Portal Imaging Research Articles

A grayscale compression method to segment bone structures for 2D-3D registration of setup images in non-coplanar radiotherapy

Purpose. To evaluate the performance of an automated 2D-3D bone registration algorithm incorporating a grayscale compression method for quantifying patient position errors in non-coplanar radiotherapy. Methods. An automated 2D-3D registration incorporating a grayscale compression method to segment bone structures was proposed. Portal images containing only bone structures ( Portalbone ) and digitally reconstructed radiographs containing only bone structures ( DRRbone ) were used for registration. First, the portal image was filtered by a high-pass finite impulse response (FIR) filter. Then the grayscale range of the filtered portal image was compressed. Thresholds were determined based on the difference in gray values of bone structures in the filtered and compressed portal image to obtain Portalbone. Another threshold was applied to generate DRRbone when the CT image uses the ray-casting algorithm to generate DRR images. The compression performance was assessed by registering the DRRbone with the Portalbone obtained by compressing the portal image into various grayscale ranges. The proposed registration method was quantitatively and visually validated using (1) a CT image of an anthropomorphic head phantom and its portal images obtained in different poses and (2) CT images and pre-treatment portal images of 20 patients treated with non-coplanar radiotherapy. Results. Mean absolute registration errors for the best compression grayscale range test were 0.642 mm, 0.574 mm, and 0.643 mm, with calculation times of 50.6 min, 42.2 min, and 49.6 min for grayscale ranges of 0–127, 0–63 and 0–31, respectively. For the accuracy validation (1), the mean absolute registration errors for couch angles 0°, 45°, 90°, 270°, and 315° were 0.694 mm, 0.839 mm, 0.726 mm, 0.833 mm, and 0.873 mm, respectively. Among the six transformation parameters, the translation error in the vertical direction contributed the most to the registration errors. Visual inspection of the patient registration results revealed success in every instance. Conclusions. The implemented grayscale compression method successfully enhances and segments bone structures in portal images, allowing for accurate determination of patient setup errors in non-coplanar radiotherapy.

Commissioning and dosimetric verification of volumetric modulated arc therapy for multiple modalities using electronic portal imaging device-based 3D dosimetry system: a novel approach.

The purpose of this study was to validate an electronic portal imaging device (EPID) based 3-dimensional (3D) dosimetry system for the commissioning of volumetric modulated arc therapy (VMAT) delivery for flattening filter (FF) and flattening filter free (FFF) modalities based on test suites developed according to American Association of Physicists in Medicine Task Group 119 (AAPM TG 119) and pre-treatment patient specific quality assurance (PSQA).With ionisation chamber, multiple-point measurement in various planes becomes extremely difficult and time-consuming, necessitating repeated exposure of the plan. The average agreement between measured and planned doses for TG plans is recommended to be within 3%, and both the ionisation chamber and PerFRACTION™ measurement were well within this prescribed limit. Both point dose differences with the planned dose and gamma passing rates are comparable with TG reported multi-institution results. From our study, we found that no significant differences were found between FF and FFF beams for measurements using PerFRACTION™ and ion chamber. Overall, PerFRACTION™ produces acceptable results to be used for commissioning and validating VMAT and for performing PSQA. The findings support the feasibility of integrating PerFRACTION™ into routine quality assurance procedures for VMAT delivery. Further multi-institutional studies are recommended to establish global baseline values and enhance the understanding of PerFRACTION™'s capabilities in diverse clinical settings.

