Performance of the Iterative OSEM and HYPER Algorithm for Total-body PET at SUVmax with a Low 18F-FDG Activity, a Short Acquisition Time and Small Lesions.

OBJECTIVE: The primary objective of this comparative investigation was to examine the qualitative attributes of image reconstructions utilizing two distinct algorithms, namely OSEM and HYPER Iterative, in total-body 18F- FDG PET/CT under various acquisition durations and injection activities. METHODS: An initial assessment was executed using a NEMA phantom to compare image quality engendered by OSEM and HYPER Iterative algorithms. Parameters such as BV, COV, and CRC were meticulously evaluated. Subsequently, a prospective cohort study was conducted on 50 patients, employing both reconstruction algorithms. The study was compartmentalized into distinct acquisition time and dosage groups. Lesions were further categorized into three size-based groups. Quantifiable metrics including SD of noise, SUVmax, SNR, and TBR were computed. Additionally, the differences in values, namely ΔSUVmax, ΔTBR, %ΔSUVmax, %ΔSD, and %ΔSNR, between OSEM and HYPER Iterative algorithms were also calculated. RESULTS: The HYPER Iterative algorithm showed reduced BV and COV compared to OSEM in the phantom study, with constant acquisition time. In the clinical study, lesion SUVmax, TBR, and SNR were significantly elevated in images reconstructed using the HYPER Iterative algorithm in comparison to those generated by OSEM (p < 0.001). Furthermore, an amplified increase in SUVmax was predominantly discernible in lesions with dimensions less than 10 mm. Metrics such as %ΔSNR and %ΔSD in HYPER Iterative exhibited improvements correlating with reduced acquisition times and dosages, wherein a more pronounced degree of enhancement was observable in both ΔSUVmax and ΔTBR. CONCLUSION: The HYPER Iterative algorithm significantly improves SUVmax and reduces noise level, with particular efficacy in lesions measuring ≤ 10 mm and under conditions of abbreviated acquisition times and lower dosages.

The efficacy of short acquisition time using 18F-FDG total-body PET/CT for the identification of pediatric epileptic foci.

BACKGROUND: 18F-FDG positron emission tomography (PET) plays a crucial part in the evaluation for pediatric epileptic patients prior to therapy. Short-term scanning holds significant importance, especially for pediatrics epileptic individuals who exhibited involuntary movements. The aim was to evaluate the effects of short acquisition time on image quality and lesion detectability in pediatric epileptic patients using total-body (TB) PET/CT. A total of 25 pediatric patients who underwent TB PET/CT using uEXPLORER scanner with an 18F-FDG administered dose of 3.7 MBq/kg and an acquisition time of 600 s were retrospectively enrolled. Short acquisition times (60 s, 150 and 300 s) were simulated by truncating PET data in list mode to reduce count density. Subjective image quality was scored on a 5-point scale. Regions of interest analysis of suspected epileptogenic zones (EZs), corresponding locations contralateral to EZs, and healthy cerebellar cortex were used to compare the semi-quantitative uptake indices of short-time images and then were compared with 600 s images. The comparison of EZs detectability based on time-dependent PET images was performed. RESULTS: Our study demonstrated that a short acquisition time of 150 s is sufficient to maintain subjective image quality and lesion significance. Statistical analysis revealed no significant difference in subjective PET image quality between imaging at 300 s and 150 s (P > 0.05). The overall impression scores of image quality and lesion conspicuity in G60s were both greater than 3 (overall quality, 3.21 ± 0.46; lesion conspicuity, 4.08 ± 0.74). As acquisition time decreased, the changes of SUVmax and SD in the cerebellar cortex gradually increased (P < 0.01). There was no significant difference in asymmetry index (AI) difference between the groups and the AIs of EZs were > 15% in all groups. In 26 EZs of 25 patients, the lesion detection rate was still 100% when the time was reduced to 60 s. CONCLUSIONS: This study proposed that TB PET/CT acquisition time could be reduced to 60 s with acceptable lesion detectability. Furthermore, it was suggested that a 150 s acquisition time would be sufficient to achieve diagnostic performance and image quality for children with epilepsy.

