Myocardial tissue phenotyping by radiomic features of native T1 maps and machine learning enhances disease detection and classification

Background

Myocardial T1 mapping by cardiac magnetic resonance (CMR) is a useful technique to detect diffuse myocardial fibrosis, but a major limitation of T1 mapping is the significant overlap in native T1 values between health and disease.

Purpose

We explored whether radiomic features from T1 maps could enhance the diagnostic value of T1 mapping in distinguishing health from disease and classifying cardiac disease phenotypes.

Methods

In a total of 149 patients (n=30 with no evidence of heart disease, n=30 with LVH of various etiologies, n=61 with hypertrophic cardiomyopathy (HCM) and n=28 with cardiac amyloidosis) undergoing a CMR scan for various indications were included in this study. In addition to measuring native myocardial T1 values from T1 maps, we extracted a total of 843 radiomic features of myocardial texture and explored their value in disease classification.

Results

We first demonstrated that T1 mapping images are a rich source of extractable, quantifiable data. The first three principal components of the T1 radiomics were significantly and distinctively correlated with cardiac disease type. Unsupervised hierarchical clustering of the population by myocardial T1 radiomics was significantly associated with myocardial disease type (chi2=55.98, p<0.0001). After machine learning for feature selection, training with internal validation and external testing, a model of T1 radiomics had good diagnostic performance (AUC 0.753) for multinomial classification of disease phenotype (normal vs. LVH vs. HCM vs. amyloid). A subset of seven radiomic features outperformed mean native T1 values for classification between myocardial health vs. disease and HCM phenocopies (for normal: T1 AUC 0.549 vs. radiomics AUC 0.888, for LVH: T1 AUC 0.645 vs. radiomics AUC 0.790, for HCM T1 AUC 0.541 vs. radiomics AUC 0.638 and for amyloid T1 AUC 0.769 vs. radiomics AUC 0.840).

Conclusions

We have shown that specific imaging patterns in myocardial native T1 maps are linked to features of cardiac disease and we have provided for the first-time evidence that radiomic phenotyping can be used to enhance the diagnostic yield of native T1 mapping for myocardial disease detection and classification.

Funding Acknowledgement

Type of funding sources: None.

Prognostic value of vascular inflammation biomarkers over clinical risk factors for cardiovascular risk : a meta-analysis

Funding Acknowledgements

Type of funding sources: None.

Background

Measurement of vascular inflammation biomarkers is supported for estimation of residual inflammatory risk and cardiovascular risk stratification, but to date there is no systematic assessment of the added value of such biomarkers in predicting cardiovascular events and their comparative performance.

Methods

We systematically searched in MEDLINE published literature before Apr 14, 2020 for prospective cohort studies assessing the prognostic value of common biomarkers of vascular inflammation in stable patients with or without cardiovascular disease. The primary outcome was the difference in the c-index (Δ[c-index]) of the best clinical model with the use of inflammatory biomarkers for the prediction of the composite endpoint of major adverse cardiovascular events (MACEs) and mortality. The secondary outcome was the Δ[c-index] for MACEs only. We calculated I² to test heterogeneity. We used random-effects modelling for the meta-analyses to assess the primary and secondary outcome.

Results

We identified 92,507 studies in MEDLINE after duplicates were removed, of which 90,882 (96%) studies were excluded after screening the titles and abstracts, and 1,507 (93%) of the 1,625 remaining studies were excluded after assessment of the full texts. We included 93 (6%) studies in our quantitative evaluation, in which 351,628 individuals participated. The combination of high-risk plaque features and Fat attenuation Index (FAI) by CCTA was associated with the highest prognostic value i.e. Δ[c-index] for the composite endpoint per biomarker type (A). In meta-analysis, both plasma and imaging biomarkers of vascular inflammation offered incremental prognostic value for the primary outcome (pooled estimate for Δ[c-index]% 2.9, 95%CI 2.1-3.6, B) and for MACEs only (pooled estimate for Δ[c-index]% 2.9, 95%CI 2.1-3.8). The prognostic value of imaging biomarkers significantly surpassed that of plasma biomarkers for the primary outcome (Δ[c-index]% 11.3, 95%CI 8.3-14.3 vs. 1.4, 95%CI 0.9-1.8 respectively, p = 1.7x10-10, C). Notably, biomarkers of vascular inflammation offered higher incremental prognostic value in studies with a shorter duration of follow-up (i.e. <5 years), in primary CHD prevention setting and lower cardiovascular risk populations i.e. (studies with <5% cumulative event incidence, D)

Conclusions

The combination of HRP features and FAI by CCTA imaging had the highest prognostic value for cardiovascular events among plasma or imaging biomarkers of vascular inflammation. CCTA imaging to detect residual inflammatory risk and the vulnerable patient at risk for events is a rational approach to improve risk stratification and prognostication.

Abstract Figure.

