CT Hounsfield numbers of soft tissues on unenhanced abdominal CT scans: variability between two different manufacturers' MDCT scanners

Misinterpretations about CT numbers, material decomposition, and elemental quantification

BACKGROUND

Quantitative CT imaging, particularly iodine and calcium quantification, is an important CT-based biomarker.

PURPOSE

This study quantifies sources of errors in quantitative CT imaging in both single-energy and spectral CT.

MATERIALS AND METHODS

This work examines the theoretical relationship between CT numbers, linear attenuation coefficient, and material quantification. We derive four understandings: (1) CT numbers are not proportional with element mass in vivo, (2) CT numbers are proportional with element mass only when contained in a voxel of pure water, (3) iodine-water material decomposition is never accurate in vivo, and (4) for error-free material decomposition a voxel must only consist of the basis decomposition vectors. Misinterpretation-based errors are calculated using the National Institute of Standards and Technology (NIST) XCOM database for: tissue chemical compositions, clinical concentrations of hydroxyapatite (HAP), and iodine. Quantification errors are also demonstrated experimentally using phantoms.

RESULTS

In single-energy CT, misinterpretation-induced errors for HAP density in adipose, muscle, lung, soft tissue, and blood ranged from 0-132%, i.e., a mass error of 0-749 mg/cm3. In spectral CT, errors with iodine in the same tissues resulted in a range of < 0.1-33% error, resulting in a mass error of < 0.1-1.2 mg/mL.

CONCLUSION

Our work demonstrates material quantification is fundamentally limited when measured in vivo due to measurement conditions differing from assumed and the errors are at or above detection limits for bone mineral density (BMD) and spectral iodine quantification. To define CT-derived biomarkers, the errors we demonstrate should either be avoided or built into uncertainty bounds.

CLINICAL RELEVANCE STATEMENT

Improving error bounds in quantitative CT biomarkers, specifically in iodine and BMD quantification, could lead to improvements in clinical care aspects based on quantitative CT.

KEY POINTS

CT numbers are only proportional with element mass only when contained in a voxel of pure water, therefore iodine-water material decomposition is never accurate in vivo. Misinterpretation-induced errors ranged from 0-132% for HAP density and < 0.1-33% in spectral CT with iodine. For error-free material decomposition, a voxel must only consist of the basis decomposition vectors.

MRI-based lesion quality score assessing ossification of the posterior longitudinal ligament of the cervical spine

BACKGROUND CONTEXT

No method currently exists for MRI-based determination of ossification of the posterior longitudinal ligament (OPLL) of the cervical spine using objective criteria.

PURPOSE

The purpose of this study was to develop an MRI-based score to determine whether a lesion represents a cervical OPLL lesion and to establish the objective diagnostic value.

STUDY DESIGN

Retrospective cohort in a single medical institution.

PATIENT SAMPLE

Thirty-five patients undergoing surgery for OPLL (Group A) and 99 patients undergoing cervical disc arthroplasty for soft disc herniation (Group B) between 2011 and 2020 were retrospectively included. All OPLL lesions on unenhanced MRI scan were correlated with a corresponding CT scan. Demographics were comparable between the two groups.

OUTCOME MEASURES (PHYSIOLOGIC MEASURES)

Using unenhanced magnetic resonance imaging (MRI), the T1- and T2- lesion quality (LQ) scores were calculated. Receiver operating characteristic (ROC) analysis was performed to calculate the area-under-the-curve (AUC) of both LQ scores as a predictor of the presence of OPLL. Computed tomography (CT)-based Hounsfield unit (HU) values of OPLL lesions were obtained and compared with both LQ scores. The LQ scores for MRI scanners from different manufacturers were compared using Student's t test to confirm the validity of the LQ score by scanner type.

METHODS

The regions of interest for signal intensity (SI) were defined as the darkest site of the lesion and the cerebrospinal fluid (CSF) at the cerebellomedullary cistern. The T1 and T2 LQ scores were measured as the ratio of the SI at the darkest site of the lesion divided by the SI of the CSF.

RESULTS

The T1 and T2 LQ scores in Group A were significantly lower than those in Group B (p<.001). ROC analysis determined that T1 and T2 LQ scores of 0.46 and 0.07, respectively, could distinguish the presence of OPLL with an accuracy of 0.93 and 0.89, respectively (p<.001). When the T1 LQ score of the lesion is <0.46, a diagnosis of OPLL may be suspected with 100% sensitivity and 92.3% specificity. The HU of the lesion had a moderate negative correlation with the T1 LQ score (r=-0.665, p<.0001). Both LQ scores were unaffected by manufacturer type.

CONCLUSIONS

This study found a correlation between the MRI-based T1 LQ scores and CT-based HU value for identifying OPLL lesions. Additional studies will be needed to validate that the T1 LQ score from the unenhanced MRI scan can identify cervical OPLL.

The Use of Low-Dose Chest Computed Tomography for the Diagnosis and Monitoring of Pulmonary Infections in Patients with Hematologic Malignancies

The study aimed to assess the image quality and diagnostic performance of low-dose Chest Computed Tomography (LDCCT) in detecting pulmonary infections in patients with hematologic malignancies. A total of 164 neutropenic patients underwent 256 consecutive CT examinations, comparing 149 LDCCT and 107 Standard-Dose Chest CT (SDCCT) between May 2015 and June 2019. LDCCT demonstrated a 47% reduction in radiation dose while maintaining acceptable image noise and quality compared to SDCCT. However, LDCCT exhibited lower sensitivity in detecting consolidation (27.5%) and ground glass opacity (64.4%) compared to SDCCT (45.8% and 82.2%, respectively) with all the respective p-values from unadjusted and adjusted for sex, age, and BMI analyses being lower than 0.006 and the corresponding Odds Ratios of detection ranging from 0.30 to 0.34. Similar trends were observed for nodules ≥3 mm and ground glass halo in nodules but were not affected by sex, age and BMI. No significant differences were found for cavitation in nodules, diffuse interlobular septal thickening, pleural effusion, pericardial effusion, and lymphadenopathy. In conclusion, LDCCT achieved substantial dose reduction with satisfactory image quality but showed limitations in detecting specific radiologic findings associated with pulmonary infections in neutropenic patients compared to SDCCT.

A phantom to simulate organ motion and its effect on dose distribution in carbon ion therapy for pancreatic cancer

Objective. Carbon ion radiotherapy is a promising radiation technique for malignancies like pancreatic cancer. However, organs' motion imposes challenges for achieving homogeneous dose delivery. In this study, an anthropomorphicPancreasPhantom forIon-beamTherapy (PPIeT) was developed to simulate breathing and gastrointestinal motion during radiotherapy.Approach. The developed phantom contains a pancreas, two kidneys, a duodenum, a spine and a spinal cord. The shell of the organs was 3D printed and filled with agarose-based mixtures. Hounsfield Units (HU) of PPIeTs' organs were measured by CT. The pancreas motion amplitude in cranial-caudal (CC) direction was evaluated from patients' 4D CT data. Motions within the obtained range were simulated and analyzed in PPIeT using MRI. Additionally, GI motion was mimicked by changing the volume of the duodenum and quantified by MRI. A patient-like treatment plan was calculated for carbon ions, and the phantom was irradiated in a static and moving condition. Dose measurements in the organs were performed using an ionization chamber and dosimetric films.Main results. PPIeT presented tissue equivalent HU and reproducible breathing-induced CC displacements of the pancreas between (3.98 ± 0.36) mm and a maximum of (18.19 ± 0.44) mm. The observed maximum change in distance of (14.28 ± 0.12) mm between pancreas and duodenum was consistent with findings in patients. Carbon ion irradiation revealed homogenous coverage of the virtual tumor at the pancreas in static condition with a 1% deviation from the treatment plan. Instead, the dose delivery during motion with the maximum amplitude yielded an underdosage of 21% at the target and an increased uncertainty by two orders of magnitude.Significance. A dedicated phantom was designed and developed for breathing motion assessment of dose deposition during carbon ion radiotherapy. PPIeT is a unique tool for dose verification in the pancreas and its organs at risk during end-to-end tests.