Improvement of Prediction Performance for Radiation Pneumonitis by Using 3-Dimensional Dosiomic Features.

INTRODUCTION: Patients with early non-small-cell lung cancer (NSCLC) have a relatively long survival time after stereotactic body radiation therapy (SBRT). Predicting radiation-induced pneumonia (RP) has important clinical and social implications for improving the quality of life of such patients. This study developed an RP prediction model by using 3-dimensional (3D) dosiomic features. The model can be used to guide radiation therapy to reduce toxicity. METHODS: Radiomic features were extracted from pre-treatment CT, dose-volume histogram (DVH) parameters and dosiomic features were extracted from the 3D dose distribution of 140 lung cancer patients. Four predictive models: (1) CT; (2) CT + DVH; (3) CT + Rtdose; and (4) Hybrid, CT + DVH + Rtdose, were trained to predict symptomatic RP by extremely randomized trees. Accuracy, sensitivity, specificity, and area under the receiver operator characteristic curve were evaluated. RESULT: Results showed that the fraction regimen was correlated with symptomatic RP (P < .001). The proposed model achieved promising prediction results. The performance metrics for CT, CT + DVH, CT + Rtdose, and Hybrid were as follows: accuracy: 0.786, 0.821, 0.821, and 0.857; sensitivity: 0.625, 1, 0.875, and 1; specificity: 0.8, 0.565, 0.5, and 0.875; and area under the receiver operator characteristic curve: 0.791, 0.809, 0.907, and 0.920, respectively. CONCLUSION: Dosiomic features can improve the performance of the predictive model for symptomatic RP compared with that obtained with the CT + DVH model. The model proposed in this study can help radiation oncologists individually predict the incidence rate of RP.

Dosimetry and treatment efficiency of SBRT using TaiChiB radiotherapy system for two-lung lesions with one overlapping organs at risk.

Purpose: This study aims to assess the dosimetry and treatment efficiency of TaiChiB-based Stereotactic Body Radiotherapy (SBRT) plans applying to treat two-lung lesions with one overlapping organs at risk. Methods: For four retrospective patients diagnosed with two-lung lesions each patient, four treatment plans were designed including Plan Edge, TaiChiB linac-based, RGS-based, and a linac-RGS hybrid (Plan TCLinac, Plan TCRGS, and Plan TCHybrid). Dosimetric metrics and beam-on time were employed to evaluate and compare the TaiChiB-based plans against Plan Edge. Results: For Conformity Index (CI), Plan TCRGS outperformed all other plans with an average CI of 1.06, as opposed to Plan Edge's 1.33. Similarly, for R50 %, Plan TCRGS was superior with an average R50 % of 3.79, better than Plan Edge's 4.28. In terms of D2 cm, Plan TCRGS also led with an average of 48.48%, compared to Plan Edge's 56.25%. For organ at risk (OAR) sparing, Plan TCRGS often displayed the lowest dosimetric values, notably for the spinal cord (Dmax 5.92 Gy) and lungs (D1500cc 1.00 Gy, D1000cc 2.61 Gy, V10 Gy 15.14%). However, its high Dmax values for the heart and great vessels sometimes exceeded safety thresholds. Plan TCHybrid presented a balanced approach, showing doses comparable to or better than Plan Edge without crossing safety limits. In terms of beam-on time, Plan TCLinac emerged as the most efficient treatment option in three out of four cases, followed closely by Plan Edge in one case. Plan TCRGS, despite its dosimetric advantages, was the least efficient, recording notably longer beam-on times, with a peak at 33.28 minutes in Case 2. Conclusion: For patients with two-lung lesions treated by SBRT whose one lesion overlaps with OARs, the Plan TCHybrid delivered by TaiChiB digital radiotherapy system can be recommended as a clinical option.

Metabolic interactions between organs in overweight and obesity using total-body positron emission tomography.