Dose and dose averaged LET comparison of ¹H, ⁴He, ⁶Li, ⁸Be, ¹⁰B, ¹²C, ¹⁴N, and ¹⁶O ion beams forming a spread-out Bragg peak

PURPOSE

Modern clinical accelerators are capable of producing ion beams from protons up to neon. This work compares the depth dose distribution and corresponding dose averaged linear energy transfer (LET) distribution, which is related to the biological effectiveness, for different ion beams (¹H, ⁴He, ⁶Li, ⁸Be, ¹⁰B, ¹²C, ¹⁴N, and ¹⁶O) using multi-energetic spectra in order to configure spread-out Bragg peaks (SOBP).

METHODS

Monte Carlo simulations were performed in order to configure a 5 cm SOBP at 8 cm depth in water for all the different ion beams. Physical dose and dose averaged LET distributions as a function of depth were then calculated and compared. The superposition of dose distribution of all ions is also presented for a two opposing fields configuration. Additional simulations were performed for (12)C beams to investigate the dependence of dose and dose averaged LET distributions on target depth and size, as well as beam configuration. These included simulations for a 3 cm SOBP at 7, 10, and 13 cm depth in water, a 6 cm SOBP at 7 depth in water, and two opposing fields of 6 cm SOBP.

RESULTS

Alpha particles and protons present superior physical depth dose distributions relative to the rest of the beams studied. Dose averaged LET distributions results suggest higher biological effectiveness in the target volume for carbon, nitrogen and oxygen ions. This is coupled, however, with relatively high LET values-especially for the last two ion species-outside the SOBP where healthy tissue would be located. Dose averaged LET distributions for ⁸Be and ¹⁰B beams show that they could be attractive alternatives to ¹²C for the treatment of small, not deeply seated lesions. The potential therapeutic effect of different ion beams studied in this work depends on target volume and position, as well as the number of beams used.

CONCLUSIONS

The optimization of beam modality for specific tumor cites remains an open question that warrants further investigation and clinically relevant results.

Real-time imaging for dose evaluation during antiproton irradiation

Online monitoring of the stopping distribution of particle beams used for radiotherapy provides the possibility of detecting possible errors in dose deposition early during a given treatment session, and may therefore help to improve the quality of the therapy. Antiproton annihilation events produce several long-range secondary particles which can be detected in real time by standard high energy particle physics detector systems. In this note, Monte Carlo calculations are performed in order to study the feasibility of real-time imaging by detecting charged pions produced during antiproton irradiation of typical biological targets. A simple treatment plan in a water phantom is simulated and the results show that by detecting pi+/- the position and the size of the planned target volume can be located with precision in the order of 1 mm.

A dosimetric comparison of Yb169 and Ir192 for HDR brachytherapy of the breast, accounting for the effect of finite patient dimensions and tissue inhomogeneities

Monte Carlo simulation dosimetry is used to compare 169Yb to 192Ir for breast high dose rate (HDR) brachytherapy applications using multiple catheter implants. Results for bare point sources show that while 169Yb delivers a greater dose rate per unit air kerma strength at the radial distance range of interest to brachytherapy in homogeneous water phantoms, it suffers a greater dose rate deficit in missing scatter conditions relative to 192Ir. As a result of these two opposing factors, in the scatter conditions defined by the presence of the lung and the finite patient dimensions in breast brachytherapy the dose distributions calculated in a patient equivalent mathematical phantom by Monte Carlo simulations for the same implant of either 169Yb or 1921r commercially available sources are found comparable. Dose volume histogram results support that 169Yb could be at least as effective as 192Ir delivering the same dose to the lung and slightly reduced dose to the breast skin. The current treatment planning systems' approach of employing dosimetry data precalculated in a homogeneous water phantom of given shape and dimensions, however, is shown to notably overestimate the delivered dose distribution for 169Yb. Especially at the skin and the lung, the treatment planning system dose overestimation is on the order of 15%-30%. These findings do not undermine the potential of 169Yb HDR sources for breast brachytherapy relative to the most commonly used 192Ir HDR sources. They imply, however, that there could be a need for the amendment of dose calculation algorithms employed in clinical treatment planning of particular brachytherapy applications, especially for intermediate photon energy sources such as 169Yb.