Detection of bone marrow edema in differential diagnoses of odontogenic cysts using dual-energy computed tomography

PURPOSE

In this study, we prospectively investigated the relationship between bone marrow edema (BME) and odontogenic cysts and explored the possibility of using dual-energy computed tomography (DECT) as an auxiliary tool for the diagnosis of odontogenic cysts.

METHODS

This cross-sectional study included 73 patients who underwent the DECT scan and surgery for odontogenic cysts or odontogenic tumors. The virtual noncalcium (VNCa) computed tomography (CT) values and CT values were measured at several sites. The predictor variable was diagnosis, and the other variables included age, sex, and sites. The primary outcome was VNCa CT value. Variables were tested using the chi-square test or the Kruskal-Wallis test. The VNCa CT and CT values were tested using the Scheffe test for multiple comparisons. All variables were analyzed as independent variables affecting the VNCa CT values around the lesion in the multiple regression analysis.

RESULT

There were 35 men and 38 women. The mean patient age was 50.0 ± 19.5 years (range: 8-86). The VNCa CT values (- 6.2 ± 34.3) around the lesion in patients with RCs were significantly higher than those in patients with dentigerous cysts (- 44.4 ± 28.6) and odontogenic keratocysts (- 67.3 ± 19.5). In multiple regression analysis, the VNCa CT values around the lesion showed a significant positive correlation with histological results (regression coefficient: - 0.605, P < 0.001).

CONCLUSION

The presence of BME is associated with radicular cysts, and DECT can be used as an auxiliary tool for radicular cyst diagnosis.

Assessment of CBCT gray value in different regions-of-interest and fields-of-view compared to Hounsfield unit

OBJECTIVES

Different factors can affect the discrepancy between the gray value (GV) measurements obtained from CBCT and the Hounsfield unit (HU) derived from multidetector CT (MDCT), which is considered the gold-standard density scale. This study aimed to explore the impact of region of interest (ROI) location and field of view (FOV) size on the difference between these two scales as a potential source of error.

METHODS

Three phantoms, each consisting of a water-filled plastic bin containing a dry dentate human skull, were prepared. CBCT scans were conducted using the NewTom VGi evo system, while MDCT scans were performed using Philips system. Three different FOV sizes (8 × 8 cm, 8 × 12 cm, and 12 × 15 cm) were used, and the GVs obtained from eight distinct ROIs were compared with the HUs from the MDCT scans. The ROIs included dental and bony regions within the anterior and posterior areas of both jaws. Statistical analyses were performed using SPSS v. 26.

RESULTS

The GVs derived from CBCT images were significantly influenced by both ROI location and FOV size (p < 0.05 for both factors). Following the comparison between GVs and HUs, the anterior mandibular bone ROI represented the minimum error, while the posterior mandibular teeth exhibited the maximum error. Moreover, the 8 × 8 cm and 12 × 15 cm FOVs resulted in the lowest and highest degrees of GV error, respectively.

CONCLUSIONS

The ROI location and the FOV size can significantly affect the GVs obtained from CBCT images. It is not recommended to use the GV scale within the posterior mandibular teeth region due to the potential for error. Additionally, selecting smaller FOV sizes, such as 8 × 8 cm, can provide GVs closer to the gold-standard numbers.

Effects of rhBMP-2 with various carriers on maxillofacial bone regeneration through computed tomography evaluation

BACKGROUND

rhBMP-2 is regarded as the most potent osteoinductive growth factor, and it has been used in the oral cavity with different carriers. The purpose of this study is to evaluate the bone-regenerative effect of rhBMP-2 delivered with different carrier systems through three-dimensional cone beam computed tomography analysis.

METHOD

A total of 112 patients underwent oral surgery with rhBMP-2 application (Group 1, n = 53) or without rhBMP-2 application (Group 2, n = 59). Group 1 was divided into 3 groups according to carriers, rhBMP-2 with allograft (Group 1-1, n = 34), rhBMP-2 with xenograft (Group 1-2, n = 5), and rhBMP-2 with absorbable collagen sponge (Group 1-3, n = 14). Cone beam computed tomography scans were taken before surgery (T0) 6 months after surgery (T1). The volume of defects was measured through the three-dimensional image analysis tool.

RESULTS

The average bone regeneration rate of Group 1 was significantly greater than that of Group 2. Within Group 1, the group that used allograft as a carrier (Group 1-1) showed significantly higher bone regeneration rates than the group that used absorbable collagen sponge as a carrier (Group 1-3).

CONCLUSION

The use of rhBMP-2 after oral surgery results in a superior bone regeneration rate compared to not using rhBMP-2, and its efficacy depends on the carriers it is used with. Allograft affects bone regeneration more than absorbable collagen sponge when it is carried with rhBMP-2. Therefore, the appropriate use of rhBMP-2 with suitable bone grafting materials is useful for promoting postoperative bone regeneration in oral surgery.

CT Number Accuracy and Association With Object Size: A Phantom Study Comparing Energy-Integrating Detector CT and Deep Silicon Photon-Counting Detector CT

BACKGROUND. Variable beam hardening based on patient size causes variation in CT numbers for energy-integrating detector (EID) CT. Photon-counting detector (PCD) CT more accurately determines effective beam energy, potentially improving CT number reliability. OBJECTIVE. The purpose of the present study was to compare EID CT and deep silicon PCD CT in terms of both the effect of changes in object size on CT number and the overall accuracy of CT numbers. METHODS. A phantom with polyethylene rings of varying sizes (mimicking patient sizes) as well as inserts of different materials was scanned on an EID CT scanner in single-energy (SE) mode (120-kV images) and in rapid-kilovoltage-switching dual-energy (DE) mode (70-keV images) and on a prototype deep silicon PCD CT scanner (70-keV images). ROIs were placed to measure the CT numbers of the materials. Slopes of CT number as a function of object size were computed. Materials' ideal CT number at 70 keV was computed using the National Institute of Standards and Technology XCOM Photon Cross Sections Database. The root mean square error (RMSE) between measured and ideal numbers was calculated across object sizes. RESULTS. Slope (expressed as Hounsfield units per centimeter) was significantly closer to zero (i.e., less variation in CT number as a function of size) for PCD CT than for SE EID CT for air (1.2 vs 2.4 HU/cm), water (-0.3 vs -1.0 HU/cm), iodine (-1.1 vs -4.5 HU/cm), and bone (-2.5 vs -10.1 HU/cm) and for PCD CT than for DE EID CT for air (1.2 vs 2.8 HU/cm), water (-0.3 vs -1.0 HU/cm), polystyrene (-0.2 vs -0.9 HU/cm), iodine (-1.1 vs -1.9 HU/cm), and bone (-2.5 vs -6.2 HU/cm) (p < .05). For all tested materials, PCD CT had the smallest RMSE, indicating CT numbers closest to ideal numbers; specifically, RMSE (expressed as Hounsfield units) for SE EID CT, DE EID CT, and PCD CT was 32, 44, and 17 HU for air; 7, 8, and 3 HU for water; 9, 10, and 4 HU for polystyrene; 31, 37, and 13 HU for iodine; and 69, 81, and 20 HU for bone, respectively. CONCLUSION. For numerous materials, deep silicon PCD CT, in comparison with SE EID CT and DE EID CT, showed lower CT number variability as a function of size and CT numbers closer to ideal numbers. CLINICAL IMPACT. Greater reliability of CT numbers for PCD CT is important given the dependence of diagnostic pathways on CT numbers.

Deep learning-based diagnosis of osteoblastic bone metastases and bone islands in computed tomograph images: a multicenter diagnostic study

OBJECTIVE

To develop and validate a deep learning (DL) model based on CT for differentiating bone islands and osteoblastic bone metastases.

MATERIALS AND METHODS

The patients with sclerosing bone lesions (SBLs) were retrospectively included in three hospitals. The images from site 1 were randomly assigned to the training (70%) and intrinsic verification (10%) datasets for developing the two-dimensional (2D) DL model (single-slice input) and "2.5-dimensional" (2.5D) DL model (three-slice input) and to the internal validation dataset (20%) for evaluating the performance of both models. The diagnostic performance was evaluated using the internal validation set from site 1 and additional external validation datasets from site 2 and site 3. And statistically analyze the performance of 2D and 2.5D DL models.