Basketball Anterior and Posterior Portal Veins Doppler Imaging of Sports Medicine Technique Exploration

INTRODUCTION: Basketball, as a high-intensity sport, has attracted much attention for its effects on the cardiovascular system of athletes. The anterior and posterior portal veins are some of the vital blood vessels in the human circulatory system, and their blood flow is closely related to the athletes' physical status. Doppler ultrasound technology is widely used in sports medicine and provides a powerful tool for an in-depth understanding of the effects of basketball on portal vein blood flow. This study aimed to explore the potential of sports medicine technology in assessing cardiovascular adaptations in athletes through portal Doppler imaging before and after basketball exercise.OBJECTIVES: The primary objective of this study was to analyze the effects of basketball exercise on portal vein blood flow in athletes before and after basketball exercise through the use of Doppler ultrasound technology. Specifically, this study aimed to explore the dynamics of pre- and post-exercise Doppler imaging of the posterior and posterior veins in order to assess the cardiovascular adaptations of athletes during exercise more comprehensively and objectively.METHODS: A group of healthy professional basketball players were selected as the study subjects, and Doppler ultrasound instruments were utilized to obtain portal Doppler images before, during, and after exercise. The functional status of the vasculature was assessed by analyzing parameters such as portal blood flow velocity and resistance index. At the same time, the physiological parameters of the athletes, such as heart rate and blood pressure, were combined to gain a comprehensive understanding of the effects of basketball on portal blood flow.RESULTS: The results of the study showed that the anterior and posterior portal blood flow velocities of the athletes changed significantly during basketball exercise. Before the exercise, the blood flow velocity was relatively low, while it rapidly increased and reached the peak state during the exercise. After exercise, blood flow velocity gradually dropped back to the baseline level. In addition, the change in resistance index also indicated that portal blood vessels experienced a particular stress and adaptation process during exercise.CONCLUSION: This study revealed the effects of exercise on the cardiovascular system of athletes by analyzing the Doppler images of the portal vein before and after basketball exercise. Basketball exercise leads to significant changes in portal hemodynamics, which provides a new perspective for sports medicine. These findings are of guiding significance for the development of training programs for athletes and the prevention of exercise-related cardiovascular problems and provide a valuable reference for further research in the field of sports medicine.

MR-linac MLC positioning QA by digitally stitching dual double-exposed films.

The picket fence (PF) test is highly recommended for multi-leaf collimator (MLC) quality assurance. However, since the electronic portal imaging device (EPID) on the Elekta Unity only covers a small area, it is not feasible to perform the PF test for the entire MLC. Here, we propose a technique for the PF test by stitching two double-exposed films. Two EBT3 films were used to encompass the entire MLC, with each one covering one half of the area. Two fields were employed to apply double exposure: a PF pattern consisting of 11 2 mm wide pickets and a 2.84cm x 22cm open field. The edges of the open field defined by the diaphragms were used to correct film rotation as well as align them horizontally. The PF pattern was also measured with the EPID where the pickets were used to align the films vertically. Individual leaf positions were detected on the merged film for quantitative analysis. Various MLC positioning errors were introduced to evaluate the technique's sensitivity. The merged films covered 72 leaf pairs properly (four leaf pairs on both sides were outside the treatment couch). With the EPID, the leaf positioning accuracy was -0.02±0.07mm (maximum: 0.29mm) and the picket width variation was 0.00±0.03mm (maximum: 0.11mm); with the films, the position accuracy and width variation were -0.03±0.13mm (maximum: 0.80mm) and 0.00±0.13mm (maximum: 0.74mm), respectively. The EPID was able to detect errors of 0.5mm or above with submillimeter accuracy; the films were only able to detect errors>1.0mm. We developed a quantitative technique for the PF test on the Elekta Unity. The merged films covered nearly the entire MLC leaf banks. The technique exhibited clinically acceptable accuracy and sensitivity to MLC positioning errors.

Noninvasive identification of SOX9 status using radiomics signatures may help construct personalized treatment strategy in hepatocellular carcinoma.