Polymer gel dosimetry for the TG-43 dosimetric characterization of a new 125I interstitial brachytherapy seed

In this work, a polymer gel-magnetic resonance (MR) imaging method is employed for the dosimetric characterization of a new 125I low dose rate seed (IsoSeed model I25.S17). Two vials filled with PABIG gel were prepared in-house and one new seed as well as one commercially available 125I seed of similar dose rate and well-known dosimetric parameters (IsoSeed model I25.S06) were positioned in each vial. Both seeds in each vial were MR scanned simultaneously on days 11 and 26 after implantation. The data obtained from the known seed in each vial are used to calibrate the gel dose response which, for the prolonged irradiation duration necessitated by the investigated dose rates, depends on the overall irradiation time. Data for this study are presented according to the AAPM TG-43 dosimetric formalism. Polymer gel results concerning the new seed are compared to corresponding, published dosimetric results obtained, for the purpose of the new seed clinical implementation, by our group using the established methods of Monte Carlo (MC) simulation and thermo-luminescence dosimetry (TLD). Polymer gel dosimetry yields an average dose rate constant value of lambda = (0.921 +/- 0.031) cGy h(-1) U(-1) relative to (MC)lambda = (0.929 +/- 0.014) cGy h(-1) U(-1), (TLD)lambda = (0.951 +/- 0.044) cGy h(-1) U(-1) and the average value of Lambda = (0.940 +/- 0.051) cGy h(-1) U(-1) proposed for the clinical implementation of the new seed. Results for radial dose function, g(L)(r), and anisotropy function, F(r, theta), also agree with corresponding MC calculations within experimental uncertainties which are smaller for the polymer gel method compared to TLD. It is concluded that the proposed polymer gel-magnetic resonance imaging methodology could be used at least as a supplement to the established techniques for the dosimetric characterization of new low energy and low dose rate interstitial brachytherapy seeds.

Relative output factor measurements of a 5 mm diameter radiosurgical photon beam using polymer gel dosimetry

Besides the fine spatial resolution inherent in polymer gel-magnetic resonance imaging (MRI) dosimetry, the method also features the potential for multiple measurements of varying sensitive volume in a single experiment by integrating results in MRI voxels of finite dimensions (i.e., in plane resolution by slice thickness). This work exploits this feature of polymer gel dosimetry to propose an experimental technique for relative output factor (OF) measurements of small radiosurgical beams. Two gel vials were irradiated with a 5 and 30 mm diameter 6 MV radiosurgery beam and MR scanned with the same slice thickness and three different in plane resolutions. Using this experimental data set, 5 mm OF measurements with the PinPoint ion chamber are simulated by integrating results over a sensitive volume equal to that of the chamber. Results are found in agreement within experimental uncertainties with actual PinPoint measurements verifying the validity of the proposed experimental procedure. The polymer gel data set is subsequently utilized for OF measurements of the 5 mm beam with varying sensitive volume to discuss the magnitude of detector volume averaging effects. Seeking to correct for volume averaging, results are extrapolated to zero sensitive volume yielding a 5 mm OF measurement of (0.66+/-5%). This result compares reasonably with corresponding ionometric and radiographic film measurements of this work and corresponding, limited, data in the literature. Overall, results suggest that polymer gel dosimetry coupled with the proposed experimental procedure helps overcome not only tissue-equivalence and beam perturbation implications but also volume averaging and positioning uncertainties which constitute the main drawback in small radiosurgical beam dosimetry.

Dose verification of single shot gamma knife applications using VIPAR polymer gel and MRI

This work describes an experimental procedure with potential to assess the overall accuracy associated with gamma knife clinical applications, from patient imaging and dosimetry planning to patient positioning and dose delivery using the automated positioning system of a Leksell Gamma Knife model C. The VIPAR polymer gel-MRI dosimetry method is employed due to its inherent three-dimensional feature and linear dose response over the range of gamma knife applications. Different polymer gel vials were irradiated with single shot gamma knife treatment plans using each of the four available collimator helmets to deliver a maximum dose of 30 Gy. Percentage relative dose results are presented not only in the form of one-dimensional profiles but also planar isocontours and isosurfaces in three dimensions. Experimental results are compared with corresponding Gammaplan treatment planning system calculations as well as acceptance test radiochromic film measurements. A good agreement, within the experimental uncertainty, is observed between measured and expected dose distributions. This experimental uncertainty is of the order of one imaging pixel in the MRI gel readout session (<1 mm) and allows for the verification of single shot gamma knife applications in terms of acceptance specifications for precision in beam alignment and accuracy. Averaging net R(2) results in the dose plateau of the 4 mm and 18 mm collimator irradiated gel vials, which were MR scanned in the same session, provides a crude estimate of the 4 mm output factor which agrees within errors with the default value of 0.870.

Monte Carlo dosimetry of 60Co HDR brachytherapy sources

Monte Carlo based dosimetry is presented for three 60Co HDR sources with a unique configuration of two active pellets in contact or spaced 9 and 11 mm apart. Results are presented in Cartesian "away and along" as well as polar coordinates following the AAPM TG-43 dosimetric formalism. Iso-dose rate contours around the sources in Cartesian coordinates reveal that significant differences between the three source designs exist only close to the source centers where dose rate distributions bear the effect of the unique source configurations. Dose rate constants of all three sources are accurately described by an equation of the form: lambda(cGyh(-1)U(-1))=lambda*pointG(1 cm,90 degrees) = 1.094*G(1 cm,90 degrees) where lambdapoint is the dose rate constant of a bare 60Co point source and G(1 cm,90 degrees) is the "exact" source geometry factor as defined by the TG-43. Radial dose and anisotropy function data extracted using the point source approximated geometry factors are tabulated for all three source designs. Finally, the dependence and variation of the above TG-43 parameters are discussed and it is demonstrated that the dosimetric properties of high-energy photon emitters are largely dependent on radionuclide source distribution.

