On the relationship between computed tomography numbers and specific gravity

Assessment of a simple method of heart weight estimation by postmortem computed tomography

BACKGROUND

Measurement of heart weight is important when investigating cause of death, but there is presently no satisfactory method of heart weight estimation by postmortem computed tomography (PMCT).

METHOD

We investigated 33 consecutive cases that underwent both PMCT and autopsy between February 2008 and June 2014. Heart and left ventricular (LV) weights were calculated by PMCT morphometry. We used a simple method to estimate LV weight: We assumed that LV was an ellipsoid and multiplied its volume on PMCT with myocardial specific gravity. We then compared the various heart and LV weights using linear regression. The calculated and estimated LV weights on PMCT were also compared.

RESULTS

It was not possible to predict heart weight at autopsy from PMCT (R² = 0.53). However, heart weight at autopsy could be accurately predicted from LV weight calculated by PMCT (R² = 0.77). In addition, there was a strong correlation between the estimated and calculated LV weights by PMCT (R² = 0.92). Heart weight at autopsy could also be accurately predicted using the PMCT-estimated LV weight (R² = 0.72).

CONCLUSION

Heart weight at autopsy could be accurately predicted using a simple method in which LV volume was assumed to be an ellipsoid on PMCT.

Attenuation Values, Volume Changes and Artifacts in Tissue Due to Freezing

Freezing of animal and human tissue may change their shape, size and other physical properties such as attenuation of roentgen rays. In correlative radiologic-anatomic studies such artifacts are potential sources of inaccuracy. In fluid phantoms changes of computed tomography (CT) numbers and volume were measured before and after freezing. Frozen saline decreased in attenuation by 80 Hounsfield units (HU) and increased in volume by 9 per cent. Pure soy oil increased in attenuation by 40 HU after freezing and shrank in volume by 4 per cent. Emulsions of 10, 20 and 30 per cent lipid expanded in inverse relationship to the changes in attenuation. In animal and human tissue, changes in CT number of −80 to + 55 HU were measured and changes in volume of —6 to +9 per cent were calculated. The attenuation of the vitreous body in the eye decreased by 80 HU and its volume was calculated to increase by 8 to 9 per cent. In the muscle and spinal cord the attenuation decreased by 40 to 60 HU, corresponding to the increase in volume of 4 to 6 per cent. Attenuation of fat was increased by 55 HU after freezing and a decrease in volume of 5 to 6 per cent was calculated. Deformation of soft tissue and herniation of the cerebellar tonsils into the foramen magnum were observed. Most anatomic, freezing artifacts can be avoided by careful preparation.

Cardiac dilatation index as an indicator of terminal central congestion evaluated using postmortem CT and forensic autopsy data

Previous studies demonstrated possible application of postmortem quantitative CT data analysis of the heart and lung in situ to investigate terminal cardiopulmonary pathophysiology. The present study analyzed virtual CT morphometric and autopsy data of the heart to investigate terminal central congestion in forensic autopsy cases (n=113, within 3 days postmortem); the virtual total heart weight in situ was estimated using CT morphometry, and the difference from and ratio to the measured weight at autopsy were calculated as indicators of heart blood pooling and the cardiac dilatation index (CDI) before dissection, respectively. There were substantial differences between the estimated heart blood pooling in situ and volume recovered at autopsy, including a characteristic decrease in drowning, alcohol/sedative-hypnotic intoxication and sudden cardiac death (SCD), possibly due to blood redistribution after thoracic dissection. The estimated in situ heart blood pool and CDI values were higher in SCD but lower in fatal hemorrhage and hemopericardium, as well as in acute mechanical asphyxiation and hyperthermia (heatstroke). In addition, there was a significant difference in heart blood pooling between mechanical asphyxiation or drowning and SCD. The CDI was significantly lower in fatal hyperthermia (heatstroke) than in drowning, fatal methamphetamine abuse, alcohol/sedative-hypnotic intoxication and SCD. These findings suggest the usefulness of applying the CDI and postmortem heart blood volume in situ as supplementary indicators of terminal central congestion, especially for investigating deaths from hemorrhage, hemopericardium, hyperthermia (heatstroke) and SCD.