BACKGROUND AND OBJECTIVES: Overweight and obesity is a complex condition resulting from unbalanced energy homeostasis among various organs. However, systemic abnormalities in overweight and obese people are seldom explored in vivo by metabolic imaging techniques. The aim of this study was to determine metabolic abnormities throughout the body in overweight and obese adults using total-body positron emission tomography (PET) glucose uptake imaging. METHODS: Thirty normal weight subjects [body mass index (BMI) < 25 kg/m2, 55.47 ± 13.94 years, 16 men and 14 women], and 26 overweight and obese subjects [BMI ≥ 25 kg/m2, 52.38 ± 9.52 years, 21 men and 5 women] received whole-body 18F-fluorodeoxyglucose PET imaging using the uEXPLORER. Whole-body standardized uptake value normalized by lean body mass (SUL) images were calculated. Metabolic networks were constructed based on the whole-body SUL images using covariance network approach. Both group-level and individual-level network differences between normal weight and overweight/obese subjects were evaluated. Correlation analysis was conducted between network properties and BMI for the overweight/obese subjects. RESULTS: Compared with normal weight subjects, overweight/obese subjects exhibited altered network connectivity strength in four network nodes, namely the pancreas (p = 0.033), spleen (p = 0.021), visceral adipose tissue (VAT) (p = 1.12 × 10-5) and bone (p = 0.021). Network deviations of overweight/obese subjects from the normal weight were positively correlated with BMI (r = 0.718, p = 3.64 × 10-5). In addition, overweight/obese subjects experienced altered connections between organs, and some of the altered connections, including pancreas-right lung and VAT-bilateral lung connections were significantly correlated with BMI. CONCLUSION: Overweight/obese individuals exhibit metabolic alterations in organ level, and altered metabolic interactions at the systemic level. The proposed approach using total-body PET imaging can reveal individual metabolic variability and metabolic deviations between organs, which would open up a new path for understanding metabolic alterations in overweight and obesity.

AS-NeSt: A Novel 3D Deep Learning Model for Radiation Therapy Dose Distribution Prediction in Esophageal Cancer Treatment With Multiple Prescriptions.

PURPOSE: Implementing artificial intelligence technologies allows for the accurate prediction of radiation therapy dose distributions, enhancing treatment planning efficiency. However, esophageal cancers present unique challenges because of tumor complexity and diverse prescription types. Additionally, limited data availability hampers the effectiveness of existing artificial intelligence models. This study developed a deep learning model, trained on a diverse data set of esophageal cancer prescriptions, to improve dose prediction accuracy. METHODS AND MATERIALS: We retrospectively collected data from 530 patients with esophageal cancer, including single-target and simultaneous integrated boost prescriptions, for model building. The proposed Asymmetric ResNeSt (AS-NeSt) model features novel 3-dimensional (3D) ResNeSt blocks and an asymmetrical architecture. We constructed a loss function targeting global and local doses and validated the model's performance against existing alternatives. Model-assisted experiments were used to validate its clinical benefits. RESULTS: The AS-NeSt model maintained an absolute prediction error below 5% for each dosimetric metric. The average Dice similarity coefficient for isodose volumes was 0.93. The model achieved an average relative prediction error of 2.02%, statistically lower than Hierarchically Densely Connected U-net (4.17%), DoseNet (2.35%), and Densely Connected Network (3.65%). It also demonstrated significantly fewer parameters and shorter prediction times. Clinically, the AS-NeSt model raised physicians' ability to accurately preassess appropriate treatment methods before planning from 95.24% to 100%, reduced planning time by over 61% for junior dosimetrists and 52% for senior dosimetrists, and decreased both inter- and intra-dosimetrist discrepancies by more than 50%. CONCLUSIONS: The AS-NeSt model, developed with innovative 3D ResNeSt blocks and an asymmetrical encoder-decoder structure, has been validated using clinical esophageal cancer patient data. It accurately predicts 3D dose distributions for various prescriptions, including simultaneous integrated boost, showing potential to improve the management of esophageal cancer treatment in a clinical setting.