Evaluation of the performance of VIPAR polymer gels using a variety of x-ray and electron beams

The aim of this investigation was the evaluation of the usefulness of N-vinyl pyrrolidone argon (VIPAR) polymer gel dosimetry for relative dose measurements using the majority of types and energies of radiation beams used in clinical practice. For this reason, VIPAR polymer gels were irradiated with the following beams: 6 and 23 MV photons (maximum dose: 15 Gy) and 6, 9, 12, 15, 18 and 21 MeV electrons (90% dose: 15 Gy). Using 6 MV x-rays, a linear gel dose response was verified for doses up to 20 Gy. Assuming linearity of response for the rest of the photon and electron beams used in this study, percentage depth dose measurements were derived. For all beams used and the range of relative doses studied, a satisfying agreement was observed between percentage depth dose measurements performed using the VIPAR gel-MRI method and an ion chamber, validating the assumption that a linear gel dose response holds for all photon and electron beams studied. VIPAR gels, therefore, can be used for relative dose distribution measurements using photons or electrons of any typical energy used in external radiotherapy applications. It is also demonstrated that two-dimensional dose distribution measurements through an irradiated (9 MeV electrons, 3 cm x 3 cm cone) VIPAR gel volume can be easily obtained.

Monte Carlo dosimetry of a new 192Ir pulsed dose rate brachytherapy source

A new microSelectron pulsed dose rate source has been designed, containing two active pellets instead of one inactive and one active pellet contained in the old design, to facilitate the incorporation of higher activity up to 74 GBq (2 Ci). In this work, Monte Carlo simulation is used to derive full dosimetric data following the AAPM TG-43 formalism, as well as the dose rate per unit air kerma strength data in Cartesian, "away and along" coordinates for both source designs. The calculated dose rate constant of the new PDR source design was found equal to lambda=(1.121 +/- 0.006) cGy h(-1) U(-1) compared to lambda = (1.124 +/- 0.006) cGy h(-1) U(-1) for the old design. Radial dose functions of the two sources calculated using the point source approximated geometry factors were found in close agreement (within 1%) except for radial distances under 2 mm. At polar angles close to the longitudinal source axis at the sources' distal end, the new design presents increased anisotropy (up to 10%) compared to the old one due to its longer active core. At polar angles close to the longitudinal source axis at the sources' drive wire end however, the old design presents increased anisotropy (up to 18%) due to attenuation of emitted photons through the inactive Ir pellet. These differences, also present in "away and along" dose rate results, necessitate the replacement of treatment planning input data for the new microSelectron pulsed dose rate source.

Polymer gel dosimetry using a three-dimensional MRI acquisition technique

In this work, three-dimensional (3-D) MRI techniques are employed in N-Vinylpyrrolidone-Argon-(VIPAR-) based polymer gel dosimetry. VIPAR gels were irradiated using a Nucletron microSelection 192Ir HDR brachytherapy remote afterloading system with single source dwell position and intravascular brachytherapy irradiation protocols. A single VIPAR gel and a single irradiation are adequate to obtain the full calibration curve needed. The 3-D dose distributions obtained with the 3-D MRI method were found to be in good agreement with the corresponding Monte Carlo calculations, for brachytherapy and intravascular irradiations. The method allows the reconstruction of isodose contours over any plane, with increased spatial resolution and accuracy following a single MR acquisition. VIPAR gel measurements using a 3-D MRI readout technique can be of particular use in the experimental dosimetry of brachytherapy sources, as well as for dose verification purposes when complex irradiation regimes and three-dimensional dose gradients are investigated.

Dosimetry comparison of 192Ir sources

192Ir sources besides being widely utilized in the field of conventional brachytherapy also find use in contemporary peripheral and coronal intravascular applications. In this study, the same Monte Carlo simulation code and input data were used to investigate differences between the dose rate distributions of the most commonly used 192Ir sources in the cm and mm distance range. Findings are discussed in view of differences in source and encapsulation dimensions as well as structural details. Results are presented in the AAPM TG-43 formalism, as generalized by AAPM TG-60, for five 192Ir HDR source designs as well as an LDR seed and an LDR wire source. Dose rate constants of the sources at r0 = 1 cm and r0 = 2 mm were found proportional to the corresponding geometry factors along the transverse source bisectors and an equation of the form lambda r0(cGyh(-1) U(-1)) = 1.12 x G(r0,90 degrees) provides results within clinical accuracy (less than 2%) for any 192Ir source. Radial dose functions do not depend significantly on source and encapsulation geometry and agree within 2% with that of a point 192Ir source. Anisotropy is of importance for accurate dosimetry at the cm distance range but it does not affect dose rate in the mm distance range significantly. At such short radial distances the source geometry factor defines the shape of isodose lines. Dose uniformity at given distances from the sources is strongly dependent on source dimensions as indicated by dose rate profiles in polar and Cartesian coordinates.