A new multi-object image thresholding method based on correlation between object class uncertainty and intensity gradient

PURPOSE

Image thresholding and gradient analysis have remained popular image preprocessing tools for several decades due to the simplicity and straight-forwardness of their definitions. Also, optimum selection of threshold and gradient strength values are hidden steps in many advanced medical imaging algorithms. A reliable method for threshold optimization may be a crucial step toward automation of several medical image based applications. Most automatic thresholding and gradient selection methods reported in literature primarily focus on image histograms ignoring a significant amount of information embedded in the spatial distribution of intensity values forming visible features in an image. Here, we present a new method that simultaneously optimizes both threshold and gradient values for different object interfaces in an image that is based on unification of information from both the histogram and spatial image features; also, the method works for unknown number of object regions.

METHODS

A new energy function is formulated by combining the object class uncertainty measure, a histogram-based feature, of each pixel with its image gradient measure, a spatial contextual feature in an image. The energy function is designed to measure the overall compliance of the theoretical premise that, in a probabilistic sense, image intensities with high class uncertainty are associated with high image gradients. Finally, it is expressed as a function of threshold and gradient parameters and optimum combinations of these parameters are sought by locating pits and valleys on the energy surface. A major strength of the algorithm lies in the fact that it does not require the number of object regions in an image to be predefined.

RESULTS

The method has been applied on several medical image datasets and it has successfully determined both threshold and gradient parameters for different object interfaces even when some of the thresholds are almost impossible to locate in the histogram. Both accuracy and reproducibility of the method have been examined on several medical image datasets including repeat scan 3D multidetector computed tomography (CT) images of cadaveric ankles specimens. Also, the new method has been qualitatively and quantitatively compared with Otsu's method along with three other algorithms based on minimum error thresholding, maximum segmented image information and minimization of homogeneity- and uncertainty-based energy and the results have demonstrated superiority of the new method.

CONCLUSIONS

We have developed a new automatic threshold and gradient strength selection algorithm by combining class uncertainty and spatial image gradient features. The performance of the method has been examined in terms of accuracy and reproducibility and the results found are better as compared to several popular automatic threshold selection methods.

An integrated Monte Carlo dosimetric verification system for radiotherapy treatment planning

An integrated Monte Carlo (MC) dose calculation system, MCRTV (Monte Carlo for radiotherapy treatment plan verification), has been developed for clinical treatment plan verification, especially for routine quality assurance (QA) of intensity-modulated radiotherapy (IMRT) plans. The MCRTV system consists of the EGS4/PRESTA MC codes originally written for particle transport through the accelerator, the multileaf collimator (MLC), and the patient/phantom, which run on a 28-CPU Linux cluster, and the associated software developed for the clinical implementation. MCRTV has an interface with a commercial treatment planning system (TPS) (Eclipse, Varian Medical Systems, Palo Alto, CA, USA) and reads the information needed for MC computation transferred in DICOM-RT format. The key features of MCRTV have been presented in detail in this paper. The phase-space data of our 15 MV photon beam from a Varian Clinac 2300C/D have been developed and several benchmarks have been performed under homogeneous and several inhomogeneous conditions (including water, aluminium, lung and bone media). The MC results agreed with the ionization chamber measurements to within 1% and 2% for homogeneous and inhomogeneous conditions, respectively. The MC calculation for a clinical prostate IMRT treatment plan validated the implementation of the beams and the patient/phantom configuration in MCRTV.

A CT-based Monte Carlo simulation tool for dosimetry planning and analysis

The Los Alamos code MCNP4A (Monte Carlo N-Particle version 4A) is currently used to simulate a variety of problems ranging from nuclear reactor analysis to boron neutron capture therapy. A graphical user interface has been developed that automatically sets up the MCNP4A geometry and radiation source requirements for a three-dimensional Monte Carlo simulation using computed tomography data. The major drawback for this dosimetry system is the amount of time to obtain a statistically significant answer. A specialized patch file has been developed that optimizes photon particle transport and dose scoring within the standard MCNP4A lattice geometry. The transport modifications produce a performance increase (number of histories per minute) of approximately 4.7 based upon a 6 MV point source centered within a 30 x 30 x 30 cm3 lattice water phantom and 1 x 1 x 1 mm3 voxels. The dose scoring modifications produce a performance increase of approximately 470 based upon a tally section of greater than 1 x 10(4) lattice elements and a voxel size of 5 mm3. Homogeneous and heterogeneous benchmark calculations produce good agreement with measurements using a standard water phantom and a high- and low-density heterogeneity phantom. The dose distribution from a typical mediastinum treatment planning setup is presented for qualitative analysis and comparison versus a conventional treatment planning system.