To develop and validate a radiomics-based model for predicting SOX9-positive hepatocellular carcinoma (HCC) using preoperative contrast-enhanced computed tomography (CT) images. From January 2013 to April 2017, patients with histologically proven HCC who received systemic sorafenib treatment after curative resection were retrospectively enrolled. Radiomic features were extracted from portal venous phase CT images and selected to build a radiomics score using logistic regression analysis. The factors associated with SOX9 expression were selected and combined by univariate and multivariate analyses to establish clinico-liver imaging (CL) model and clinico-liver imaging-radiomics (CLR) model. Diagnostic performance was measured by area under curve (AUC). Overall survival (OS) and recurrence-free survival (RFS) rates were compared using Kaplan-Meier method. A total of 108 patients (training cohort: n = 80; validation cohort: n = 28) were enrolled. Multivariate analyses revealed that the albumin-bilirubin grade and tumor size were significant independent factors for predicting SOX9-positive HCCs and were included in the CL model. The CLR model integrating the radiomics score with albumin-bilirubin grade and tumor size showed better discriminative performance than the CL model with AUCs of 0.912 and 0.790 in the training and validation cohorts. Survival curves for RFS and OS showed that SOX9 expression was closely related to the prognosis of HCC patients. RFS and OS rates were significantly lower in patients with SOX9-positive than SOX9-negative (51.02% vs. 75.00% at 1-year RFS rates; 76.92% vs. 94.94% at 2-year OS rates). Radiomics signatures may serve as noninvasive predictors for SOX9 status evaluation in patients with HCC and may aid in constructing individualized treatment strategies.

A Novel Approach for Position Verification and Dose Calculation through Local MVCT Reconstruction.

Traditional positioning verification using cone-beam computed tomography (CBCT) may incur errors due to potential misalignments between the isocenter of CBCT and the treatment beams in radiotherapy. This study introduces an innovative method for verifying patient positioning in radiotherapy. Initially, the transmission images from an electronic portal imaging device (EPID) are acquired from 10 distinct angles. Utilizing the ART-TV algorithm, a sparse reconstruction of local megavoltage computed tomography (MVCT) is performed. Subsequently, this MVCT is aligned with the planning CT via a three-dimensional mutual information registration technique, pinpointing any patient-positioning discrepancies and facilitating corrective adjustments to the treatment setup. Notably, this approach employs the same radiation source as used in treatment to obtain three-dimensional images, thereby circumventing errors stemming from misalignment between the isocenter of CBCT and the accelerator. The registration process requires only 10 EPID images, and the dose absorbed during this process is included in the total dose calculation. The results show that our method's reconstructed MVCT images fulfill the requirements for registration, and the registration algorithm accurately detects positioning errors, thus allowing for adjustments in the patient's treatment position and precise calculation of the absorbed dose.

Feasibility of Ethos adaptive treatments of lung tumors and associated quality assurance.

Online adaptive radiotherapy with Ethos is based on the anatomy determined from daily cone beam computed tomography (CBCT) images. Dose optimization and computation are performed on the density map of a synthetic CT (sCT), a deformable registration of the initial planning CT (pCT) onto the current CBCT. Large density changes as present in the lung region are challenging the system. Treatment plans for Ethos were created and delivered for 1, 2, and 3cm diameter lung lesions in an anthropomorphic phantom, combining different insets in the pCT and during adaptive and non-adaptive treatment sessions. Primary and secondary dose calculations as well as back-projected dose from portal images were evaluated. Density changes due to changed insets were not considered in the sCTs. This resulted in errors in the dose; for example, -15.9% of the mean dose for a plan when changing from a 3cm inset in the pCT to 1cm at the time of treatment. Secondary dose calculation is based on the sCT and could therefore not reveal these dose errors. However, dose calculation on the CBCT, either as a recalculation in the treatment planning system or as pre-treatment quality assurance (QA) before the treatment, indicated the differences. EPID in-vivo QA also reported discrepancies between calculated and delivered dose distributions. An incorrect density distribution in the sCT has an impact on the dose calculation accuracy in the adaptive treatment workflow with the Ethos system. Additional quality checks of the sCT can detect such errors.

Dual-energy CT Portal Venography: Clinical Application Values and Future Opportunities.

Standard multidetector computed tomography (MDCT) uses a single X-ray tube to emit a mixed energy X-ray beam, which is received by a single detector. The difference is that dual-energy CT (DECT), a new equipment in recent years, employs a single X-ray tube or two X-ray tubes to emit two single-energy X-ray beams, which are received by a single or two detectors. The application of dual-energy technology to portal venography has become one of the research hotspots. This paper will elaborate on the clinical application values of DECT portal venography in improving portal vein image quality, distinguishing the nature of portal vein thrombus, reducing contrast agent dose and radiation dose, and will discuss the possibility of its movement from research to routine practice and future development opportunities.

