the CT Clinical Innovation Center include:
- Cardiovascular radiology
- CT protocol management and optimization
- Gastrointestinal radiology
- Genitourinary radiology
- Human observer performance studies and analysis
- Kinematic wrist imaging
- Musculoskeletal radiology
- Neurological radiology
- Pediatric dose reduction
- Polycystic kidney disease
- Renal and cardiac physiology
- Spectral CT and vascular plaque imaging
- Thoracic radiology