Dosimetry close to an 192Ir HDR source using N-vinylpyrrolidone based polymer gels and magnetic resonance imaging

In this work, the utilization of polymer gel-MRI dosimetry for measurements at distances relevant to clinical brachytherapy and intravascular applications [i.e., in the mm range, where steep three-dimensional (3-D) dose gradients exist] is investigated using N-vinylpyrrolidone-based gels. Transverse axis radial dose distributions, dose distributions parallel to the source axis, and 2-D dose distributions around the commonly used microSelectron 192Ir HDR source are measured for single source dwell position irradiations. Experimental results are found in good agreement with verified Monte Carlo calculations, even for distances less than 3 mm from the source. The effect of various MRI parameters, such as slice thickness, slice mispositioning, and in-plane resolution, on the accuracy of the method is also investigated. Possible limitations of the method are discussed, and its' overall potential in brachytherapy dosimetry is evaluated. Experimental 2-D dose distributions for an intravascular application following the Paris irradiation protocol are compared to corresponding commercial treatment planning system calculations. Results suggest that polymer gel-MRI dosimetry is capable of experimentally verifying dose distributions in relevant clinical intravascular applications.

Narrow stereotactic beam profile measurements using N-vinylpyrrolidone based polymer gels and magnetic resonance imaging

In this work, polymer gel-MRI dosimetry (using VIPAR gels), radiographic film and a PinPoint ion chamber were used for profile measurements of 6 MV x-ray stereotactic beams of 5 and 10 mm diameter. The VIPAR gel-MRI method exhibited a linear dose response up to 32 Gy. VIPAR gels were found to resolve the penumbra region quite accurately, provided that the in-plane image resolution of the related T2-map is adequate (< or = 0.53 mm). T2-map slice thickness had no significant effect on beam profile data. VIPAR measurements performed with a spatial resolution of 0.13 mm provided penumbra widths (80%-20% distance) of 1.34 and 1.70 mm for the 5 and 10 mm cones respectively. These widths were found to be significantly smaller than those obtained with the film (2.23 mm for the 5 mm cone, 2.45 mm for the 10 mm cone) and PinPoint (2.25 mm for the 5 mm cone, 2.52 mm for the 10 mm cone) methods. Regarding relative depth dose measurements, good correlation between VIPAR gel and PinPoint data was observed. In conclusion, polymer gel-MRI dosimetry can provide relatively accurate profile data for very small beams used in stereotactic radiosurgery since it can overcome, to some extent, the problems related to the finite size of conventional detectors.

Use of membrane and bone grafts in the reconstruction of orbital fractures

OBJECTIVE

To present and analyze the clinical results derived from the use of different grafts for the reconstruction of orbital defects during a 10-year period.

STUDY DESIGN

Fifty-five fracture cases with orbital bony defect, requiring a graft, are presented. The surgical treatment includes the reconstruction of the fracture (osteosynthesis) and the repair of the remaining bone defect by graft, with the type of graft dependent on the size of the defect. For minor defects membranes were used (lyophilized dura or alloplastic dura mater), whereas major defects were repaired with bone grafts (autografts, heterografts, or bone substitute material). All patients have been regularly evaluated for at least one year postoperatively.

RESULTS

All grafts were well tolerated by the patients. Diplopia subsided in all but 5 cases, motility disturbance was fully repaired in all but 3 cases. Esthetics were improved in cases with severe bone defect.

CONCLUSION

The wide variety of grafts available allows successful reconstruction of all types of orbital bony defects. The clinician should be able to use different types of grafts depending on the type and size of the defect.

Dosimetry of 192Ir wires for LDR interstitial brachytherapy following the AAPM TG-43 dosimetric formalism

Implementation of the AAPM Task Group 43 dosimetric formalism for 192Ir wires used as interstitial sources in low dose-rate (LDR) brachytherapy applications is investigated. Geometry factors, dose-rate constant values, radial dose functions, and anisotropy functions, utilized in this formalism, were calculated for various lengths of all commercially available wire source designs by means of a well-established Monte Carlo simulation code and an improved modification of the Sievert integral method. Results are presented in the form of look up tables that allow interpolation for dose-rate calculations around all practically used wire lengths, with accuracy acceptable for clinical applications.