Assessing quality assurance of multi-leaf collimator using the structural similarity index.

This study aims to evaluate the viability of utilizing the Structural Similarity Index (SSI*) as an innovative imaging metric for quality assurance (QA) of the multi-leaf collimator (MLC). Additionally, we compared the results obtained through SSI* with those derived from a conventional Gamma index test for three types of Varian machines (Trilogy, Truebeam, and Edge) over a 12-week period of MLC QA in our clinic. To assess sensitivity to MLC positioning errors, we designed a 1cm slit on the reference MLC, subsequently shifted by 0.5-5mm on the target MLC. For evaluating sensitivity to output error, we irradiated five 25cm×25cm open fields on the portal image with varying Monitor Units (MUs) of 96-100. We compared SSI* and Gamma index tests using three linear accelerator (LINAC) machines: Varian Trilogy, Truebeam, and Edge, with MLC leaf widths of 1, 0.5, and 0.25mm. Weekly QA included VMAT and static field modes, with Picket fence test images acquired. Mechanical uncertainties related to the LINAC head, electronic portal imaging device (EPID), and MLC during gantry rotation and leaf motion were monitored. The Gamma index test started detecting the MLC shift at a threshold of 4mm, whereas the SSI* metric showed sensitivity to shifts as small as 2mm. Moreover, the Gamma index test identified dose changes at 95MUs, indicating a 5% dose difference based on the distance to agreement (DTA)/dose difference (DD) criteria of 1mm/3%. In contrast, the SSI* metric alerted to dose differences starting from 97MUs, corresponding to a 3% dose difference. The Gamma index test passed all measurements conducted on each machine. However, the SSI* metric rejected all measurements from the Edge and Trilogy machines and two from the Truebeam. Our findings demonstrate that the SSI* exhibits greater sensitivity than the Gamma index test in detecting MLC positioning errors and dose changes between static and VMAT modes. The SSI* metric outperformed the Gamma index test regarding sensitivity across these parameters.

Radiomics analysis of contrast-enhanced computerized tomography for differentiation of gastric schwannomas from gastric gastrointestinal stromal tumors

PurposeTo assess the performance of radiomics-based analysis of contrast-enhanced computerized tomography (CE-CT) images for distinguishing GS from gastric GIST.MethodsForty-nine patients with GS and two hundred fifty-three with GIST were enrolled in this retrospective study. CT features were evaluated by two associate chief radiologists. Radiomics features were extracted from portal venous phase images using Pyradiomics software. A non-radiomics dataset (combination of clinical characteristics and radiologist-determined CT features) and a radiomics dataset were used to build stepwise logistic regression and least absolute shrinkage and selection operator (LASSO) logistic regression models, respectively. Model performance was evaluated according to sensitivity, specificity, accuracy, and receiver operating characteristic (ROC) curve, and Delong’s test was applied to compare the area under the curve (AUC) between different models.ResultsA total of 1223 radiomics features were extracted from portal venous phase images. After reducing dimensions by calculating Pearson correlation coefficients (PCCs), 20 radiomics features, 20 clinical characteristics + CT features were used to build the models, respectively. The AUC values for the models using radiomics features and those using clinical features were more than 0.900 for both the training and validation groups. There were no significant differences in predictive performance between the radiomic and clinical data models according to Delong’s test.ConclusionA radiomics-based model applied to CE-CT images showed comparable predictive performance to senior physicians in the differentiation of GS from GIST.