Investigation of the quarter-dose 18 F-FDG total-body PET in routine clinical practice and its clinical value.

OBJECTIVE: The purpose of the study was to evaluate the routine clinical application of total-body PET with quarter-dose 18 F-FDG. METHODS: The contrast recovery coefficient (CRC) and coefficient of variation (COV) were evaluated among full-, half-, and quarter-dose groups with an acquisition duration of 10-, 5-, 3-, and 1-min in the NEMA (IQ) phantom test. Fifty patients undergoing total-body PET/CT with quarter-dose (0.925MBq/kg) of 18 F-FDG were included in the prospective study. The acquisition time was 10 min, divided into duration groups of 5-, 3-, and 1-min, referred to as G10, G5, G3, and G1. Visual scores were assessed based on overall visual assessment, noise scoring, and lesion conspicuity. Lesion SUV max and TBR were evaluated in semi-quantitative analysis. G10 was used as the gold reference to evaluate lesion detectability. RESULTS: In the phantom study, the COV value of the images with quarter-dose 18 F-FDG and 10-min acquisition time was 11.52%. For spheres with 10 mm diameter, the CRC of quarter-dose PET images was relatively stable compared to that of full-dose groups with all acquisition durations. In the human study, the visual score in G10, G5, and G3 was significantly higher than that in G1. The differences in lesion SUV max and TBR for G1-G10 were significantly higher than that for G5-G10 and G3-G10. All lesions in G10 could be identified in G5 and G3. CONCLUSION: The phantom and human findings demonstrated the feasibility of quarter-dose 18 F-FDG PET with 3-min acquisition time, which can maintain image quality with reduced radiation dose.

Rabbit systemic glucose metabolism map by total-body dynamic PET/CT technology.

BACKGROUND: This study evaluated total-body glucose metabolism in a preclinical lab animal, the rabbit, by employing a dynamic glucose metabolic image obtained with total-body fluorine-18 fluorodeoxyglucose ( 18 F-FDG) PET/computed tomography (PET/CT). METHODS: The dynamic total-body PET/CT system was used to obtain glucose metabolic imaging from 10 sedated body-matched rabbits. The standard uptake value (SUV) of 18 F-FDG was used to evaluate glucose metabolism. In addition, the correlation between glucose metabolism and sexes was assessed, as well as metabolic differences between left- and right sides. RESULTS: We found significant distribution heterogeneity of glucose in several organs across the entire body. There were no significant metabolic differences between sexes and between bilateral sides in the 10 rabbits. Thereafter, we assayed the major organ SUV changes by dynamic PET/CT of the major organs. The heart, liver, and urinary system showed more 18 F-FDG, whereas the skeletal muscle, brain, spinal cord, and lungs incorporated less 18 F-FDG. The phenotype of 18 F-FDG uptake was highly correlated with the physiological functions. The 18 F-FDG accumulation in urinary system were observed which could reflect the renal parenchyma glucose metabolism indirectly. However, the low 18 F-FDG uptake in the brain and spinal cord was due to sedation. CONCLUSION: The total-body glucose metabolic atlas depicted with 18 F-FDG dynamic PET/CT may be used as a reference for assessing pathological 18 F-FDG uptake. Furthermore, this study could be a reference for preclinical research involving abnormality of glucose metabolism.

The feasibility of quantitative assessment of dynamic 18F-fluorodeoxyglucose PET in Takayasu's arteritis: a pilot study.