RESULTS

In total, 1918 SBLs in 728 patients in site 1, 122 SBLs in 71 patients in site 2, and 71 SBLs in 47 patients in site 3 were used to develop and test the 2D and 2.5D DL models. The best performance was obtained using the 2.5D DL model, which achieved an AUC of 0.996 (95% confidence interval [CI], 0.995-0.996), 0.958 (95% CI, 0.958-0.960), and 0.952 (95% CI, 0.951-0.953) and accuracies of 0.950, 0.902, and 0.863 for the internal validation set, the external validation set from site 2 and site 3, respectively.

CONCLUSION

A DL model based on a three-slice CT image input (2.5D DL model) can improve the prediction of osteoblastic bone metastases, which can facilitate clinical decision-making.

KEY POINTS

• This study investigated the value of deep learning models in identifying bone islands and osteoblastic bone metastases. • Three-slice CT image input (2.5D DL model) outweighed the 2D model in the classification of sclerosing bone lesions. • The 2.5D deep learning model showed excellent performance using the internal (AUC, 0.996) and two external (AUC, 0.958; AUC, 0.952) validation sets.

3D-printed iodine-ink CT phantom for radiomics feature extraction - advantages and challenges

BACKGROUND

To test and validate novel CT techniques, such as texture analysis in radiomics, repeat measurements are required. Current anthropomorphic phantoms lack fine texture and true anatomic representation. 3D-printing of iodinated ink on paper is a promising phantom manufacturing technique. Previously acquired or artificially created CT data can be used to generate realistic phantoms.

PURPOSE

To present the design process of an anthropomorphic 3D-printed iodine ink phantom, highlighting the different advantages and pitfalls in its use. To analyze the phantom's X-ray attenuation properties, and the influences of the printing process on the imaging characteristics, by comparing it to the original input dataset.

METHODS

Two patient CT scans and artificially generated test patterns were combined in a single dataset for phantom printing and cropped to a size of 26 × 19 × 30 cm3 . This DICOM dataset was printed on paper using iodinated ink. The phantom was CT-scanned and compared to the original image dataset used for printing the phantom. The water-equivalent diameter of the phantom was compared to that of a patient cohort (N = 104). Iodine concentrations in the phantom were measured using dual-energy CT. 86 radiomics features were extracted from 10 repeat phantom scans and the input dataset. Features were compared using a histogram analysis and a PCA individually and overall, respectively. The frequency content was compared using the normalized spectrum modulus.

RESULTS

Low density structures are depicted incorrectly, while soft tissue structures show excellent visual accordance with the input dataset. Maximum deviations of around 30 HU between the original dataset and phantom HU values were observed. The phantom has X-ray attenuation properties comparable to a lightweight adult patient (∼54 kg, BMI 19 kg/m2 ). Iodine concentrations in the phantom varied between 0 and 50 mg/ml. PCA of radiomics features shows different tissue types separate in similar areas of PCA representation in the phantom scans as in the input dataset. Individual feature analysis revealed systematic shift of first order radiomics features compared to the original dataset, while some higher order radiomics features did not. The normalized frequency modulus |f(ω)| of the phantom data agrees well with the original data. However, all frequencies systematically occur more frequently in the phantom compared to the maximum of the spectrum modulus than in the original data set, especially for mid-frequencies (e.g., for ω = 0.3942 mm-1 , |f(ω)|original  = 0.09 * |fmax |original and |f(ω)|phantom  = 0.12 * |fmax |phantom ).

CONCLUSIONS

3D-iodine-ink-printing technology can be used to print anthropomorphic phantoms with a water-equivalent diameter of a lightweight adult patient. Challenges include small residual air enclosures and the fidelity of HU values. For soft tissue, there is a good agreement between the HU values of the phantom and input data set. Radiomics texture features of the phantom scans are similar to the input data set, but systematic shifts of radiomics features in first order features, due to differences in HU values, need to be considered. The paper substrate influences the spatial frequency distribution of the phantom scans. This phantom type is of very limited use for dual-energy CT analyses.

Reporting breast density on chest computed tomography

Women are encouraged to have a yearly mammogram and in addition to screening for breast cancer, the radiologist reports the patient's breast density. High breast density increases a woman's risk of developing breast cancer. The number of chest computed tomography (CT) scans performed each year is increasing. Chest CT scans for lung cancer screening in high-risk patients are the standard of care. Important additional findings can be identified on these exams including coronary artery calcifications, thyroid nodules, and breast density. Our previous research has shown that breast density can be reliably graded on chest CT and is comparable to mammographic grading. However, the inter-reader agreement was higher for chest CT. It is important that thoracic radiologists include the grading of breast density in their chest CT reports. According to mammography literature, this information has proven to be helpful for early detection of breast cancer. Federal legislation recommends notifying both providers and patients about breast density on mammography and so it follows that if we see the same information on chest CT, we should report it so that at the very least the clinician can encourage their patient to have a routine mammogram.

Extracellular volume fraction using contrast-enhanced CT is useful in differentiating intrahepatic cholangiocellular carcinoma from hepatocellular carcinoma

Objectives

To evaluate whether tumor extracellular volume fraction (fECV) on contrast-enhanced computed tomography (CT) aids in the differentiation between intrahepatic cholangiocarcinoma (ICC) and hepatocellular carcinoma (HCC).

Methods

In this retrospective study, 113 patients with pathologically confirmed ICC (n = 39) or HCC (n = 74) who had undergone preoperative contrast-enhanced CT were enrolled. Enhancement values of the tumor (Etumor) and aorta (Eaorta) were obtained in the precontrast and equilibrium phase CT images. fECV was calculated using the following equation: fECV [%] = Etumor/Eaorta × (100 - hematocrit [%]). fECV values were compared between the ICC and HCC groups using Welch's t-test. The diagnostic performance of fECV for differentiating ICC and HCC was assessed using receiver-operating characteristic (ROC) analysis. fECV and the CT imaging features of tumors were evaluated by two radiologists. Multivariate logistic regression analysis was performed to identify factors predicting a diagnosis of ICC.

Results

Mean fECV was significantly higher in ICCs (43.8% ± 13.2%) than that in HCCs (31.6% ± 9.0%, p < 0.001). The area under the curve for differentiating ICC from HCC was 0.763 when the cutoff value of fECV was 41.5%. The multivariate analysis identified fECV (unit OR: 1.10; 95% CI: 1.01-1.21; p < 0.05), peripheral rim enhancement during the arterial phase (OR: 17.0; 95% CI: 1.29-225; p < 0.05), and absence of washout pattern (OR: 235; 95% CI: 14.03-3933; p < 0.001) as independent CT features for differentiating between the two tumor types.

Conclusions

A high value of fECV, peripheral rim enhancement during the arterial phase, and absence of washout pattern were independent factors in the differentiation of ICC from HCC.

Compton scattering geometry: a tool to study radiation interaction characteristics of rare earth compounds doped in low-Z organic compound

The present measurements comprise of Compton scattering technique at six energies from 0.242 MeV to 0.402 MeV (not available from conventional radioisotopes) by scattering of primary gamma photon beam of 0.662 MeV energy from cylindrical aluminium target at different scattering angles. Two inorganic (rare-earth) compounds, Lanthanum (3+) nitrate hexahydrate and Samarium (3+) nitrate hexahydrate in a Low-Z organic solvent (acetone), have shown certain radiation interaction characteristics are the subject of study. The collimated beam of scattered gamma rays impinges on the plastic container having solution of rare earth compounds in the acetone of different concentrations. The transmitted gamma ray beam is detected by a well-collimated 2″ × 2″ NaI(Tl) scintillation detector and is analysed by a PC-based ORTEC Mastero-32 MCA. Attenuation coefficients along with some shielding parameters i.e. molar extinction coefficients, half value layer, tenth value layer and mean free path are evaluated. Besides this, Computed Tomography numbers and photon interaction cross-sections (photoelectric, coherent and incoherent) are also determined. The measured values of these parameters are compared with WinXCom software package’s values and are found to be in good agreement. The available data on rare earth salts’ solutions, in the current measurements, is scientifically important in nuclear and radiation physics, bridging the gap in which radiation workers do not have access to such data.