Analysis of clinicopathologic and imaging features of dual-phenotype hepatocellular carcinoma

Dual-phenotype hepatocellular carcinoma (DPHCC) is a new subtype of hepatocellular carcinoma (HCC). This study aimed to investigate the relationship between the computerized tomography scan (CT) imaging and clinicopathologic features of DPHCC. The CT imaging and clinicopathologic data of 97 HCC cases who underwent radical resection were collected retrospectively. The CT imaging feature was evaluated by the ratio of the average CT value of tumor to liver (TLR) in the plain scan, arterial, portal vein and delayed phases. The association between CT imaging and clinicopathologic features was analyzed using the t-test or chi-square test. Univariate and multivariate recurrence-free survival (RFS) analysis and overall survival (OS) were performed. The positive rates of cytokeratin 7 (CK7) and CK19 were 35.1% and 20.6% respectively. The positive rate of CK19 was significantly higher in cases with age < 47 years (P = 0.005), tumor diameter > 4 cm (P = 0.016) or AFP ≥ 400 ng/ml (P = 0.007). The TLR in the portal vein phase was significantly lower in CK19 positive group (P = 0.024). The recurrence risk was significantly higher in cases with CK19 positive (HR: 2.17, 95% CI 1.16 to 4.04, P = 0.013), tumor diameter > 4 cm (HR: 2.05, 95% CI 1.11 to 3.78, P = 0.019), AFP ≥ 400 ng/ml (HR: 2.50, 95% CI 1.37 to 4.54, P = 0.002) or CA199 ≥ 37 U/ml (HR: 2.23, 95% CI 1.12 to 4.42, P = 0.020). However, imaging features, pathological subtype, CK7 or CK19 expression were not significantly related to HCC OS in the univariate and multivariate analysis (all P > 0.05). The expression of CK19 may be associated with the enhancement feature of the portal vein phase CT image, and CK19 positive may suggest a worse RFS.

Multi-sequence MRI radiomics of colorectal liver metastases: Which features are reproducible across readers?

PurposeTo assess the inter-reader reproducibility of radiomics features on multiple MRI sequences after segmentations of colorectal liver metastases (CRLM). Method30 CRLM (in 23 patients) were manually delineated by three readers on MRI before the start of chemotherapy on the contrast enhanced T1-weighted images (CE-T1W) in the portal venous phase, T2-weighted images (T2W) and b800 diffusion weighted images (DWI). DWI delineations were copied to the ADC-maps. 107 radiomics features were extracted per sequence. The intraclass correlation coefficient (ICC) was calculated per feature. Features were considered reproducible if ICC > 0.9. Results90% of CE-T1W features were reproducible with a median ICC of 0.98 (range 0.76–1.00). 81% of DWI features were robust with median ICC = 0.97 (range 0.38–1.00). The T2W features had a median ICC of 0.96 (range 0.55–0.99) and were reproducible in 80%. ADC showed the lowest number of reproducible features with 58% and median ICC = 0.91 (range 0.38–0.99) When considering the lower bound of the ICC 95% confidence intervals, 58%, 66%, 54% and 29% reached 0.9 for the CE-T1W, DWI, T2W and ADC features, respectively. The feature class with the best reproducibility differed per sequence. ConclusionsThe majority of MRI radiomics features from CE-T1W, T2W, DWI and ADC in colorectal liver metastases were robust for segmentation variability between readers. The CE-T1W yielded slightly better reproducibility results compared to DWI and T2W. The ADC features seem more susceptible to reader differences compared to the other three sequences.

Uncertainties in prostate localization: A study in a laboratory setting.

Purpose: The main goal of this study is to minimize systematic errors by utilizing linear regression to determine the position of gold seed markers that were implanted in a wax phantom, using EPID images in a controlled laboratory setting. Materials and Methods: Pairs of orthogonal portal images were generated and carefully examined, while the phantom was subjected to a known displacement in either the lateral or superior-inferior direction. The measured seed displacement was then compared with the average seed displacement determined from the orthogonal EPID images. The accuracy of the inferred seed positions from the portal images was evaluated using linear regression, concordance correlation coefficient (CCC), intraclass correlation coefficients (ICC), and Bland-Altman plots. Results: The average displacement errors were found to be 3.2 mm, 2.6 mm, 2.4 mm, and 6.7 mm in the lateral left, lateral right, cranio-caudal (inward), and cranio-caudal (outward) directions, respectively. Conclusion: The results suggest that utilizing linear regression can significantly reduce the systematic errors described in this study and are consistent with previously published findings on setup errors.

