CNN-based multi-modal radiomics analysis of pseudo-CT utilization in MRI-only brain stereotactic radiotherapy: a feasibility study.

BACKGROUND: Pseudo-computed tomography (pCT) quality is a crucial issue in magnetic resonance image (MRI)-only brain stereotactic radiotherapy (SRT), so this study systematically evaluated it from the multi-modal radiomics perspective. METHODS: 34 cases (< 30 cm³) were retrospectively included (2021.9-2022.10). For each case, both CT and MRI scans were performed at simulation, and pCT was generated by a convolutional neural network (CNN) from planning MRI. Conformal arc or volumetric modulated arc technique was used to optimize the dose distribution. The SRT dose was compared between pCT and planning CT with dose volume histogram (DVH) metrics and gamma index. Wilcoxon test and Spearman analysis were used to identify key factors associated with dose deviations. Additionally, original image features were extracted for radiomic analysis. Tumor control probability (TCP) and normal tissue complication probability (NTCP) were employed for efficacy evaluation. RESULTS: There was no significant difference between pCT and planning CT except for radiomics. The mean value of Hounsfield unit of the planning CT was slightly higher than that of pCT. The Gadolinium-based agents in planning MRI could increase DVH metrics deviation slightly. The median local gamma passing rates (1%/1 mm) between planning CTs and pCTs (non-contrast) was 92.6% (range 63.5-99.6%). Also, differences were observed in more than 85% of original radiomic features. The mean absolute deviation in TCP was 0.03%, and the NTCP difference was below 0.02%, except for the normal brain, which had a 0.16% difference. In addition, the number of SRT fractions and lesions, and lesion morphology could influence dose deviation. CONCLUSIONS: This is the first multi-modal radiomics analysis of CNN-based pCT from planning MRI for SRT of small brain lesions, covering dosiomics and radiomics. The findings suggest the potential of pCT in SRT plan design and efficacy prediction, but caution needs to be taken for radiomic analysis.

Dynamic Changes in Bladder Morphology Over Time in Cervical Cancer Patients.

OBJECTIVES: Continuous surveillance of bladder volume (BV) is beneficial during the treatment of various urogenital diseases because the bladder is always changing its position, size and even shape at different filling phases. For this purpose, we quantified the motion of the urinary bladder. METHODS: Daily ultrasound measurements and weekly cone-beam computed tomography scans were obtained from 89 patients in the supine position. BV, bladder centroid positions, and triaxial lengths in the left-right (LR), anterior-posterior (AP), and superior-inferior (SI) directions were compared across different time points. RESULTS: BV linearly increased over time, and the mean urinary filling rate (vtot) was correlated with the patients' age and water consumption. The greatest bladder centroid motion occurred longitudinally, with less movement observed laterally. The maximum bladder centroid movement was 18.8 ± 2.2 mm inferiorly and 1.8 ± 0.9 mm posteriorly for every 10% decrease in BV. The rates of changes in triaxial lengths differed across the 4 filling phases. The rate was the largest at a BV range of 10-80 mL, especially in the LR direction, with values of 5.9 ± 1.0, 3.6 ± 1.0, and 3.9 ± 1.0 mm per every 10-mL BV increase for LR, AP, and SI, respectively. With bladder filling (<80 mL), the maximum increase in triaxial length was observed in the SI direction and the rates of all changes considerably decreased, especially at BV > 600 mL. CONCLUSION: The vtot could be used to evaluate the temporal changes in the bladder. The spatial changes should be assessed according to different filling phases based on the centroid position and triaxial lengths.

[Comparative Study on Evaluating the Bladder Volume between BladderScan BVI9400 and Ultrasound System iU22].