Cervical Paraspinal Muscle Fatty Infiltration is Directly Related to Extension Reserve in Patients With Cervical Spine Pathology

STUDY DESIGN/SETTING

Retrospective review of a prospectively collected database.

OBJECTIVE

The objective of this study was to determine the relationship between paracervical muscle area, density, and fat infiltration and cervical alignment among patients presenting with cervical spine pathology.

BACKGROUND CONTEXT

The impact of cervical spine alignment on clinical outcomes has been extensively studied, but little is known about the association between spinal alignment and cervical paraspinal musculature.

METHODS

We examined computed tomography scans and radiographs for patients presenting with cervical spine pathology. The posterior paracervical muscle area, density, and fat infiltration was calculated on axial slices at C2, C4, C6, and T1. We measured radiographic parameters including cervical sagittal vertical axis, cervical lordosis, T1 slope (T1S), range of motion of the cervical spine. We performed Pearson correlation tests to determine if there were significant relationships between muscle measurements and alignment parameters.

RESULTS

The study included 51 patients. The paracervical muscle area was higher for males at C2 ( P =0.005), C4 ( P =0.001), and T1 ( P =0.002). There was a positive correlation between age and fat infiltration at C2, C4, C6, and T1 (all P <0.05). The cervical sagittal vertical axis positively correlated with muscle cross-sectional area at C2 ( P =0.013) and C4 ( P =0.013). Overall cervical range of motion directly correlated with muscle density at C2 ( r =0.48, P =0.003), C4 ( r =0.41, P =0.01), and C6 ( r =0.53. P <0.001) and indirectly correlated with fat infiltration at C2 ( r =-0.40, P =0.02), C4 ( r =-0.32, P =0.04), and C6 ( r =-0.35, P =0.02). Muscle density correlated directly with reserve of extension at C2 ( r =0.57, P =0.009), C4 ( r =0.48, P =0.037), and C6 ( r =0.47, P =0.033). Reserve of extension indirectly correlated with fat infiltration at C2 ( r =0.65, P =0.006), C4 ( r =0.47, P =0.037), and C6 ( r =0.48, P =0.029).

CONCLUSIONS

We have identified specific changes in paracervical muscle that are associated with a patient's ability to extend their cervical spine.

The automated measurement of CT number linearity using an ACR accreditation phantom

We developed a software to automatically measure the linearity between the CT numbers and densities of objects using an ACR 464 CT phantom, and investigated the CT number linearity of 16 different CT scanners. The software included a segmentation-rotation method. After segmenting five objects within the phantom image, the software computed the mean CT number of each object and plotted a graph between the CT numbers and densities of the objects. Linear regression and coefficients of regression, R², were automatically calculated. The software was used to investigate the CT number linearity of 16 CT scanners from Toshiba, Siemens, Hitachi, and GE installed at 16 hospitals in Indonesia. The linearity of the CT number obtained on most of the scanners showed a strong linear correlation (R²> 0.99) between the CT numbers and densities of the five phantom materials. Two scanners (Siemens Emotion 16) had the strongest linear correlation withR²= 0.999, and two Hitachi Eclos scanners had the weakest linear correlation withR²< 0.99.

The effect of computed tomography parameters on sarcopenia and myosteatosis assessment: a scoping review

Computed tomography (CT) is a valuable assessment method for muscle pathologies such as sarcopenia, cachexia, and myosteatosis. However, several key underappreciated scan imaging parameters need consideration for both research and clinical use, specifically CT kilovoltage and the use of contrast material. We conducted a scoping review to assess these effects on CT muscle measures. We reviewed articles from PubMed, Scopus, and Web of Science from 1970 to 2020 on the effect of intravenous contrast material and variation in CT kilovoltage on muscle mass and density. We identified 971 articles on contrast and 277 articles on kilovoltage. The number of articles that met inclusion criteria for contrast and kilovoltage was 11 and 7, respectively. Ten studies evaluated the effect of contrast on muscle density of which nine found that contrast significantly increases CT muscle density (arterial phase 6-23% increase, venous phase 19-57% increase, and delayed phase 23-43% increase). Seven out of 10 studies evaluating the effect of contrast on muscle area found significant increases in area due to contrast (≤2.58%). Six studies evaluating kilovoltage on muscle density found that lower kilovoltage resulted in a higher muscle density (14-40% increase). One study reported a significant decrease in muscle area when reducing kilovoltage (2.9%). The use of contrast and kilovoltage variations can have dramatic effects on skeletal muscle analysis and should be considered and reported in CT muscle analysis research. These significant factors in CT skeletal muscle analysis can alter clinical and research outcomes and are therefore a barrier to clinical application unless better appreciated.

Study of the Natural Crystalline Lens Characteristics Using Dual-Energy Computed Tomography

There is a paucity of radiologic literature regarding age-related cataract, and little is known about any differences in the imaging appearance of the natural crystalline lens on computed tomography (CT) exams among different demographic groups. In this retrospective review of 198 eyes in 103 adults who underwent dual-energy computed tomography (DECT) exams of the head, regions of interest spanning 3−5 mm were placed over the center of the lens, and the x-ray attenuation of each lens was recorded in Hounsfield Units (HU) at 3 energy levels: 40 keV, 70 keV, and 190 keV. Generalized estimating equations (GEEs) were used to assess the association of clinical or demographic data with lens attenuation. The mean HU values were significantly lower for the older vs. younger group at 40 keV (GEE p-value = 0.022), but there was no significant difference at higher energy levels (p > 0.05). Mean HU values were significantly higher for females vs. males and non-whites vs. non-Hispanic whites at all 3 energy levels in bivariate and multivariable analyses (all p-value < 0.05). There was no significant association between lens attenuation and either diabetes or smoking status. The crystalline lens of females and non-whites had higher attenuation on DECT which may suggest higher density or increased concentration of materials like calcium and increased potential for cataract formation. Given the large scope of cataracts as a cause of visual impairment and the racial disparities that exist in its detection and treatment, further investigation into the role of opportunistic imaging to detect cataract formation is warranted.

Evaluation of femoral head bone quality by Hounsfield units: a comparison with dual-energy X-ray absorptiometry

BACKGROUND

Osteoporosis is associated with decreases in bone mineral density (BMD) and is diagnosed using dual-energy X-ray absorptiometry (DXA). Computed tomography (CT), performed in routine practice, can also be used to evaluate bone quality without additional cost.

PURPOSE

To determine whether Hounsfield units (HU), a standardized CT attenuation coefficient, measured from the femoral head correlated with DXA-measured BMD.

MATERIAL AND METHODS

We evaluated 82 patients (14 men, 68 women; mean age, 67 years) undergoing femoral DXA and CT (non-enhanced abdominopelvic and hip scans) with 130 kV to determine whether HU correlated with T-scores. HU were measured by two radiologists using the largest spherical region of interest including the medullary bone of the femoral head from the junction point of the most caudal section of the femoral head with the femoral neck in 5-mm axial sections. The correlations of both sides' HU values with their ages and DXA femur T-score were evaluated.

RESULTS

HU values obtained from both femoral heads showed significant variation between the osteoporotic and non-osteoporotic groups (both P = 0.000) and strongly correlated with each other and DXA femur T-scores (left r = 0.75, right r = 0.73, respectively). In ROC curve analysis, predictive power of left HU values in identifying patients with osteoporotic femur DXA T-score was 0.905, and for right HU values it was 0.924. Osteoporosis cutoff values were 198 HU and 204 HU for the left and right hips, respectively.

CONCLUSIONS

HU obtained from CT performed in routine practice correlated with the DXA scores, thus providing an alternative method to determine regional bone quality without additional cost. This may be useful when choosing a fixation method, especially in trauma cases with already-performed abdominopelvic or pelvic CT in emergency services.