Optimization of a commercial portal dose image prediction algorithm for pre-treatment verifications of plans using unflattened photon beams.

The aim was to improve the portal dosimetry-based quality assurance results of conventional treatment plans by adjusting the multileaf collimator (MLC) dosimetric leaf gap (DLG) and transmission (T) values of the anisotropic analytic algorithm (AAA) used for portal dose image prediction (PDIP). The AAA-based PDIP v. 16.1 algorithm (PDIP-AAA) of the Eclipse TPS was configured for 6 MV FFF energy. Optimal DLG and T values were achieved for this algorithm by comparing predicted versus measured portal images of the Chair pattern. Twenty clinical plans using 6 MV FFF beams were verified using the optimal PDIP-AAA algorithm and the standard PDIP v. 16 algorithm (PDIP-vE), configured using the van Esch package. The 3% global/2 mm gamma passing rates (GPRs) and average gamma indexes (AGIs) were computed for each acquired image. For each plan, the mean GPR (GPRmean) and mean GAI (GAImean) were compared for both algorithms. A 2-tailed Student t-test (α = 0.05) was used to evaluate whether there was a statistically significant difference. Optimal values of DLG = 0.1 mm and T = 0.01 were found for the PDIP-AAA algorithm, providing significantly better values of GPRmean and AGImean than PDIP-vE (p < 0.001). All plans verified with PIDP-AAA showed GPRmean ≥ 95%. In contrast, only 45% of the plans reported GPRmean ≥ 95% with the PDIP-vE algorithm. The MLC parameters available in the PDIP-AAA model must be tuned to improve the accuracy of the predicted dose image. This work-around is not possible using the standard PDIP algorithm. The adjusted PDIP-AAA resulted in significantly better results than PDIP-vE.

In silico study of the electronic portal imaging devices sensitivity to parotid glands shrinkage during radiotherapy for nasopharyngeal cases

This computational investigation examines the capability of electronic portal imaging devices (EPIDs) in detecting anatomical changes during radiation therapy. A photon beam for 6 MV was simulated using compliant phase-space files via BEAMnrc (an EGSnrc user code). Initially, the dose calculation was carried out using DOSXYZnrc, an EGSnrc user code, on a water-based calibration phantom made of water. Phantoms, constructed utilizing patient computed tomography data, were subsequently used to simulate clinically relevant parotid gland (PG) shrinkage through image deformation using ImSimQA software. Two EPID geometries were implemented; a simple water slab, and a comprehensive multi-layer representation of the EPID (referred to as the 17-slab model). Using a 1%/1 mm gamma index the 17-slab geometry was sensitive to volume reductions in the PG ≥ − 26% from its original volume, while the water slab detected volume reduction of ≥ − 28.5%. In a clinical setting, replanning of head and neck patients is initiated when a reduction of 24–30% in PG volume is detected through cone-beam computed tomography. The sensitivity index, indicative of the signal to error ratio between the reference and evaluated images, showed an increase as the gland volume decreased for both models. Notably, the 17-slab model consistently more sensitive than the water slab. This computational study highlights the prospective use of EPIDs in monitoring and detecting clinically significant volume changes in the PG during radiation treatment. Advances in knowledge: While EPIDs are frequently employed for patient setup and alignment, this computational work constructed a virtual EPID to assess its detection abilities for the anatomical changes in the PG due to radiation exposure in head and neck cancer cases.

CT-based radiomics signature for differentiating pyelocaliceal upper urinary tract urothelial carcinoma from infiltrative renal cell carcinoma.