OBJECTIVE: To evaluate the accuracy of the latest BladderScan BVI9400 on measuring bladder volume. METHODS: Two bladder phantoms were selected for investigating the accuracy of BVI9400. 341 patients with the iU22 ultrasound examinations were followed by BVI 9400. The difference and correlation between BVI9400 and iU22 were contrastively analyzed. RESULTS: The relative difference between results from BVI9400 and phantom volume was 2.5% and 1.36%. There was a strong correlation for patients between BVI9400 and iU22 (R = 0.96, P < 0.001). The relative difference between BVI9400 and iU22 decreased with the increasing of bladder volume and had no significant difference with patient's gender (P > 0.1). CONCLUSION: BladderScan BVI9400 had the ability of high accuracy and good stability of measured data. In view of quick and conveniences, BVI9400 could be as auxiliary equipment on pelvic tumor to evaluate whether the bladder volume during fractional radiotherapy was consistency with that during CT positioning.

The impact of plan complexity on calculation and measurement-based pre-treatment verifications for sliding-window intensity-modulated radiotherapy.

Background and purpose: In sliding-window intensity-modulated radiotherapy, increased plan modulation often leads to increased plan complexities and dose uncertainties. Dose calculation and/or measurement checks are usually adopted for pre-treatment verification. This study aims to evaluate the relationship among plan complexities, calculated doses and measured doses. Materials and methods: A total of 53 plan complexity metrics (PCMs) were selected, emphasizing small field characteristics and leaf speed/acceleration. Doses were retrieved from two beam-matched treatment devices. The intended dose was computed employing the Anisotropic Analytical Algorithm and validated through Monte Carlo (MC) and Collapsed Cone Convolution (CCC) algorithms. To measure the delivered dose, 3D diode arrays of various geometries, encompassing helical, cross, and oblique cross shapes, were utilized. Their interrelation was assessed via Spearman correlation analysis and principal component linear regression (PCR). Results: The correlation coefficients between calculation-based (CQA) and measurement-based verification quality assurance (MQA) were below 0.53. Most PCMs showed higher correlation rpcm-QA with CQA (max: 0.84) than MQA (max: 0.65). The proportion of rpcm-QA  ≥ 0.5 was the largest in the pelvis compared to head-and-neck and chest-and-abdomen, and the highest rpcm-QA occurred at 1 %/1mm. Some modulation indices for the MLC speed and acceleration were significantly correlated with CQA and MQA. PCR's determination coefficients (R2 ) indicated PCMs had higher accuracy in predicting CQA (max: 0.75) than MQA (max: 0.42). Conclusions: CQA and MQA demonstrated a weak correlation. Compared to MQA, CQA exhibited a stronger correlation with PCMs. Certain PCMs related to MLC movement effectively indicated variations in both quality assurances.

The role of image-guided radiotherapy in prostate cancer: A systematic review and meta-analysis.

Background: Image-guided radiotherapy (IGRT) has gradually been widely promoted in clinical procedure. However, there has been no consensus on the effects of IGRT on toxicity and survival, and no clear level 1 evidence has even been promulgated. Methods: Medline, EMBASE, PubMed, Cochrane databases and ClinicalTrials.gov were searched for studies comparing IGRT vs non-IGRT or higher frequency IGRT vs lower frequency IGRT during prostate radiotherapy, indexed from database inception to April 2022. Results: The review included 18 studies (3 randomized clinical trial and 15 cohort studies) involving 6521 men, with a median duration of patient follow-up of 46.2 months in the IGRT group vs 52.7 months in the control group. The meta-analysis demonstrated that IGRT significantly reduced acute GU (risk ratio [RR], 0.78; 95 % confidence interval [CI], 0.69-0.88; P < 0.001 [9 studies]) and GI toxicity (RR, 0.49; 95 % CI, 0.35-0.68; P < 0.001 [4 studies]) and late GI toxicity (HR, 0.25; 95 % CI, 0.07-0.87; P = 0.03 [3 studies]) compared with non-IGRT. Meanwhile, compared with prospective studies, retrospective studies showed that IGRT had a more significant effect in reducing the late GI toxicity. Compared with non-daily IGRT, daily IGRT significantly improved 3-year PRFS (HR, 0.45; 95 % CI, 0.28-0.72; P = 0.001 [2 studies]) and BFFS (HR, 0.57; 95 % CI, 0.39-0.83; P = 0.003 [3 studies]). Furthermore, high-frequency daily IGRT could lead to greater 3-year BFFS benefit in prostate cancer patients than weekly IGRT. However, no significant effects of IGRT on acute rectal toxicity, late GU toxicity, 5-year OS and SCM were found. Conclusions: For men receiving prostate radiotherapy, IGRT was associated with an improvement in biochemical tumor control and a reduction in GI and acute GU toxicity, but did not significantly improve 5-year OS or increase 5-year SCM.