Value-added Opportunistic CT Screening: State of the Art

Opportunistic CT screening leverages robust imaging data embedded within abdominal and thoracic scans that are generally unrelated to the specific clinical indication and have heretofore gone largely unused. This incidental imaging information may prove beneficial to patients in terms of wellness, prevention, risk profiling, and presymptomatic detection of relevant disease. The growing interest in CT-based opportunistic screening relates to a confluence of factors: the objective and generalizable nature of CT-based body composition measures, the emergence of fully automated explainable AI solutions, the sheer volume of body CT scans performed, and the increasing emphasis on precision medicine and value-added initiatives. With a systematic approach to body composition and other useful CT markers, initial evidence suggests that their ability to help radiologists assess biologic age and predict future adverse cardiometabolic events rivals even the best available clinical reference standards. Emerging data suggest that standalone "intended" CT screening over an unorganized opportunistic approach may be justified, especially when combined with established cancer screening. This review will discuss the current status of opportunistic CT screening, including specific body composition markers and the various disease processes that may be impacted. The remaining hurdles to widespread clinical adoption include generalization to more diverse patient populations, disparate technical settings, and reimbursement.

The Effect of COVID 19 on Liver Parenchyma Detected and measured by CT scan Hounsfield Units

BACKGROUND

COVID 19 may affect organs other than lungs, including liver, leading to parenchymal changes. These changes are best assessed by unenhanced computed tomography (CT). We aim to investigate the effect of COVID 19 on liver parenchyma by measuring the attenuation in CT scan Hounsfield unit (HU).

MATERIALS AND METHODS

A cohort of patients, who tested COVID 19 polymerase chain reaction positive, were enrolled and divided into two groups: fatty liver (FL) group (HU ≤ 40) and nonfatty liver (NFL) group (HU > 40) according to liver parenchyma attenuation measurements by high resolution noncontrast CT scan. The CT scan was performed on admission and on follow up (10-14 days later). Liver enzyme tests were submitted on admission and follow up.

RESULTS

Three hundred and two patients were enrolled. Liver HU increased significantly from 48.9 on admission to 53.4 on follow up CT scan (P<0.001) in all patients. This increase was more significant in the FL group (increased from 31.9 to 42.9 [P =0.018]) Liver enzymes were abnormal in 22.6% of the full cohort. However, there was no significant change in liver enzymes between the admission and follow up in both groups.

CONCLUSION

The use of unenhanced CT scan for assessment of liver parenchymal represents an objective and noninvasive method. The significant changes in parenchymal HU are not always accompanied by significant changes in liver enzymes. Increased HU values caused by COVID 19 may be due to either a decrease in the fat or an increase in the fibrosis in the liver.

An abdominal phantom with anthropomorphic organ motion and multimodal imaging contrast for MR-guided radiotherapy

Purpose. Improvements in image-guided radiotherapy (IGRT) enable accurate and precise treatment of moving tumors in the abdomen while simultaneously sparing healthy tissue. However, the lack of validation tools for newly developed MR-guided radiotherapy hybrid devices such as the MR-Linac is an open issue. This study presents a custom developed abdominal phantom with respiratory organ motion and multimodal imaging contrast to perform end-to-end tests for IGRT treatment planning scenarios.Methods. The abdominal phantom contains deformable and anatomically shaped liver and kidney models made of Ni-DTPA and KCl-doped agarose mixtures that can be reproducibly positioned within the phantom. Organ models are wrapped in foil to avoid ion exchange with the surrounding agarose and to provide stable T1 and T2 relaxation times as well as HU numbers. Breathing motion is realized by a diaphragm connected to an actuator that is hydraulically controlled via a programmable logic controller. With this system, artificial and patient-specific breathing patterns can be carried out. In 1.5 T magnetic resonance imaging (MRI), diaphragm, liver and kidney motion was measured and compared to the breathing motion of a healthy male volunteer for different breathing amplitudes including shallow, normal and deep breathing.Results. The constructed abdominal phantom demonstrated organ-equivalent intensity values in CT as well as in MRI. T1-weighted (T1w) and T2-weighted (T2w) relaxation times for 1.5 T and CT numbers were 552.9 ms, 48.2 ms and 48.8 HU (liver) as well as 950.42 ms, 79 ms and 28.2 HU (kidney), respectively. These values were stable for more than six months. Extracted breathing motion from a healthy volunteer revealed a liver to diaphragm motion ratio (LDMR) of 64.4% and a kidney to diaphragm motion ratio (KDMR) of 30.7%. Well-comparable values were obtained for the phantom (LDMR: 65.5%, KDMR: 27.5%).Conclusions. The abdominal phantom demonstrated anthropomorphic T1 and T2 relaxation times as well as HU numbers and physiological motion pattern in MRI and CT. This allows for wide use in the validation of IGRT including MRgRT.

Virtual Unenhanced Images: Qualitative and Quantitative Comparison Between Different Dual-Energy CT Scanners in a Patient and Phantom Study

MATERIALS AND METHODS

Forty-four patients with clinical contrast-enhanced abdominal examinations on each of the 3 DECT scanner types and a phantom scanned with the same protocols were included in this retrospective study. Qualitative and quantitative assessment was performed on VUE images. Quantitative evaluation included measurement of attenuation and image noise for various tissues and the phantom. Virtual unenhanced image attenuation and noise were compared between scanner types, and intrapatient interscanner reproducibility of virtual unenhanced image attenuation was calculated as the percentage of measurement pairs with an interscanner difference ≤ 10 HU. Image quality, noise, sharpness, and iodine elimination were assessed qualitatively by 2 radiologists.

RESULTS

Significant interscanner differences in VUE attenuation and noise were found in all tissues. dlDECT and rsDECT showed significantly higher VUE attenuation than dsDECT in the aorta, portal vein, and kidneys (P < 0.05). Conversely, VUE attenuation in dsDECT was significantly higher than in dlDECT/rsDECT for subcutaneous and retroperitoneal fat (both P < 0.05). A total of 91.9% (385/419) of measurements were reproducible between rsDECT and dlDECT, 70.9% (297/419) between dsDECT and rsDECT, and 66.8% (280/419) between dsDECT and dlDECT. Virtual unenhanced image attenuation in the contrast media-filled phantom cavity was 12.7 ± 4.7 HU in dlDECT, -5.3 ± 4.2 HU in rsDECT, and -4.0 ± 10.7 HU in dsDECT with significant differences between dlDECT and rsDECT/dsDECT, respectively (P < 0.05), between which attenuation was comparable in the unenhanced extraluminal phantom component (P = 0.11-0.62). Qualitatively, dsDECT yielded best iodine elimination, whereas sharpness, image noise, and overall image quality were rated higher in dlDECT and rsDECT.

CONCLUSIONS

There are significant interscanner differences in the attenuation measurements and qualitative assessment of VUE images, which should be acknowledged when using these images in patients that are being scanned on different DECT scanner types during imaging follow-up.

A Metric for Quantification of Iodine Contrast Enhancement (Q-ICE) in Computed Tomography

BACKGROUND

Poor contrast enhancement is related to issues with examination execution, contrast prescription, computed tomography (CT) protocols, and patient conditions. Currently, our community has no metric to monitor true enhancement on routine single-phase examinations because this requires knowledge of both pre- and postcontrast CT number.

PURPOSE

We propose an automatable solution to quantifying contrast enhancement without requiring a dedicated noncontrast series.

METHODS

The difference in CT number between a target region in an enhanced and unenhanced image defines the metric "quantification of iodine contrast enhancement" (Q-ICE). Quantification of iodine contrast enhancement uses the noncontrast bolus tracking baseline image from routine abdominal examinations, which mitigates the need for a dedicated noncontrast series. We applied this method retrospectively to 312 patient livers from 2 sites between 2017 and 2020. Each site used a weight-based contrast injection protocol for weights 60 to 113 kg and a constant volume less than 60 kg and greater than 113 kg. Hypothesis testing was performed to compare Q-ICE between sites and detect Q-ICE dependence on weight and kilovoltage (kV).