To develop a CT-based radiomics model and a combined model for preoperatively discriminating infiltrative renal cell carcinoma (RCC) and pyelocaliceal upper urinary tract urothelial carcinoma (UTUC), which invades the renal parenchyma. Eighty patients (37 pathologically proven infiltrative RCCs and 43 pathologically proven pyelocaliceal UTUCs) were retrospectively enrolled and randomly divided into a training set (n = 56) and a testing set (n = 24) at a ratio of 7:3. Traditional CT imaging characteristics in the portal venous phase were collected by two radiologists (SPH and ZXL, who have 4 and 30 years of experience in abdominal radiology, respectively). Patient demographics and traditional CT imaging characteristics were used to construct the clinical model. The radiomics score was calculated based on the radiomics features extracted from the portal venous CT images and the random forest (RF) algorithm to construct the radiomics model. The combined model was constructed using the radiomics score and significant clinical factors according to the multivariate logistic regression. The diagnostic efficacy of the models was evaluated using receiver operating characteristic (ROC) curve analysis and the area under the curve (AUC). The RF score based on the eight validated features extracted from the portal venous CT images was used to build the radiomics model. Painless hematuria as an independent risk factor was used to build the clinical model. The combined model was constructed using the RF score and the selected clinical factor. Both the radiomics model and combined model showed higher efficacy in differentiating infiltrative RCC and pyelocaliceal UTUC in the training and testing cohorts with AUC values of 0.95 and 0.90, respectively, for the radiomics model and 0.99 and 0.90, respectively, for the combined model. The decision curves of the combined model as well as the radiomics model indicated an overall net benefit over the clinical model. Both the radiomics model and the combined model achieved a notable reduction in false-positive and false-negativerates, resulting in significantly higher accuracy compared to the visual assessments in both the training and testing cohorts. The radiomics model and combined model had the potential to accurately differentiate infiltrative RCC and pyelocaliceal UTUC, which invades the renal parenchyma, and provide a new potentially non-invasive method to guide surgery strategies.

Automated MRI liver segmentation for anatomical segmentation, liver volumetry, and the extraction of radiomics.

To develop and evaluate a deep convolutional neural network (DCNN) for automated liver segmentation, volumetry, and radiomic feature extraction on contrast-enhanced portal venous phase magnetic resonance imaging (MRI). This retrospective study included hepatocellular carcinoma patients from an institutional database with portal venous MRI. After manual segmentation, the data was randomly split into independent training, validation, and internal testing sets. From a collaborating institution, de-identified scans were used for external testing. The public LiverHccSeg dataset was used for further external validation. A 3D DCNN was trained to automatically segment the liver. Segmentation accuracy was quantified by theDice similarity coefficient (DSC) with respect to manual segmentation. A Mann-Whitney U test was used to compare the internal and external test sets. Agreement of volumetry and radiomic features was assessed using the intraclass correlation coefficient (ICC). In total, 470 patients met the inclusion criteria (63.9±8.2 years; 376 males) and 20 patients were used for external validation (41±12 years; 13 males). DSC segmentation accuracy of the DCNN was similarly high between the internal (0.97±0.01) and external (0.96±0.03) test sets (p=0.28) and demonstrated robust segmentation performance on public testing (0.93±0.03). Agreement of liver volumetry was satisfactory in the internal (ICC, 0.99), external (ICC, 0.97), and public (ICC, 0.85) test sets. Radiomic features demonstrated excellent agreement in the internal (mean ICC, 0.98±0.04), external (mean ICC, 0.94±0.10), and public (mean ICC, 0.91±0.09) datasets. Automated liver segmentation yields robust and generalizable segmentation performance on MRI data and can be used for volumetry and radiomic feature extraction. Liver volumetry, anatomic localization, and extraction of quantitative imaging biomarkers require accurate segmentation, but manual segmentation is time-consuming. A deep convolutional neural network demonstrates fast and accurate segmentation performance on T1-weighted portal venous MRI. • This deep convolutional neural network yields robust and generalizable liver segmentation performance on internal, external, and public testing data. • Automated liver volumetry demonstrated excellent agreement with manual volumetry. • Automated liver segmentations can be used for robust and reproducible radiomic feature extraction.