PURPOSE: PET has been demonstrated to be sensitive for detecting active inflammation in Takayasu's arteritis (TAK) patients, but semi-quantitative-based assessment may be susceptible to various biological and technical factors. Absolute quantification via dynamic PET (dPET) may provide a more reliable and quantitative assessment of TAK-active arteries. The purpose of this study was to investigate the feasibility and efficacy of dPET in quantifying TAK-active arteries compared to static PET. MATERIALS AND METHODS: This prospective study enrolled 10 TAK-active patients (fulfilled the NIH criteria) and 5 control participants from March to October 2022. One-hour dPET scan (all TAK and control participants) and delayed static PET scan at 2-h (all TAK patients) were acquired. For 1-h static PET, summed images from 50 to 60 min of the dPET were extracted. PET parameters derived from 1- and 2-h static PET including SUV (SUV1H and SUV2H), target-to-background ratio (TBR) (TBR1H and TBR2H), net influx rate (Ki), and TBRKi extracted from dPET were obtained. The detectability of TAK-active arteries was compared among different scanning methods using the generalized estimating equation (GEE) with a logistic regression with repeated measures, and the GEE with gamma distribution and log link function was used to evaluate the different study groups or scanning methods. RESULTS: Based on the disease states, 5 cases of TAK were classified as untreated and relapsed, respectively. The SUVmax on 2-h PET was higher than that on 1-h PET in the untreated patients (P < 0.05). However, no significant differences were observed in the median SUVmax between 1-h PET and 2-h PET in the relapsed patients (P > 0.05). The TBRKi was significantly higher than both TBR1H and TBR2H (all P < 0.001). Moreover, the detectability of TAK-active arteries by dPET-derived Ki was significantly higher than 1-h and 2-h PET (all P < 0.001). Significant differences were observed in Kimax, SUVmax-1H, TBR1H, and TBRKi among untreated, relapsed, and control groups (all P < 0.05). CONCLUSIONS: Absolute quantitative assessment by dPET provides an improved sensitivity and detectability in both visualization and quantification of TAK-active arteries. This elucidates the clinical significance of dPET in the early detection of active inflammation and monitoring recurrence.

A planning strategy may reduce the risk of heart diseases and radiation pneumonia: Avoiding the specific heart substructures.

BACKGROUND: Dose to heart substructures is a better predictor for major adverse cardiac events (MACE) than mean heart dose (MHD). We propose an avoidance planning strategy for important cardiac substructures. MATERIAL AND METHODS: Two plans, clinical and cardiac substructure-avoidance plan, were generated for twenty patients. Five dose-sensitive substructures, including left ventricle, pulmonary artery, left anterior descending branch, left circumflex branch and the coronary artery were chosen. The avoidance plan aims to meet the target criteria and organ-at-risk (OARs) constraints while minimizing the dose parameters of the above five substructures. The dosimetric assessments included the mean dose and the maximum dose of cardiac substructures and several volume parameters. In addition, we also evaluated the relative risk of coronary artery disease (CAD), chronic heart failure (CHF), and radiation pneumonia (RP). RESULTS: Pearson correlation coefficient and R2 value of linear regression fitting demonstrated that MHD had poor prediction ability for the mean dose of the cardiac substructures. Compared to clinical plans, an avoidance plan is able to statistically significantly decrease the dose to key substructures. Meanwhile, the dose to OARs and the coverage of the target are comparable in the two plans. In addition, it can be observed that the avoidance plan statistically decreases the relative risks of CAD, CHF, and RP. CONCLUSIONS: The substructure-avoidance planning strategy that incorporates the cardiac substructures into optimization process, can protect the important heart substructures, such as left ventricle, left anterior descending branch and pulmonary artery, achieving the substantive sparing of dose-sensitive cardiac structures, and have the potential to decrease the relative risks of CAD, CHF, and RP.

Glucose uptake and distribution across the human skeleton using state-of-the-art total-body PET/CT.

A growing number of studies have demonstrated that the skeleton is an endocrine organ that is involved in glucose metabolism and plays a significant role in human glucose homeostasis. However, there is still a limited understanding of the in vivo glucose uptake and distribution across the human skeleton. To address this issue, we aimed to elucidate the detailed profile of glucose uptake across the skeleton using a total-body positron emission tomography (PET) scanner. A total of 41 healthy participants were recruited. Two of them received a 1-hour dynamic total-body 18F-fluorodeoxyglucose (18F-FDG) PET scan, and all of them received a 10-minute static total-body 18F-FDG PET scan. The net influx rate (Ki) and standardized uptake value normalized by lean body mass (SUL) were calculated as indicators of glucose uptake from the dynamic and static PET data, respectively. The results showed that the vertebrae, hip bone and skull had relatively high Ki and SUL values compared with metabolic organs such as the liver. Both the Ki and SUL were higher in the epiphyseal, metaphyseal and cortical regions of long bones. Moreover, trends associated with age and overweight with glucose uptake (SULmax and SULmean) in bones were uncovered. Overall, these results indicate that the skeleton is a site with significant glucose uptake, and skeletal glucose uptake can be affected by age and dysregulated metabolism.