Bladder wall dosimetry for 131I administered activities

Detailed dosimetric calculations of bladder dose from 131I accumulations are presented, using Monte Carlo simulations and the dynamic bladder model proposed in the recent revision of MIRD Pamphlet No 14. Penetrating radiation and electron dose to the bladder wall inner surface, as well as within the bladder wall, per unit of 131I activity accumulated in the bladder, are calculated for various bladder filling volumes. The effect of backscatter was taken into account for photon dose calculations while electron dose was determined for all the beta spectral components of 131I. Dosimetric results were used to derive simple parameterisation functions of photon and electron dose with respect to bladder filling volume in combination with any bladder physiology model. Calculations of bladder dose per unit of 131I-OIH and 131I-NaI administered activity, using the derived parametrisations, differ by less than 12% from corresponding results given in the recent revision of MIRD Pamphlet No 14.

Monte Carlo dosimetry of a new 192Ir high dose rate brachytherapy source

This work provides full dosimetric data for a new high-strength 192Ir source currently launched by Varian Oncology Systems for use in their high dose rate remote afterloading systems. The active core length of the new source is reduced to 5 mm compared to a value of 10 mm for the existing VariSource source design, with all other geometric source and encapsulation details being similar. Dose-rate constant, radial dose functions, geometry factors, and anisotropy functions, utilized in the AAPM Task Group 43 dose calculation formalism, were calculated using Monte Carlo simulation. Results are compared with corresponding data published for the existing VariSource and microSelectron high dose rate sources. The dose-rate constant for the new Varian source was found to be equal to 1.101 +/- 0.006cGyh(-1) U(-1), compared to values of 1.043 +/- 0.005 and 1.116 +/- 0.006 cGyh(-1) U(-1) calculated for the existing VariSource and microSelectron sources, respectively. The radial dose functions between the three sources are similar with the exception of their values at radial distances very close to the source (r approximately 2 mm) where differences of approximately 3% are observed. The new Varian source demonstrates a smaller anisotropy relative to the existing VariSource source design for polar angles close to the source longitudinal axis, due to its smaller active core length.

Dose rate calculations around 192Ir brachytherapy sources using a Sievert integration model

The classical Sievert integral method is a valuable tool for dose rate calculations around brachytherapy sources, combining simplicity with reasonable computational times. However, its accuracy in predicting dose rate anisotropy around 192Ir brachytherapy sources has been repeatedly put into question. In this work, we used a primary and scatter separation technique to improve an existing modification of the Sievert integral (Williamson's isotropic scatter model) that determines dose rate anisotropy around commercially available 192Ir brachytherapy sources. The proposed Sievert formalism provides increased accuracy while maintaining the simplicity and computational time efficiency of the Sievert integral method. To describe transmission within the materials encountered, the formalism makes use of narrow beam attenuation coefficients which can be directly and easily calculated from the initially emitted 192Ir spectrum. The other numerical parameters required for its implementation, once calculated with the aid of our home-made Monte Carlo simulation code, can be used for any 192Ir source design. Calculations of dose rate and anisotropy functions with the proposed Sievert expression, around commonly used 192Ir high dose rate sources and other 192Ir elongated source designs, are in good agreement with corresponding accurate Monte Carlo results which have been reported by our group and other authors.

A new polymer gel for magnetic resonance imaging (MRI) radiation dosimetry

New composition polymer gels, the N-vinylpyrrolidone argon (VIPAR) gels, were developed and investigated as MRI dosimeters. VIPAR gels were irradiated in the dose range of 0-12 Gy by a 6 MV x-ray linear accelerator and MR-scanned in a 1.5 T magnetic resonance imager. A linear relationship was found between absorbed dose and spin spin relaxation rate R2. The dose sensitivity was found to be approximately 0.1 s(-1) Gy(-1) for a gel composition of 4% w/w in N-vinylpyrrolidone, 4% w/w in N,N'-methylene-bisacrylamide, 5% w/w in gelatine type A and 87% w/w in water. This dose sensitivity was stable with time and did not deteriorate even when a boost radiation dose of 2.5 Gy was applied 15 days after the first irradiation. Good reproducibility of these results was observed when a new batch of gels was produced and used for corresponding measurements and analysis.

A Monte Carlo investigation of the dosimetric characteristics of the VariSource 192Ir high dose rate brachytherapy source

An analytical Monte Carlo simulation code, incorporating in detail source construction and dimensions, was used to investigate the dosimetric characteristics of the VariSource 192Ir high dose rate brachytherapy source. Dose-rate constant, radial dose functions, and anisotropy functions, utilized in the AAPM Task Group 43 dose estimation formalism, were calculated with the source centered in a spherical water phantom of 30 cm in diameter. The results, which are in agreement with the corresponding data available in the literature, are compared to those obtained in our previous study for the widely used microSelectron 192Ir high dose-rate brachytherapy source. The dose-rate constant was found to be equal to 1.043 +/- 0.005 cGy h(-1) U(-1) for the VariSource, compared to a value of 1.116 +/- 0.006 cGy h(-1) U(-1) calculated for the microSelectron. Significant differences in the anisotropy of the two sources are observed only for polar angles close to their long axis and are due to their different dimensions.