RESULTS

Mean Q-ICE differed between sites (P = 0.004) by 4.96 Hounsfield unit with 95% confidence interval (1.63-8.28), albeit this difference was roughly 2 times smaller than the SD in Q-ICE across patients at a single site. For patients between 60 and 113 kg, we did not observe evidence of Q-ICE varying with patient weight (P = 0.920 and 0.064 for 120 and 140 kV, respectively). The Q-ICE did vary with patient weight for patients less than 60 kg (P = 0.003) and greater than 113 kg (P = 0.04). We observed a roughly 10 Hounsfield unit reduction in Q-ICE liver for patients scanned with 140 versus 120 kV. We observed several underenhancing examinations with an arterial phase appearance motivating our CT protocol optimization team to consider increasing the delay for slowly enhancing patients.

CONCLUSIONS

A quality metric for quantifying CT contrast enhancement was developed and suggested tangible opportunities for quality improvement and potential financial savings.

Quantitative Analysis of Deltoid Ligament Degradation in Patients With Chronic Ankle Instability Using Computed Tomographic Images

BACKGROUND

Rotational ankle instability (RAI) is associated with the faster onset of severe ankle osteoarthritis via dysfunction of the anterior talofibular ligament, calcaneofibular ligament, and deltoid ligament. No specific clinical examination is available for RAI, and diagnostic imaging has limitations in evaluating ligament degradation. This study investigated the deltoid ligament degeneration using Hounsfield unit (HU) values on computed tomography (CT) images.

METHODS

Patients were enrolled in this retrospective analysis if they had undergone magnetic resonance imaging (MRI) and CT scans of the ankle. The chronic ankle instability (CAI) group comprised 20 ankles with CAI (9 men, 11 women; mean age, 28.7 years) and the control group comprised 28 ankles (16 men, 12 women, mean age, 41.3 years). The average HU values of the deep posterior tibiotalar ligament (dPTL) that constitutes the deltoid ligament were measured on coronal CT images, and MRI results were used as a reference. All patients were subdivided based on the MRI findings of dPTL injury such as fascicular disruption, irregularity, and the loss of striation.

RESULTS

A strong negative correlation was identified between age and HU values for all patients (Spearman ρ = -0.63; P < .001). The mean HU values of the dPTL for participants aged <60 years were 81.0 HU for the control group (21 ankles) and 69.5 HU for the CAI group (P = .0075). No significant differences in the HU values were observed for the dPTL among the MRI subgroups.

CONCLUSION

In addition to the conventional imaging examination such as stress radiographs and MRI, HU measurements of CT images could be useful for quantitatively and noninvasively evaluating degenerative changes in the deltoid ligament for CAI patients to assist the diagnosis of RAI.

LEVEL OF EVIDENCE

Level III. case-control study.

Histopathological and Radiographic Features of Osteolysis After Fixation of Osteochondral Fragments Using Poly-L-Lactic Acid Pins for Osteochondral Lesions of the Talus

BACKGROUND

Fixation of osteochondral fragments is a potential option for treating an osteochondral lesion of the talus (OLT) involving large lesions in the remaining articular cartilage surface. Bioabsorbable devices, especially those made of poly-L-lactic acid (PLLA), can be used for the fixation of an OLT. Postoperative osteolysis surrounding the PLLA pins is occasionally observed; however, the significance of osteolysis remains unknown.

PURPOSE

To elucidate the association between osteolysis surrounding the PLLA pins, histopathological findings in subchondral bone, and preoperative Hounsfield unit (HU) values at the pin fixation site.

STUDY DESIGN

Case Series; Level of evidence, 4.

METHODS

This retrospective analysis included 20 patients with OLT (11 men and 9 women; mean age, 20.9 years; 1 bilateral case). Tissue from the osteochondral fragment was collected intraoperatively using a bone biopsy needle for histological evaluation. The fragment was fixed through the biopsy hole using a PLLA pin. Osteolysis surrounding the PLLA pin was assessed at 1 year postoperatively using magnetic resonance imaging (MRI). Histopathological scores were assigned based on trabecular bone loss, empty lacunae, inflammatory granulation tissue, cartilage-like tissue, and the presence of osteoclasts. The HU values around the pin insertion site, detected on the postoperative MRI scans, were measured using the region of interest based on the preoperative coronal and sagittal computed tomography (CT) images.

RESULTS

Osteolysis was observed postoperatively in 9 ankles (42.9%). Histopathological evaluation revealed that the osteolysis group had a significantly higher pathological score than the nonosteolysis group (10.2 vs 6.3; P < .001). Lower HU values were identified in the osteolysis group on preoperative coronal and sagittal CT images (P < .05). The histopathological score negatively correlated with preoperative HU values (Pearson r = -0.46; P = .037).

CONCLUSION

Intraoperative biopsy of the OLT allowed for histopathological evaluation of the same site as that of the PLLA pin fixation. Our findings suggest that preoperative subchondral trabecular deterioration is associated with the incidence of postoperative osteolysis surrounding the PLLA pin. Additionally, low preoperative HU values in subchondral bone under OLT may serve as a predictor of osteolysis surrounding the PLLA pin.

Challenges estimating patient organs doses undergoing enhanced chest CT examination: exploratory study

Purpose: Estimate organs doses (ODs) of patients subjected to unenhanced (S1) and enhanced (S2) chest CT studies relying on image parameters such as Hounsfield Units (HUs).Materials and Methods: CT scans and images of a total of 16 patients who underwent two series of chest CT studies were obtained and retrospectively examined. OD increments of liver and pancreas for both series (S1 & S2) were estimated using two different independent methods, namely simulation approach using CT-EXPO and Amato's phantom-based fitting model (APFM). HUs were quantified for each organ by manually drawing fixed area-sized regions of interest (ROIs). The mean HUs were collected to obtain the ODs increments following APFM. Regression analysis was applied to find and assess the relationship between the HUs and the OD increments estimated using APFM and that using CT-EXPO. Spearman Coefficient and Wilcoxon Matched Pairedt-testwere conducted to show statistical correlation and difference between ODs increments using the two methods.Results:A strong significant difference was depicted between S1 and S2 scan series of liver and pancreas using CT-EXPO simulation. Mean HU values for S1 were lower than S2, resulting in statistically significant (p < 0.0001) HU changes. CT-EXPO simulation yielded significantly higher difference in ODs compared to the APFM for liver (p = 0.0455) and pancreas (p = 0.0031). Regression analysis revealed a strong relationship between HU of S1 and S2 and ODs increments using APFM in both organs (R² = 0.99), dissimilar to CT-EXPO (R² = 0.39 in liver andR² = 0.05 in pancreas).Conclusions: Although CT-EXPO allows for estimating ODs accounting for major acquisition scan parameters, it is not a reliable tool to evaluate the impact of contrast enhancement on ODs. On the other hand, the APFM accounts for contrast enhancement accumulation yet only provides relative OD increments, an information of limited clinical use.