Optimization of isocenter position for multiple brain metastases single-isocenter stereotactic radiosurgery to minimize dosimetric variations due to rotational uncertainty.

PURPOSE: This paper studied a novel calculation framework that can determine the optimal value isocenter position of single isocenter SRS treatment plan for multiple brain metastases, in order to minimize the dosimetric variations caused by rotational uncertainty. MATERIALS AND METHODS: 21 patients with 2-4 GTVswho received SRS treatment for multiple brain metastases in our institution were selected for the retrospective study. The PTVwas obtained by expanding GTV 1 mm isotropic margin. We studied a stochastic optimization framework, which determined the optimal value isocenter location by maximizing the average target dose coverageCtarget,meanwith a rotation error of no more than 1°. We evaluated the performance of the optimal isocenter by comparing theCtarget,meanand average dice similarity coefficient (DSC)with the optimal value and the center of mass (CM) respectively as the treatment isocenter. The extra PTV margin to achieve 100% target dose coverage was calculated by our framework. RESULTS: Compared to the CM method, the optimal value isocenter method increased the averageCtarget,meanof all targets from 97.0% to 97.7%and the average DSC from 0.794to 0.799. Throughout all the cases, the average extra PTV margin to obtain full target dose coverage was 0.7 mmwhen using the optimal value isocenter as the treatment isocenter. CONCLUSION: We studied a novel computational framework using stochastic optimization to determine the optimal isocenter position of SRS treatment plan for multiple brain metastases. At the same time, our framework gave the extra PTV margin to obtain full target dose coverage.

Role of breath-hold lung PET in stage IA pulmonary adenocarcinoma.

BACKGROUND: Respiratory motion during PET acquisition may result in image blurring and resolution loss, reduced measurement of radiotracer uptake, and consequently, inaccurate lesion quantification and description. With the introduction of the total-body PET system, short-time PET acquisition is feasible due to its high sensitivity and spatial resolution. The purpose of this study was to evaluate the additional value of 20-s breath-hold (BH) lung PET in patients with stage IA pulmonary adenocarcinoma. METHODS: Forty-seven patients with confirmed stage IA pulmonary adenocarcinoma were enrolled in this retrospective study. All patients underwent a 300-s FB whole-body PET, followed by a BH lung PET. The SUVmax, TBR of the lesions and the percentage difference in nodule SUVmax (%ΔSUVmax) and TBR (%ΔTBR) between the two acquisitions was also calculated. The lesions were further divided by distance from pleura for subgroup analysis. The lesion detectability on PET images was the percentage of FDG-positive lesions. RESULTS: Among 47 patients, the BH lung PET images identified all lung nodules, and there was a significant difference in overall nodule SUVmax and TBR between BH PET and FB PET (both p < 0.01). The %ΔSUVmax and %ΔTBR were significantly higher in nodules adjacent to pleura (≤ 10 mm in distance) than those away from pleura (both p < 0.05). The lesion detectability of BH lung PET was significantly higher than that of FB PET (p < 0.01). CONCLUSION: BH PET acquisition is a practical way to minimize motion artifacts in PET which has the potential to improve lesion detection for stage IA pulmonary adenocarcinoma. CRITICAL RELEVANCE STATEMENT: BH PET acquisition is a practical way to minimize motion artifacts in PET which has the potential to improve lesion detection for stage IA pulmonary adenocarcinoma.

Image-based features in machine learning to identify delivery errors and predict error magnitude for patient-specific IMRT quality assurance.