Monte Carlo and TLD dosimetry of an 192Ir high dose-rate brachytherapy source

An analytical Monte Carlo simulation code has been used to perform dosimetry calculations around an 192Ir high dose-rate brachytherapy source utilized in the widely used microSelectron afterloaded system. Radial dose functions, dose rate constant and anisotropy functions, utilized in the AAPM Task Group 43 dose estimation formalism, have been calculated. In addition, measurements of anisotropy functions using LiF TLD-100 rods have been performed in a polystyrene phantom to support our Monte Carlo calculations. The energy dependence of LiF TLD response was investigated over the whole range of measurement distances and angles. TLD measurements and Monte Carlo calculations are in agreement to each other and agree with published data. The influence of phantom dimensions on calculations was also investigated. Radial dose functions were found to depend significantly on phantom dimensions at radial distances near phantom edges. Deviations of up to 25% are observed at these distances due to the lack of full scattering conditions, indicating that body dimensions should be taken into account in treatment planning when the absorbed dose is calculated near body edges. On the other hand, anisotropy functions do not demonstrate a strong dependence on phantom dimensions. However, these functions depend on radial distance at angles close to the longitudinal axis of the source, where deviations of up to 20% are observed.

Epidemiological aspects of tuberculosis in the Padua Health District 1985-1996

The present study was carried out in order to monitor the epidemiology of Mycobacterium tuberculosis disease in the Padua Health District over the period 1985-1996 with a view to the situation in some population subgroups. The average annual rate per 100,000 persons per year was 7.9 with a peak in 1994 (15.6 per 100,000 persons). The most significant variations are not per annum but between the periods 1985-1990 and 1991-1996 (average rates per 100,000 persons: 4.8 and 10.9, respectively). The trend shows an increasing frequency in all age groups and especially among the elderly. Immigration from endemic areas and human immunodeficiency virus (HIV) infection appear to contribute to the increase the number of notified cases among adults. In the last four years an increase of tuberculosis cases was also observed among subjects with professional exposure risk. Improvements of programs both at national and regional level are believed to be essential to the tuberculosis control.

Dose rate distributions around 60Co, 137Cs, 198Au, 192Ir, 241Am, 125I (models 6702 and 6711) brachytherapy sources and the nuclide 99Tcm

Simple analytical functions derived from our point source Monte Carlo calculations on the combined attenuation and scatter factor, B exp(-mu r), for 60Co, 137Cs, 198Au, 192Ir, 241Am, 125I (models 6702 and 6711) brachytherapy sources and the nuclide 99Tcm, for water spherical geometries of radii R = 15 and 20 cm, are presented. Our results for the broadly used 60Co, 137Cs, 198Au and 192Ir brachytherapy sources can be compared directly and found in excellent agreement with the widely accepted data of Meisberger et al in the limited distance range for which the latter are valid. Our data, however, can be used with high accuracy outside this distance range. Many discrepancies observed among different data sets available in recent literature are attributable to differences in geometries used. The results for the recently introduced 241Am source are very dissimilar to those produced by any other currently used brachytherapy source. Dose rate distributions, based on the above simple functions, are proposed in accordance with the recommendations for calibration of the brachytherapy sources in terms of reference air kerma rate and were found to be in good agreement with data available in the literature. Our calculations for 125I sources (models 6702 and 6711), provided that the characteristic x-rays from titanium encapsulation are taken into account, support recent experimental and theoretical dose rate distributions indicating that currently accepted values for 125I may be overestimated.

Accurate Monte Carlo calculations of the combined attenuation and build-up factors, for energies (20-1500 keV) and distances (0-10 cm) relevant in brachytherapy

The combined build-up and attenuation factor, B exp (-mu r), of point isotropic photon sources in a water medium has been calculated using the Monte Carlo method, for energies (20-1500 keV) and distances (1-10 cm) relevant in brachytherapy. For the transport of photons and electrons, up-to-date and self-consistent total, partial and differential cross sections were used. The influence of coherent (Rayleigh) and incoherent (Compton) scattering, as well as the effects of the source and medium geometries on the calculations, were investigated in detail and it was found that these effects can lead to significant deviations from published data, especially at low energies and/or large distances from the sources. Our results can be used for any mono- or multi-energetic photon source in the energy range 20-1500 keV with uncertainties of the order of 2-3%, and they may influence treatment planning especially in the case of organs at risk which are usually near the edge of the body.

Direct immunofluorescence in oral lichen planus

Direct immunofluorescent staining (DIF) was performed on biopsy specimens from thirty-five patients with oral lichen planus. The results showed fibrin deposition in all cases at the mucosal-submucosal interface, within colloid bodies (fourteen of thirty-five) and within vascular walls (five of thirty-five). Deposition of IgG, IgA and IgM was detected to a lesser extent, while complement (C3) could not be identified in any case. The significance of these findings was assessed by comparison with the IF results obtained in thirty-five biopsies from various oral diseases other than lichen planus and ten healty persons. Although the presence of fibrin deposition at the mucosal-submucosal junction, within vessels and cytoid bodies, was found to be highly characteristic of lichen planus, these findings were not specifically diagnostic. Morphologically identical deposits were also seen in lupus erythematosus. It is known at present whether immunologic reactions may play a role in the pathogenesis of lichen planus. However, the immunopathologic findings may occasionally be additional suggestive markers in the diagnosis of the disease.