Whole-Body Low-Dose CT in Multiple Myeloma: Diagnostic Value of Appendicular Medullary Patterns of Attenuation

OBJECTIVE. The purpose of this article is to analyze whole-body low-dose CT-detected appendicular medullary patterns of attenuation in patients with newly diagnosed multiple myeloma and to determine the diagnostic performance of whole-body low-dose CT in detecting diffuse marrow infiltration. MATERIALS AND METHODS. A total of 76 patients with myeloma who underwent whole-body low-dose CT and spinal MRI at initial assessment were retrospectively analyzed. The medullary cavities of femurs and humeri were evaluated qualitatively and quantitatively on CT. Medullary attenuation and SD-to-mean attenuation ratio were recorded for each long bone. The pattern of marrow involvement on spinal MRI was used as reference. The chi-square test was used to evaluate the relationship between the CT-based appendicular medullary cavity pattern and the MRI pattern, and ROC analysis was performed to assess the diagnostic accuracy of CT attenuation measurements for the differentiation between diffuse and mixed CT-based appendicular medullary cavity patterns. RESULTS. Medullary attenuation differed significantly among mixed, nodular, and diffuse CT-based appendicular medullary cavity patterns in the femurs (mean, 34.23 HU and range, 15-61 HU; mean, 66.26 HU and range, 26-104 HU; mean, 92.80 HU and range, 53-127 HU, respectively) and humeri (mean, 22.18 HU and range, 9-41; mean, 61.18 HU and range, 23-93 HU; mean, 77.50 and range, 25-105 HU, respectively). To discriminate between diffuse and mixed CT-based appendicular medullary cavity patterns, optimal cutoff attenuation values were 63 HU (sensitivity, 97.7%; specificity, 100.0%) for the femurs, and 52 HU (sensitivity, 97.4%; specificity, 100.0%) for the humeri. A total of 24 of 30 (80.0%) patients with a diffuse MRI pattern showed a diffuse CT-based appendicular medullary cavity pattern on whole-body low-dose CT, and all patients with a diffuse CT-based appendicular medullary cavity pattern also showed a diffuse pattern on MRI. CONCLUSION. According to analysis of peripheral medullary patterns of attenuation, whole-body low-dose CT can identify patients with multiple myeloma with diffuse marrow involvement.

Machine-learning-based multiple abnormality prediction with large-scale chest computed tomography volumes

Machine learning models for radiology benefit from large-scale data sets with high quality labels for abnormalities. We curated and analyzed a chest computed tomography (CT) data set of 36,316 volumes from 19,993 unique patients. This is the largest multiply-annotated volumetric medical imaging data set reported. To annotate this data set, we developed a rule-based method for automatically extracting abnormality labels from free-text radiology reports with an average F-score of 0.976 (min 0.941, max 1.0). We also developed a model for multi-organ, multi-disease classification of chest CT volumes that uses a deep convolutional neural network (CNN). This model reached a classification performance of AUROC >0.90 for 18 abnormalities, with an average AUROC of 0.773 for all 83 abnormalities, demonstrating the feasibility of learning from unfiltered whole volume CT data. We show that training on more labels improves performance significantly: for a subset of 9 labels - nodule, opacity, atelectasis, pleural effusion, consolidation, mass, pericardial effusion, cardiomegaly, and pneumothorax - the model's average AUROC increased by 10% when the number of training labels was increased from 9 to all 83. All code for volume preprocessing, automated label extraction, and the volume abnormality prediction model is publicly available. The 36,316 CT volumes and labels will also be made publicly available pending institutional approval.

Assessment of Hounsfield unit in the differential diagnosis of odontogenic cysts

OBJECTIVES

The purpose of this study was to evaluate the usefulness of Hounsfield unit (HU) assessment with multislice-CT in the differentiation of radicular cysts (RCs), dentigerous cysts (DCs) and odontogenic keratocysts (OKCs).

METHODS

In total, 307 odontogenic cysts (RCs, DCs and OKCs) were included in this study. Cysts with lesion diameter <10 mm, cysts with artefacts affecting measurement of HU values, cysts involving infection and recurrent cysts were regarded as exclusion criteria. Images were acquired in three different types of CT scanners: Aquilion ONE, Discovery CT750 HD and SOMATOM Definition Flash. Differences in HU values among scanners and among types of odontogenic cysts were assessed using one-way analysis of variance; multiple comparisons were performed post hoc, using the Tukey-Kramer honestly significant difference test.

RESULTS

In total, 164 cysts were analysed in this study (64 RCs, 57 DCs and 43 OKCs). Regardless of the type of lesion, the Aquilion ONE scanner demonstrated a significant difference in HU value, compared with the Discovery CT750 HD scanner. Regardless of CT scanner model, HU values significantly differed between DCs and OKCs (p < 0.0001), as well as between OKCs and RCs (p < 0.0001).

CONCLUSIONS

HU values were found to vary among CT scanners and should always be associated with other lesion imaging features while interpreting and elaboration diagnostic hypothesis. Notably, the results suggested that OKCs might be able to be differentiated from DCs and RCs by using HU values.

Electron Density and Biologically Effective Dose (BED) Radiomics-Based Machine Learning Models to Predict Late Radiation-Induced Subcutaneous Fibrosis

Purpose: to predict the occurrence of late subcutaneous radiation induced fibrosis (RIF) after partial breast irradiation (PBI) for breast carcinoma by using machine learning (ML) models and radiomic features from 3D Biologically Effective Dose (3D-BED) and Relative Electron Density (3D-RED). Methods: 165 patients underwent external PBI following a hypo-fractionation protocol consisting of 40 Gy/10 fractions, 35 Gy/7 fractions, and 28 Gy/4 fractions, for 73, 60, and 32 patients, respectively. Physicians evaluated toxicity at regular intervals by the Common Terminology Adverse Events (CTAE) version 4.0. RIF was assessed every 3 months after the completion of radiation course and scored prospectively. RIF was experienced by 41 (24.8%) patients after average 5 years of follow up. The Hounsfield Units (HU) of the CT-images were converted into relative electron density (3D-RED) and Dose maps into Biologically Effective Dose (3D-BED), respectively. Shape, first-order and textural features of 3D-RED and 3D-BED were calculated in the planning target volume (PTV) and breast. Clinical and demographic variables were also considered (954 features in total). Imbalance of the dataset was addressed by data augmentation using ADASYN technique. A subset of non-redundant features that best predict the data was identified by sequential feature selection. Support Vector Machines (SVM), ensemble machine learning (EML) using various aggregation algorithms and Naive Bayes (NB) classifiers were trained on patient dataset to predict RIF occurrence. Models were assessed using sensitivity and specificity of the ML classifiers and the area under the receiver operator characteristic curve (AUC) of the score functions in repeated 5-fold cross validation on the augmented dataset. Results: The SVM model with seven features was preferred for RIF prediction and scored sensitivity 0.83 (95% CI 0.80-0.86), specificity 0.75 (95% CI 0.71-0.77) and AUC of the score function 0.86 (0.85-0.88) on cross-validation. The selected features included cluster shade and Run Length Non-uniformity of breast 3D-BED, kurtosis and cluster shade from PTV 3D-RED, and 10th percentile of PTV 3D-BED. Conclusion: Textures extracted from 3D-BED and 3D-RED in the breast and PTV can predict late RIF and may help better select patient candidates to exclusive PBI.

Cross-anatomical, radiographic and computed tomographic study of the stifle joint of donkeys (Equus africanus asinus)

The present investigation was conducted to provide a full anatomical description of the stifle joint of donkeys using 3D computed tomography imaging technique, in addition to the classic anatomical methods, such as radiography and cross-anatomical sectioning. The radiography and CT imaging of stifle joint were interpreted in comparison with cross-sectional anatomical sections. Volume-rendering reconstruction techniques (3D-CT) were used to describe the anatomical structure of stifle joint. The used twelve adult healthy donkeys were free from any musculoskeletal disorders. Four donkeys were used for the gross anatomical observations, four for CT and radiography and two live animals for determination the site of injections. The results of this study revealed that the complex stifle joint was formed from three joints: femorotibial, femoropatellar and proximal tibiofibular. The articular surfaces were described for each joint, and the synovial layer of the articular capsule formed three main joint sacs: femoropatellar, medial femorotibial and lateral femorotibial sacs. The ligaments of stifle joint were recorded, and meniscal ligaments included cranial and caudal ligaments of medial and lateral menisci and meniscofemoral ligament of lateral meniscus. The cruciate ligaments were also described and they included the cranial and caudal cruciate ligaments, while the patellar ligament included the medial, middle and lateral patellar ligaments. The arterial supply and the site of injection of the stifle joint were described. In conclusion, the 3D reconstruction CT provided well-defined baseline reference image for the stifle joint of donkeys for anatomist, radiologist, surgeons and researchers.

Dose reduction for CT coronary calcium scoring with a calcium-aware image reconstruction technique: a phantom study

Objective

To assess the dose reduction potential of a calcium-aware reconstruction technique, which aims at tube voltage-independent computed tomography (CT) numbers for calcium.