Segmental inertial parameters of the human trunk as determined from computed tomography

This study used computed tomography (CT) imaging to determine in vivo mass, center of mass (CM), and moments of inertia (Icm) about the CM of discrete segments of the human torso. Four subjects, two males and two females, underwent serial transverse CT scans that were collected at 1-cm intervals for the full length of the trunk. The pixel intensity values of transverse images were correlated to tissue densities, thereby allowing trunk section mass properties to be calculated. The percentage of body mass observed by vertebral levels ranged from 1.1% at T1 to 2.6% at L5. The masses of the upper, middle, and lower trunk segments as percentages of body mass were estimated to be 18.5, 12.2, and 10.7%, respectively. The whole trunk mass was estimated to comprise 41.6% of the total body mass. Transverse vertebral CM values were found to lie anterior to their respective vertebral centroids by up to 5.0 cm in the lower thoracic region. For the upper, middle, and lower trunk segments, the average CM positions were found to be 25.9, 62.5, and 86.9% of the distance from the superior to inferior ends of the trunk. The upper and middle trunk CMs corresponded to approximately 4.0 cm anterior to T7/T8 vertebral centroid levels and 1.0 cm anterior to L3/L4 vertebral centroid levels, respectively. For the whole trunk, the CM was 52.7% of the distance from the xiphoid process and approximately 2.0 cm anterior to L1/L2 vertebral centroid levels. Variations in CM and Icm values were observed between subject, but these were within the range of previous reports of body segment parameters. Differences from previous studies were attributable to variations in boundary definitions, measurement techniques, population groups, and body states (live versus cadaver) examined. The disparity between previous findings and findings of this study emphasizes the need to better define the segmental properties of the trunk so that improved biomechanical representation of the body can be achieved.

In vivo tissue characterization using quantitative computed tomography: a review

Over the past 15 years many attempts have been made to make use of the quantitative information contained within computed tomography (CT) scanner images. A survey of the various approaches to quantitative computed tomography (QCT) is reported. The technical limitations of QCT are discussed. It is concluded that the measurement of bone mineral content is currently the only clinically well-established QCT technique.

Influence of radiation therapy on the lung-tissue in breast cancer patients: CT-assessed density changes and associated symptoms

The relative electron density of lung tissue was measured from computer tomography (CT) slices in 33 breast cancer patients treated by various techniques of adjuvant radiotherapy. The measurements were made before radiotherapy, 3 months and 9 months after completion of radiation therapy. The changes in lung densities at 3 months and 9 months were compared to radiation induced radiological (CT) findings. In addition, subjective symptoms such as cough and dyspnoea were assessed before and after radiotherapy. It was observed that the mean of the relative electron density of lung tissue varied from 0.25 when the whole lung was considered to 0.17 when only the anterior lateral quarter of the lung was taken into account. In patients with positive radiological (CT) findings the mean lung density of the anterior lateral quarter increased 2.1 times 3 months after radiotherapy and was still increased 1.6 times 6 months later. For those patients without findings, in the CT pictures the corresponding values were 1.2 and 1.1, respectively. The standard deviation of the pixel values within the anterior lateral quarter of the lung increased 3.8 times and 3.2 times at 3 months and 9 months, respectively, in the former group, as opposed to 1.2 and 1.1 in the latter group. Thirteen patients had an increase in either cough or dyspnoea as observed 3 months after completion of radiotherapy. In eleven patients these symptoms persisted 6 months later. No significant correlation was found between radiological findings and subjective symptoms. However, when three different treatment techniques were compared among 29 patients the highest rate of radiological findings was observed in patients in which the largest lung volumes received the target dose. A tendency towards an increased rate of subjective symptoms was also found in this group.

Lung density changes observed in vivo in rat lungs after irradiation: variations among and within individual lungs

Lung density measurements using Computed Tomography have been used before at various intervals after irradiation to monitor radiation-induced changes in the lung. The average lung density, its standard deviation which was used as a measure of the density homogeneity throughout the lung, and the densities of smaller lung regions were measured before and up to 76 weeks after irradiation in rat lungs. Large differences in individual response to irradiation were observed. Both increases and decreases in lung density were measured. Regions of very low density were often found adjacent to dense foci of radiation damage. These compensatory changes made the measurement of changes in average lung density an insensitive index of radiation damage. However, the measurement of regional densities in smaller lung volumes, a method not previously applied to rodents, was a much more sensitive index of radiation damage. Changes from non-irradiated control lung densities were observed at earlier times and for lower radiation doses.