Practical use of the central venous access port for contrast-enhanced CT: comparison with peripheral intravenous access regarding enhancement and safety

To evaluate the efficacy of using the central venous (CV) port compared with peripheral intravenous access for contrast-material injection for contrast enhancement during the portal venous phase. Patients were divided into three groups: CV delay, CV routine, and peripheral access (PA) groups. Patients in the CV delay group underwent injection in the arm-down position with an additional delay, while those in the CV routine and PA groups underwent injections with the routine injection protocol for portal venous phase imaging. Contrast enhancement was evaluated by measuring the mean radiodensity (Hounsfield units) values for the aortic arch, abdominal aorta, inferior vena cava, portal vein, and spleen. The peak injection pressure was recorded and compared among the three groups. No complications related to power injection were observed during 119 contrast-material injections performed using the CV port device. The CV delay group showed significantly lower radiodensity values than the PA group (165.7±20.1 versus 181±19 HU [p<0.01] for the portal vein); however, no significant differences in mean radiodensity values were observed between the CV routine and PA groups (p>0.05). The median peak injection pressure was 73.5, 67, and 47 psi in the CV delay, CV routine, and PA groups, respectively (p<0.01). The CV port can be used for safe contrast-material injection while maintaining contrast enhancement on portal venous phase comparable to that achieved with peripheral intravenous access.

Comparison of deep learning models for building two-dimensional non-transit EPID Dosimetry on Varian Halcyon.

This study compared the effectiveness of five deep learning models in constructing non-transit dosimetry with an a-Si electronic portal imaging device (EPID) on Varian Halcyon. Deep learning model is increasingly used to support prediction and decision-making in several fields including oncology and radiotherapy. Forty-seven unique plans of data obtained from breast cancer patients were calculated using Eclipse treatment planning system (TPS) and extracted from DICOM format as the ground truth. Varian Halcyon was then used to irradiate the a-Si 1200 EPID detector without an attenuator. The EPID and TPS images were augmented and divided randomly into two groups of equal sizes to distinguish the validation and training-test data. Five different deep learning models were then created and validated using a gamma index of 3%/3 mm. Four models successfully improved the similarity of the EPID images and the TPS-generated planned dose images. Meanwhile, the mismatch of the constituent components and number of parameters could cause the models to produce wrong results. The average gamma pass rates were 90.07 ± 4.96% for A-model, 77.42 ± 7.18% for B-model, 79.60 ± 6.56% for C-model, 80.21 ± 5.88% for D-model, and 80.47 ± 5.98% for E-model. The deep learning model is proven to run fast and can increase the similarity of EPID images with TPS images to build non-transit dosimetry. However, more cases are needed to validate this model before being used in clinical activities.

Feasibility Study of Computed Tomographic Radiomics Model for the Prediction of Early and Intermediate Stage Hepatocellular Carcinoma Using BCLC Staging.

Hepatocellular carcinoma (HCC) is a serious health concern because of its high morbidity and mortality. The prognosis of HCC largely depends on the disease stage at diagnosis. Computed tomography (CT) image textural analysis is an image analysis technique that has emerged in recent years. To probe the feasibility of a CT radiomic model for predicting early (stages 0, A) and intermediate (stage B) HCC using Barcelona Clinic Liver Cancer (BCLC) staging. A total of 190 patients with stages 0, A, or B HCC according to CT-enhanced arterial and portal vein phase images were retrospectively assessed. The lesions were delineated manually to construct a region of interest (ROI) consisting of the entire tumor mass. Consequently, the textural profiles of the ROIs were extracted by specific software. Least absolute shrinkage and selection operator dimensionality reduction was used to screen the textural profiles and obtain the area under the receiver operating characteristic curve values. Within the test cohort, the area under the curve (AUC) values associated with arterial-phase images and BCLC stages 0, A, and B disease were 0.99, 0.98, and 0.99, respectively. The overall accuracy rate was 92.7%. The AUC values associated with portal vein phase images and BCLC stages 0, A, and B disease were 0.98, 0.95, and 0.99, respectively, with an overall accuracy of 90.9%. The CT radiomic model can be used to predict the BCLC stage of early-stage and intermediate-stage HCC.