Methods and materials

A cardiothoracic phantom, mimicking three different patient sizes, was scanned with two calcium inserts (named D100 and CCI), containing calcifications varying in size and density. Tube voltage was varied both manually (range 70–150 and Sn100 kVp) and automatically. Tube current was automatically adapted to maintain reference image quality defined at 120 kVp. Data was reconstructed with the standard reconstruction technique (kernel Qr36) and the calcium-aware reconstruction technique (kernel Sa36). We assessed the radiation dose reduction potential (volumetric CT dose index values (CTDIvol)), noise (standard deviation (SD)), mean CT number (HU) of each calcification, and Agatston scores for varying kVp. Results were compared with the reference acquired at 120 kVp and reconstructed with Qr36.

Results

Automatic selection of the optimal tube voltage resulted in a CTDIvol reduction of 22%, 15%, and 12% compared with the reference for the small, medium, and large phantom, respectively. CT numbers differed up to 64% for the standard reconstruction and 11% for the calcium-aware reconstruction. Similarly, Agatston scores deviated up to 40% and 8% for the standard and calcium-aware reconstruction technique, respectively.

Conclusion

CT numbers remained consistent with comparable calcium scores when the calcium-aware image reconstruction technique was applied with varying tube voltage. Less consistency was observed in small calcifications with low density. Automatic reduction of tube voltage resulted in a dose reduction of up to 22%.

Key Points

• The calcium-aware image reconstruction technique allows for consistent CT numbers when varying the tube voltage.

• Automatic reduction of tube voltage results in a reduced radiation exposure of up to 22%.

• This study stresses the known limitations of the current Agatston score technique.

Electronic supplementary material

The online version of this article (10.1007/s00330-020-06709-9) contains supplementary material, which is available to authorized users.

3D printed CT-based abdominal structure mannequin for enabling research

An anthropomorphic phantom is a radiologically accurate, tissue realistic model of the human body that can be used for research into innovative imaging and interventional techniques, education simulation and calibration of medical imaging equipment. Currently available CT phantoms are appropriate tools for calibration of medical imaging equipment but have major disadvantages for research and educational simulation. They are expensive, lacking the realistic appearance and characteristics of anatomical organs when visualized during X-ray based image scanning. In addition, CT phantoms are not modular hence users are not able to remove specific organs from inside the phantom for research or training purposes. 3D printing technology has evolved and can be used to print anatomically accurate abdominal organs for a modular anthropomorphic mannequin to address limitations of existing phantoms. In this study, CT images from a clinical patient were used to 3D print the following organ shells: liver, kidneys, spleen, and large and small intestines. In addition, fatty tissue was made using modelling beeswax and musculature was modeled using liquid urethane rubber to match the radiological density of real tissue in CT Hounsfield Units at 120kVp. Similarly, all 3D printed organ shells were filled with an agar-based solution to mimic the radiological density of real tissue in CT Hounsfield Units at 120kVp. The mannequin has scope for applications in various aspects of medical imaging and education, allowing us to address key areas of clinical importance without the need for scanning patients.

A novel computed tomography image synthesis method for correcting the spectrum dependence of CT numbers

The quantitative evaluation of computed tomography (CT) images is widely investigated and applied in clinical diagnosis. However, the CT number of tissue can vary with scanners or applied tube voltages because of the x-ray spectrum dependence of measured linear attenuation coefficients that degrades evaluation accuracy and limits multicenter or multimodality research. This study proposed a novel CT image synthesis method to correct the spectrum dependence of CT numbers by normalizing them to the same spectrum condition. Stoichiometric calibration was performed to derive the spectrum characteristic parameters (SCPs) of six spectra from two CT scanners with different applied tube voltages. Subsequently, conversion relationships between CT numbers and tissue parameters (TPs) were determined using the SCPs and standard tissue data. The CT number of a tissue measured from a spectrum condition was converted to TPs using these relationships, and the results were used to estimate the CT number of the tissue in another spectrum condition using the corresponding SCPs. Phantom, cadaver, and patient studies were performed to evaluate the proposed method. In the phantom study, image synthesis reduced the mean difference between the CT numbers of tissue-equivalent phantoms measured using different spectra from 57.96 to 33.94 HU. In the cadaveric study, the mean difference between the CT numbers of a temporal bone flap measured using different spectra was lowered by over 57%. In the patient image study, a significant difference of 81.5 HU was observed between the mean CT numbers of femoral shafts obtained from the two scanners; this difference was reduced to less than 17 HU, which was nonsignificant, when the proposed method was used. The proposed image synthesis method could reduce the spectrum dependence of CT numbers measured with different spectra and could be applied clinically to improve the accuracy of multicenter and multimodality evaluation and research.

Experiences with image quality and radiation dose of cone beam computed tomography (CBCT) and multidetector computed tomography (MDCT) in pediatric extremity trauma

INTRODUCTION

Novel dedicated extremity cone beam computed tomography (CBCT) devices, recently introduced to the market, raised attention as a possible alternative in advanced diagnostic pediatric trauma imaging, today usually performed by multidetector computed tomography (MDCT). This work aimed to compare image quality and radiation dose of CBCT and MDCT.

MATERIALS AND METHODS

Fifty-four CBCT-MDCT examination pairs, containing nine MDCTs acquired in parallel prospectively and 45 MDCTs matched in retrospect, were included in this study. Image quality was analyzed semi-objectively by measuring noise, contrast-to-noise ratio (CNR), and signal-to-noise ratios (SNR) and subjectively by performing image impression ratings. CT dose records were readout.

RESULTS

Image noise was significantly lower in CBCT compared with MDCT, both semi-objectively and subjectively (both p < 0.001). CNR and SNRs were also in favor of CBCT, though CBCT examinations exhibited significantly more beam hardening artifacts that diminished the advantages of the superior semi-objective image quality. These artifacts were believed to occur more often in children due to numerous bone-cartilage transitions in open growth plates and may have led to a better subjective diagnostic certainty rating (p = 0.001). Motion artifacts were infrequently, but exclusively observed in CBCT. CT dose index (CTDI_vol) was substantially lower in CBCT (p < 0.001).

CONCLUSION

Dedicated extremity CBCT could be an alternative low-dose modality in the diagnostic pathway of pediatric fractures. At lower doses compared with MDCT and commonly affected by beam hardening artifacts, semi-objective CBCT image quality parameters were generally better than in MDCT.

Effects of energy level, reconstruction kernel, and tube rotation time on Hounsfield units of hydroxyapatite in virtual monochromatic images obtained with dual-energy CT

Purpose

This study was performed to investigate the effects of energy level, reconstruction kernel, and tube rotation time on Hounsfield unit (HU) values of hydroxyapatite (HA) in virtual monochromatic images (VMIs) obtained with dual-energy computed tomography (DECT) (Siemens Healthineers, Erlangen, Germany).

Materials and Methods

A bone density calibration phantom with 3 HA inserts of different densities (CTWATER®; 0, 100, and 200 mg of HA/cm³) was scanned using a twin-beam DECT scanner at 120 kVp with tube rotation times of 0.5 and 1.0 seconds. The VMIs were reconstructed by changing the energy level (with options of 40 keV, 70 keV, and 140 keV). In order to investigate the impact of the reconstruction kernel, virtual monochromatic images were reconstructed after changing the kernel from body regular 40 (Br40) to head regular 40 (Hr40) in the reconstruction phase. The mean HU value was measured by placing a circular region of interests (ROIs) in the middle of each insert obtained from the VMIs. The HU values were compared with regard to energy level, reconstruction kernel, and tube rotation time.

Results

Hydroxyapatite density was strongly correlated with HU values (correlation coefficient=0.678, P<0.05). For the HA 100 and 200 inserts, HU decreased significantly at increased energy levels (correlation coefficient= −0.538, P<0.05) but increased by 70 HU when using Hr40 rather than Br40 (correlation coefficient=0.158, P<0.05). The tube rotation time did not significantly affect the HU (P>0.05).

Conclusion

The HU values of hydroxyapatite were strongly correlated with hydroxyapatite density and energy level in VMIs obtained with DECT.