The Uniform Density Assumption: Its Effect upon the Estimation of Body Segment Inertial Parameters

The effect of the uniform density assumption upon estimation of body segment inertial parameters was examined by employing directly measured, CT-derived, and cadaver-derived density values. Sectional and average density values for the right leg segments of a patient 29 years of age and a cadaver (65 years) were obtained with a GE 9800 computed tomography scanner using dual energy radiographic factors of 80 kV, 200 mAs, and 140 kV, 200 mAs. Careful sectioning of the cadaver leg following these scans permitted mass and density measurements to be directly performed. The results for both legs showed marked variation in cross-section density values throughout their lengths, which highlighted the limitations of the assumption of uniform segment density. The effect of employing this assumption was tested using a series of inertial parameter estimation strategies by means of mathematical modeling. Adoption of the uniform density assumption when estimating inertial parameters of the human leg segment was shown to produce only minor errors. However, greater errors were shown to be caused by inaccurate estimates of segment volume.

CT measurement of lung density: the role of patient position and value for total body irradiation

To obtain more precise data on pulmonary doses in preparation for total body irradiation, the lung density of patients was systematically analyzed in treatment positions using data obtained by computed tomography (CT). With the patient supine, the lung density was not significantly different for the right and left lungs. In contrast, considerable differences were noted between the two lungs in lateral decubitus positions owing to variations in ventilation and perfusion. The relative electron density of lung was also found to decrease with age, dropping to pe = 0.160 at 71 years.

Computed tomographic assessment of radiation induced damage in the lung of normal and WR 2721 protected LAF1 mice

LAF1 mice were irradiated with single, graded doses of X rays to the thorax in the range of 0 to 14 Gy unprotected, or 0 to 18 Gy after injecting the radioprotective aminothiol compound, WR 2721. Computed tomographic (CT) scanning of the thorax was performed at intervals for a period of 42 weeks after irradiation. The gravimetric density was determined for both left and right lungs by averaging the CT numerical data within lung slices traced on a magnified video image of the thorax. Significant elevations in CT density occurred at post-irradiation times corresponding to pneumonitis and late phase, as evidenced by the pneumopathic decline in survival. The threshold dose yielding a significant increase in CT density in the pneumonitis phase was 11 Gy, a dose at which only 3% of the animals died. A single peak of increased CT density was observed for the pneumonitis phase for unprotected animals, whereas a transient return of CT density toward control values at 21-22 weeks produced two peaks from the WR 2721 treated group. The CT density of lung increased in a stepwise manner in the dose range of 11-14 Gy. For the isoeffect dose that produced equal animal survival (14 Gy and 18 Gy + WR 2721), the lung density increased by approximately 27% over control values for both treatments, suggesting that CT density is related to survival. Periodic computed tomographic analysis of the lungs of patients sustaining radiotherapy to large pulmonary fields may be of value in assessing the degree and progression of pulmonary complications.

Lung dose calculations using computerized tomography: is there a need for pixel based procedures?

Radiotherapy of tumors within the thorax often requires radiation beams that traverse lung tissues. Not only do lungs consist of large volumes of low density tissues but, in addition, lung tissues are more radiosensitive than most other tissues. Computed tomography (CT) provides the detailed anatomic and geometric data ideally suited for accurate lung dose calculations. However, most radiotherapy centers do not have direct access to a CT scanner for planning purposes. In addition, it is argued by some that it is not necessary to use the CT pixel data directly for lung dose calculations but that it is sufficient to have the lung geometry outlined and to assume some average bulk density (not related to that specific patient) for the total lung volume. This study analyzes the potential errors in lung dose calculations when different lung density assumptions are made. The conclusions show that different levels of accuracy can be achieved with different levels of sophistication in lung density assumptions. However, the delivery of radiation absorbed dose to lung to an accuracy of 5% for all patients will require the detailed anatomical density data that are provided by CT.