OBJECTIVE: To identify delivery error type and predict associated error magnitude by image-based features using machine learning (ML). METHODS: In this study, a total of 40 thoracic plans (including 208 beams) were selected, and four error types with different magnitudes were introduced into the original plans, including 1) collimator misalignment (COLL), 2) monitor unit (MU) variation, 3) systematic multileaf collimator misalignment (MLCS), and 4) random MLC misalignment (MLCR). These dose distributions of portal dose predictions for the original plans were defined as the reference dose distributions (RDD), while those for the error-introduced plans were defined as the error-introduced dose distributions (EDD). Both distributions were calculated for all beams with portal dose image prediction (PDIP). Besides, 14 image-based features were extracted from RDD and EDD of portal dose predictions to obtain the feature vectors. In addition, a random forest was adopted for the multiclass classification task, and regression prediction for error magnitude. RESULTS: The top five features extracted with the highest weight included 1) the relative displacement in the x direction, 2) the ratio of the absolute minimum residual error to the maximal RDD value, 3) the product of the maximum and minimum residuals, 4) the ratio of the absolute maximum residual error to the maximal RDD value, and 5) the ratio of the absolute mean residual value to the maximal RDD value. The relative displacement in the x direction had the highest weight. The overall accuracy of the five-class classification model was 99.85% for the validation set and 99.30% for the testing set. This model could be applied to the classification of the error-free plan, COLL, MU, MLCS, and MLCR with an accuracy of 100%, 98.4%, 99.9%, 98.0%, and 98.3%, respectively. MLCR had the worst performance in error magnitude prediction (70.1-96.6%), while others had better performance in error magnitude prediction (higher than 93%). In the error magnitude prediction, the mean absolute error (MAE) between predicted error magnitude and actual error ranged from 0.03 to 0.33, with the root mean squared error (RMSE) varying from 0.17 to 0.56 for the validation set. The MAE and RMSE ranged from 0.03 to 0.50 and 0.44 to 0.59 for the test set, respectively. CONCLUSION: It could be demonstrated in this study that the image-based features extracted from RDD and EDD can be employed to identify different types of delivery errors and accurately predict error magnitude with the assistance of ML techniques. They can be used to associate traditional gamma analysis with clinically based analysis for error classification and magnitude prediction in patient-specific IMRT quality assurance.

Improving CBCT image quality to the CT level using RegGAN in esophageal cancer adaptive radiotherapy.

OBJECTIVE: This study aimed to improve the image quality and CT Hounsfield unit accuracy of daily cone-beam computed tomography (CBCT) using registration generative adversarial networks (RegGAN) and apply synthetic CT (sCT) images to dose calculations in radiotherapy. METHODS: The CBCT/planning CT images of 150 esophageal cancer patients undergoing radiotherapy were used for training (120 patients) and testing (30 patients). An unsupervised deep-learning method, the 2.5D RegGAN model with an adaptively trained registration network, was proposed, through which sCT images were generated. The quality of deep-learning-generated sCT images was quantitatively compared to the reference deformed CT (dCT) image using mean absolute error (MAE), root mean square error (RMSE) of Hounsfield units (HU), and peak signal-to-noise ratio (PSNR). The dose calculation accuracy was further evaluated for esophageal cancer radiotherapy plans, and the same plans were calculated on dCT, CBCT, and sCT images. RESULTS: The quality of sCT images produced by RegGAN was significantly improved compared to the original CBCT images. ReGAN achieved image quality in the testing patients with MAE sCT vs. CBCT: 43.7 ± 4.8 vs. 80.1 ± 9.1; RMSE sCT vs. CBCT: 67.2 ± 12.4 vs. 124.2 ± 21.8; and PSNR sCT vs. CBCT: 27.9 ± 5.6 vs. 21.3 ± 4.2. The sCT images generated by the RegGAN model showed superior accuracy on dose calculation, with higher gamma passing rates (93.3 ± 4.4, 90.4 ± 5.2, and 84.3 ± 6.6) compared to original CBCT images (89.6 ± 5.7, 85.7 ± 6.9, and 72.5 ± 12.5) under the criteria of 3 mm/3%, 2 mm/2%, and 1 mm/1%, respectively. CONCLUSION: The proposed deep-learning RegGAN model seems promising for generation of high-quality sCT images from stand-alone thoracic CBCT images in an efficient way and thus has the potential to support CBCT-based esophageal cancer adaptive radiotherapy.

