Radiological Physics and Technology最新文献

Current imaging methods for diagnosing breast cancer (BC) are associated with limited sensitivity and specificity and modest positive predictive power. The recent progress in image analysis using artificial intelligence (AI) has created great promise to improve BC diagnosis and subtype differentiation. In this case, novel quantitative computational methods, such as radiomics, have been developed to enhance the sensitivity and specificity of early BC diagnosis and classification. The potential of radiomics in improving the diagnostic efficacy of imaging studies has been shown in several studies. In this review article, we discuss the radiomics workflow and current handcrafted radiomics methods in the diagnosis and classification of BC based on the most recent studies on different imaging modalities, e.g., MRI, mammography, contrast-enhanced spectral mammography (CESM), ultrasound imaging, and digital breast tumosynthesis (DBT). We also discuss current challenges and potential strategies to improve the specificity and sensitivity of radiomics in breast cancer to help achieve a higher level of BC classification and diagnosis in the clinical setting. The growing field of AI incorporation with imaging information has opened a great opportunity to provide a higher level of care for BC patients.

To verify the effect of the frame rate on image quality in cardiology, we used an indirect conversion dynamic flat-panel detector (FPD). We quantified the input-output characteristics, and determined the modulation transfer function (MTF) and normalized noise power spectrum (NNPS) of the equipment used in cardiology at 7.5, 10, 15, and 30 frames per second (fps). We also calculated the noise power spectrum for still images and videos at all frame rates and obtained the image lag correction factor r. The input-output characteristics and the MTF agreed even when the frame rate was varied. The NNPS tended to decrease uniformly as a function of frequency at increasing frame rates. The factor r decreased as a function of the frame rate, and its minimum value was 30 fps. Our results suggest that high-frame-rate imaging in cardiology using indirect conversion dynamic FPDs is affected by image lag.

Beam-matched linear accelerators (LA's) require accurate and precise dosimetry for fractionated stereotactic treatment. In this study, the beam data were validated by comparing the three-beam-matched LA's measured data and the vendor reference data. Upon its validation, the accuracy of the volumetric dose delivery for eighty patient-specific fractionated stereotactic treatment plans was evaluated. Measurements of the percentage depth dose (PDD), beam profiles, output factors (OFs), absolute output, and dynamic multi-leaf collimator (MLC) transmission factors for 6 MV and 10 MV flattening filter (FF) and flattening filter-free (FFF) photon beams were obtained from three-beam-matched LA's. The patient-specific quality assurance evaluation for all eighty plans was performed using PTW Octavius 1000 SRS™ array detectors for two-dimensional (2D) fluence measurement. The following 2D gamma passing criteria were used: 1%/1 mm, 2%/1 mm, 1%/2 mm, 2%/2 mm and 3%/2 mm. In all three LA's, gamma analysis for PDD and profile were above 97% with gamma criteria of 1%/1 mm. The differences OFs, absolute output, and dynamic MLC transmission factors were less than ± 1% of base value. For all eighty cases, the median passing rates on the three LA's were above 76%, 88%, 92%, 96%, and 98% for the above-mentioned gamma criteria of the three LA's. The beam-matched LA's showed good agreement between the measured and treatment planning system (TPS) calculated values for fractionated stereotactic VMAT plans with 6 MV and 10 MV (FF and FFF) photon beams. Patients can be shifted and treated on any beam-matched linac without the need of re-planning.

The aim of this study was to optimise the vessel angle as well as the stack number from the profiles of carbon dioxide digital subtraction angiography (CO₂-DSA) images of a water phantom containing an artificial vessel tilted at different angles which imitate arteries in the body. The artificial vessel was tilted at 0°, 15°, and 30° relative to the horizontal axis with its centre as the pivot point, and CO₂-DSA images were acquired at each vessel tilt angle. The maximum opacity method was used to stack up to four images of the next frame one by one. The signal-to-noise ratio (SNR) was determined from the profile curves. The Wilcoxon rank sum test was used to evaluate whether the profile curve and SNR differed depending on the vessel tilt angle or stack number, and a p-value of less than 0.05 was considered statistically significant. Images acquired at 0° had a significantly lower SNR than images acquired at 15° (p = 0.10). When the vessel angle was 30°, the profile curves were significantly improved (p < 0.05) when two or more images were stacked over the original image. Images with a good SNR were acquired at the vessel tilt angle of 15°, and the shape of the profile curve was improved when two or more images were stacked on the original image. This study demonstrates that the quality of images acquired using CO₂-DSA can be significantly improved through parameter optimisation for image acquisition and post-processing.

This study aimed to develop a novel internal target volume (ITV) definition for respiratory motion targets, considering target motion velocity and time. The proposed ITV was evaluated in respiratory-gated radiotherapy. An ITV modified with target motion velocity and time (ITVvt) was defined as an ITV that includes a target motion based on target motion velocity and time. The target motion velocity was calculated using four-dimensional computed tomography (4DCT) images. The ITVvts were created from phantom and clinical 4DCT images. The phantom 4DCT images were acquired using a solid phantom that moved with a sinusoidal waveform (peak-to-peak amplitudes of 10 and 20 mm and cycles of 2-6 s). The clinical 4DCT images were obtained from eight lung cancer cases. In respiratory-gated radiotherapy, the ITVvt was compared with conventional ITVs for beam times of 0.5-2 s within the gating window. The conventional ITV was created by adding a uniform margin as the maximum motion within the gating window. In the phantom images, the maximum volume difference between the ITVvt and conventional ITV was -81.9%. In the clinical images, the maximum volume difference was -53.6%. Shorter respiratory cycles and longer BTs resulted in smaller ITVvt compared with the conventional ITV. Therefore, the proposed ITVvt plan could be used to reduce treatment volumes and doses to normal tissues.