The density of mouse lung in vivo following X irradiation

The lungs of mice were irradiated with single X radiation doses of 5 to 14 Gy. Six weeks after irradiation, computed tomographic (CT) scans of the mice were performed at two-week intervals. Beyond 14 weeks after irradiation, the animals were scanned at 1-week intervals. The mice irradiated to 5 and 7 Gy exhibited no change in lung density, in comparison with the unirradiated lungs of control mice up to times of 48 weeks. The mice irradiated to doses of greater than 10 Gy exhibited marked increases in lung density at 15 weeks after irradiation. Increases in density followed a similar time course for these doses, but the magnitude of the density increase was dependent on the radiation dose. An interpretation of these findings in terms of radiation pneumonitis is presented, and the possibility of using CT to monitor lung density in radiotherapy patients is discussed.

Post-irradiation lung density changes measured by computerized tomography

To investigate the possibility of using computerized tomography (CT) as a prognostic indicator of radiation induced lung toxicity, a series of CT scans were performed on one patient. These scans suggested an increase in lung density at day 73 after an upper half body irradiation. Because of the difficulty in lung density follow-up in patients irradiated for palliation, further studies were performed with LAF1 mice. Serial scans were taken on three groups of mice: (1) control group, (2) irradiated to 11.0 Gy and (3) irradiated to 14.0 Gy. Lung density increases between 10 and 15% were observed in the two irradiated groups. The time course was dependent on dose with an earlier onset for the high dose group (14 weeks) than for the low dose group (24 weeks). These density changes were observed only a few weeks prior to the death of the animals, indicating that small animals such as mice will not likely be useful for assessing CT scanning as an early predictor of radiation damage to lungs.

Errors in measuring trabecular bone by computed tomography due to marrow and bone composition

The linear attenuation coefficient (mu in cm-1) of trabecular bone was modeled for different conditions of bone and marrow composition in order to assess their influence on computed tomography (CT) quantitation. A large relative change (10% of TBV at 15% TBV) of bone concentration resulted in small changes of mu: 2.3% at 60 keV, 3.4% at 44 keV, 5.2% at 29 keV. Relative changes of trabecular bone volume (TBV) on the order of 3% could be detected in vivo by CT were it not for errors of relocation and for compositional influences on accuracy. The mu (and density) depended critically not only on amounts of bone substance and marrow but on their compositions. Normal variation in the composition of bone substance produced an uncertainty in mu equivalent to 0.5 to 1% TBV. Increases of yellow marrow produced a decrease of mu which could be mistaken for a decrease of bone concentration. The biological variation (90% confidence limit) of marrow composition gives an uncertainty at 15% TBV of about 2.4% TBV at 60 keV, 1.7% at 44 keV, and 1.3% at 29 keV. These correspond to relative uncertainties of 16, 11, and 9% respectively. These factors help explain the large accuracy errors (30%) observed in all studies of trabecular bone where single-energy CT was used. Marrow composition also can affect precision of bone measurement. Systematic shifts of red and yellow marrow could mask biological changes such as those occurring with aging or treatment.

Lung density as measured by computerized tomography: implications for radiotherapy

Accurate dose calculations for radiotherapy planning require a detailed knowledge of the internal anatomy of the patient both in terms of geometry and density. Computed tomography (CT) is presently the best means of providing these data. Fifty-eight patients who had scans of the thorax for radiotherapy planning were studied. The CT numbers were converted to relative electron densities and average lung densities were obtained for every patient. A linear correlation of lung density with age was found with the mean lung density of 0.35 at an age 5 years and 0.19 at an age of 80. The effect of scanning the patient under full inspiration, full expiration or normal shallow breathing conditions was analyzed. At the age of 5 years the expiration and inspiration average lung densities were 0.36 and 0.20, while at the age of 80 years they were 0.22 and 0.16, respectively. Respiratory volume changes were linearly correlated with changes in relative electron density. Differences in lung density between expiration and inspiration scans were found to demonstrate a similar trend with age as the relationship between vital capacity and age. The dosimetric and the possible clinical implications of lung density measurements for radiotherapy are considered. In particular, dose calculations were performed using scans taken under a number of different respiratory conditions. Doses calculated for a single cobalt-60 beam can differ by more than 25% when comparing a full inspiration scan to a normal breathing scan. A similar comparison for a parallel pair distribution on the lung yields a difference of about 3% while a typical three field technique for treating cancer of the esophagus shows a difference of nearly 10%.