Total-body low-dose CT image denoising using a prior knowledge transfer technique with a contrastive regularization mechanism.

PURPOSE: Reducing the radiation exposure experienced by patients in total-body computed tomography (CT) imaging has attracted extensive attention in the medical imaging community. A low radiation dose may result in increased noise and artifacts that greatly affect the subsequent clinical diagnosis. To obtain high-quality total-body low-dose CT (LDCT) images, previous deep learning-based research works developed various network architectures. However, most of these methods only employ normal-dose CT (NDCT) images as ground truths to guide the training process of the constructed denoising network. As a result of this simple restriction, the reconstructed images tend to lose favorable image details and easily generate oversmoothed textures. This study explores how to better utilize the information contained in the feature spaces of NDCT images to guide the LDCT image reconstruction process and achieve high-quality results. METHODS: We propose a novel intratask knowledge transfer (KT) method that leverages the knowledge distilled from NDCT images as an auxiliary component of the LDCT image reconstruction process. Our proposed architecture is named the teacher-student consistency network (TSC-Net), which consists of teacher and student networks with identical architectures. By employing the designed KT loss, the student network is encouraged to emulate the teacher network in the representation space and gain robust prior content. In addition, to further exploit the information contained in CT scans, a contrastive regularization mechanism (CRM) built upon contrastive learning is introduced. The CRM aims to minimize and maximize the L2 distances from the predicted CT images to the NDCT samples and to the LDCT samples in the latent space, respectively. Moreover, based on attention and the deformable convolution approach, we design a dynamic enhancement module (DEM) to improve the network capability to transform input information flows. RESULTS: By conducting ablation studies, we prove the effectiveness of the proposed KT loss, CRM, and DEM. Extensive experimental results demonstrate that the TSC-Net outperforms the state-of-the-art methods in both quantitative and qualitative evaluations. Additionally, the excellent results obtained for clinical readings also prove that our proposed method can reconstruct high-quality CT images for clinical applications. CONCLUSIONS: Based on the experimental results and clinical readings, the TSC-Net has better performance than other approaches. In our future work, we may explore the reconstruction of LDCT images by fusing the positron emission tomography (PET) and CT modalities to further improve the visual quality of the reconstructed CT images.

A critical review of on-site inorganic arsenic screening methods.

Approximately 94 to 220 million people worldwide are at risk of drinking well water containing arsenic > 10 µg/L, the WHO guideline value. To identify non-compliant domestic wells, assess health risks and reduce exposure, accurate and rapid on-site inorganic arsenic screening methods are desirable because all domestic wells worldwide need to be tested. Here, the principles, advantages and limitations of commonly used colorimetry, electrochemistry, and biosensing methods are critically reviewed, with the performance compared with laboratory-based benchmark methods. Most commercial kits are based on the classic Gutzeit reaction. Despite being semi-quantitative, the more recent and more expensive products display improved and acceptable accuracy and shorter testing time (∼10 min). Carried out by trained professionals, electrochemical methods are also feasible for on-site analysis, although miniaturization is desirable yet challenging. Biosensing using whole bacterial cells or bio-engineered materials such as aptamers is promising, if incorporated with function specific nanomaterials and biomaterials. Since arsenic is frequently found as arsenite in reducing groundwater and subject to oxidation during sampling, transportation and storage, on-site separation and sample preservation are feasible but the specific methods should be chosen based on sample matrix and tested before use. To eliminate arsenic exposure among hundreds of millions of mostly rural residents worldwide, we call for concerted efforts in research community and regulatory authority to develop accurate, rapid, and affordable tests for on-site screening and monitoring of arsenic in drinking water. Access to affordable testing will benefit people who are socioeconomically disadvantaged.