Verruciform xanthoma of the oral mucosa

An additional case of verruciform xanthoma in the oral cavity is presented. The clinical and light microscopic findings in this case are similar to those previously described. In addition, the clinical features of 47 cases that could be found in the literature are summarized and discussed.

Immunofluorescent studies in desquamative gingivitis

Twenty-seven patients (19 females and 8 males) with a clinically diagnosed desquamative gingivitis were used in this study. Twenty-seven additional patients with oral and skin diseases and with gingival involvement other than that of the desquamative gingivitis type served as controls. Biopsy gingival specimens were sectioned and stained with H and E. The rest of the tissue specimen was cut with a cryostat and used for immunofluorescent assays of the tissue-bound antibodies (IgG, IgA, IgM, C3 anf fibrin). Circulating antibodies were determined by using the indirect immunofluorescence method in serum samples. Normal human oral mucosa and several animal epithelial tissues were used as substrates. The immunopathologic findings suggest the diagnosis of lichen planus in the presence of linear or granular fibrin deposition at the basement membrane zone/mucosal submucosal interface with or without cytoid bodies and cicatricial pemphigoid, in the presence of circulating and/or tissue-bound immunoglobulins and C3 in a linear continuous pattern along the BMZ. It is suggested that an accurate diagnosis of the underlying disease of the desquamative gingivitis can be made on the bases of the clinical, histopathologic, immunopathologic and follow-up findings.

Melanin pigmentation and inflammation in human gingiva

The study was carried out to determine the distribution of melanin granules in different anatomical areas of the gingiva and to relate the density of melanin granules to the degree of gingival inflammation. Specimens including free and part of the attached gingiva were secured from the vestibular surfaces of the six maxillary or mandibular teeth of 21 Caucasian subjects during periodontal surgery. Serial histologic sections 5 to 6 mu thick were prepared and stained with Hematoxylin and Eosin or Masson-Fontana. Evaluation of the H & E stained sections disclosed that the number of melanophores and melanin granules in the epithelium and the inflammatory cells of the subepithelial connective tissue gradually decreased from the free gingival groove area to the free gingival crest and from the free gingival groove to the mucogingival junction. The density of melanophores in the vestibular epithelium showed a significant positive correlation (r = 0.38, P less than 0.00003) with the severity of inflammation (numbers of inflammatory cells) in the attached gingiva but not in the free gingiva (r = -0.033, P less than 0.2643). There was a positive significant correlation between the density of melanin granules of the vestibular epithelium and that of the adjacent connective tissue in both the free (r = 0.57, P less than 0.0003) and attached gingiva (r = 0.46, P less than 0.0003).

Immunofluorescent study of cytologic smears in oral pemphigus: A simple diagnostic technique

This study was undertaken in order to explore the diagnostic efficacy of the direct immunofluorescent technique for demonstration in vivo bound immunoglobulins to desquamated epithelial cells from patients with oral pemphigus. Cytologic smears were obtained with a wooden spatula from the oral lesions of thirty patients with known pemphigus vulgaris. Control smears were taken from thirty-five patients twenty suffering from various other mucosal diseases and fifteen devoid of oral lesions. The smears were stored at -30 degrees C. and prepared within 48 hours for examination by the direct immunofluorescent technique. Deposition of IgG was demonstrated in all thirty smears from the oral pemphigus lesions. None of the control smears gave positive results. These findings suggest that the direct immunofluorescent technique using cytologic smears may provide an additional simple and reliable method in the diagnosis or oral pemphigus vulgaris.

Cicatricial pemphigoid: direct and indirect immunofluorescent studies

Oral mucosa, skin tissue, and serum samples from thirty-three patients with cicatricial pemphigoid were studied by the direct and indirect immunofluorescent techniques to determine the presence of tissue-bound and circulating antibodies. A linear continuous basement membrane zone pattern was observed in 96.9 percent of the oral mucosa biopsy specimens studied. This pattern is indistinguishable from the pattern observed in bullous pemphigoid. In 97 percent of the cases clinically healthy skin biopsy specimens were negative for basement membrane zone staining. Circulating anti-basement membrane zone antibodies were present in 36.4 percent of thirty-three patients with cicatricial pemphigoid in low titer (1:10 to 1:40), when normal oral mucosa was used as substrate. The demonstration of tissue-bound and circulating basement membrane zone antibodies in cicatricial pemphigoid morphologically identical to those found in bullous pemphigoid provides further support for the concept that the two diseases may represent variants of the same entity.