Urinary toxicities are one of the serious complications of radiotherapy for prostate cancer, and dose-volume histogram of prostatic urethra has been associated with such toxicities in previous reports. Previous research has focused on estimating the prostatic urethra, which is difficult to delineate in CT images; however, these studies, which are limited in number, mainly focused on cases undergoing brachytherapy uses low-dose-rate sources and do not involve external beam radiation therapy (EBRT). In this study, we aimed to develop a deep learning-based method of determining the position of the prostatic urethra in patients eligible for EBRT. We used contour data from 430 patients with localized prostate cancer. In all cases, a urethral catheter was placed when planning CT to identify the prostatic urethra. We used 2D and 3D U-Net segmentation models. The input images included the bladder and prostate, while the output images focused on the prostatic urethra. The 2D model determined the prostate's position based on results from both coronal and sagittal directions. Evaluation metrics included the average distance between centerlines. The average centerline distances for the 2D and 3D models were 2.07 ± 0.87 mm and 2.05 ± 0.92 mm, respectively. Increasing the number of cases while maintaining equivalent accuracy as we did in this study suggests the potential for high generalization performance and the feasibility of using deep learning technology for estimating the position of the prostatic urethra.

We aimed to evaluate the accuracy and repeatability of the T1, T2*, and proton density (PD) values obtained by quantitative parameter mapping (QPM) using the ISMRM/NIST MRI system phantom and compared them with computer simulations. We compared the relaxation times and PD obtained through QPM with the reference values of the ISMRM/NIST MRI system phantom and conventional methods. Furthermore, we evaluated the presence or absence of influences other than noise in T1 and T2* values obtained by QPM by comparing the obtained coefficient of variation (CV) with simulation results. The T1, T2*, and PD values by QPM showed a strong correlation with the measured values and the referenced values. The simulated CVs of QPM calculated for each sphere showed similar trends to those of the actual scans.

Deep-learning-based methods can improve robustness against individual variations in computed tomography (CT) images of the sternocleidomastoid muscle, which is a challenge when using conventional methods based on probabilistic atlases are used for automatic segmentation. Thus, this study proposes a novel multiclass learning approach for the joint segmentation of the sternocleidomastoid and skeletal muscles in CT images, and it employs a two-dimensional U-Net architecture. The proposed method concurrently learns and segmented segments the sternocleidomastoid muscle and the entire skeletal musculature. Consequently, three-dimensional segmentation results are generated for both muscle groups. Experiments conducted on a dataset of 30 body CT images demonstrated segmentation accuracies of 82.94% and 92.73% for the sternocleidomastoid muscle and entire skeletal muscle compartment, respectively. These results outperformed those of conventional methods, such as the single-region learning of a target muscle and multiclass learning of specific muscle pairs. Moreover, the multiclass learning paradigm facilitated a robust segmentation performance regardless of the input image range. This highlights the method's potential for cases that present muscle atrophy or reduced muscle strength. The proposed method exhibits promising capabilities for the high-accuracy joint segmentation of the sternocleidomastoid and skeletal muscles and is effective in recognizing skeletal muscles, thus, it holds promise for integration into computer-aided diagnostic systems for comprehensive musculoskeletal analysis. These findings are expected to enhance medical image analysis techniques and their applications in clinical decision support systems.

Proton dose calculation in media other than water may be of interest for either research purposes or clinical practice. Current study aims to quantify the required parameters for analytical proton dosimetry in muscle, bone, and PMMA. Required analytical dosimetry parameters were extracted from ICRU-49 report and Janni study. Geant4 Toolkit was also used for Bragg curve simulation inside the investigated media at different proton energies. Calculated and simulated dosimetry data were compared using gamma analysis. Simulated and calculated Bragg curves are consistent, a fact that confirms the validity of reported parameters for analytical proton dosimetry inside considered media. Furthermore, derived analytical parameters for these media are different from those of water. Listed parameters can be reliably utilized for analytical proton dosimetry inside muscle, bone, and PMMA. Furthermore, accurate proton dosimetry inside each medium demands dedicated analytical parameters and one is not allowed to use the water coefficients for non-water media.

This study investigates the influence of calculation accuracy in peripheral low-dose regions on the gamma pass rate (GPR), utilizing the Acuros XB (AXB) algorithm and ArcCHECK™ measurement. The effects of varying small field sizes, dose grid sizes, and split-arc techniques on GPR were analyzed. Various small field sizes were employed. Thirty-two single-arc plans with dose grid sizes of 2 mm and 1 mm and prescribed doses of 2, 5, 10, and 20 Gy were calculated using the AXB algorithm. In total, 128 GPR plans were examined. These plans were categorized into three sub-fields (3SF), four sub-fields (4SF), and six sub-fields (6SF). The GPR results deteriorated with smaller target sizes and a 2 mm dose grid size in a single arc. A similar degradation in GPR was observed with smaller target sizes and a 1 mm dose grid size. However, the 1 mm dose grid size generally resulted in better GPR compared with the 2 mm dose grid size for the same target sizes. The GPR improved with finer split angles and a 2 mm dose grid size in the split-arc method. However, no statistically significant improvement was observed with finer split angles and a 1 mm dose grid size. This study demonstrates that coarser dose grid sizes result in lower GPRs in peripheral low-dose regions as calculated by AXB with ArcCHECK™ measurement. To enhance GPR, employing split-arc methods and finer dose grid sizes could be beneficial.