Pretreatment patient-specific quality assurance prediction based on 1D complexity metrics and 3D planning dose: classification, gamma passing rates, and DVH metrics.

PURPOSE: Highly modulated radiotherapy plans aim to achieve target conformality and spare organs at risk, but the high complexity of the plan may increase the uncertainty of treatment. Thus, patient-specific quality assurance (PSQA) plays a crucial role in ensuring treatment accuracy and providing clinical guidance. This study aims to propose a prediction model based on complexity metrics and patient planning dose for PSQA results. MATERIALS AND METHODS: Planning dose, measurement-based reconstructed dose and plan complexity metrics of the 687 radiotherapy plans of patients treated in our institution were collected for model establishing. Global gamma passing rate (GPR, 3%/2mm,10% threshold) of 90% was used as QA criterion. Neural architecture models based on Swin-transformer were adapted to process 3D dose and incorporate 1D metrics to predict QA results. The dataset was divided into training (447), validation (90), and testing (150) sets. Evaluation of predictions was performed using mean absolute error (MAE) for GPR, planning target volume (PTV) HI and PTV CI, mean absolute percentage error (MAPE) for PTV D95, PTV D2 and PTV Dmean, and the area under the receiver operating characteristic (ROC) curve (AUC) for classification. Furthermore, we also compare the prediction results with other models based on either only 1D or 3D inputs. RESULTS: In this dataset, 72.8% (500/687) plans passed the pretreatment QA under the criterion. On the testing set, our model achieves the highest performance, with the 1D model slightly surpassing the 3D model. The performance results are as follows (combine, 1D, and 3D transformer): The AUCs are 0.92, 0.88 and 0.86 for QA classification. The MAEs of prediction are 0.039, 0.046, and 0.040 for 3D GPR, 0.018, 0.021, and 0.019 for PTV HI, and 0.075, 0.078, and 0.084 for PTV CI. Specifically, for cases with 3D GPRs greater than 90%, the MAE could achieve 0.020 (combine). The MAPE of prediction is 1.23%, 1.52%, and 1.66% for PTV D95, 2.36%, 2.67%, and 2.45% for PTV D2, and 1.46%, 1.70%, and 1.71% for PTV Dmean. CONCLUSION: The model based on 1D complexity metrics and 3D planning dose could predict pretreatment PSQA results with high accuracy and the complexity metrics play a leading role in the model. Furthermore, dose-volume metric deviations of PTV could be predicted and more clinically valuable information could be provided.

Optimal fractionation and timing of weekly cone-beam CT in daily surface-guided radiotherapy for breast cancer.

PURPOSE: Surface-guided radiotherapy (SGRT) has been demonstrated to be a promising supplement to cone-beam computed tomography (CBCT) in adjuvant breast cancer radiotherapy, but a rational combination mode is lacking in clinical practice. The aim of this study was to explore this mode and investigate its impact on the setup and dose accuracy. METHODS AND MATERIALS: Daily SGRT and weekly CBCT images were acquired for 23 patients with breast cancer who received conventional fractionated radiotherapy after lumpectomy. Sixteen modes were acquired by randomly selecting one (CBCT1), two (CBCTij), three (CBCTijk), four (CBCTijkl), and five (CBCT12345) images from the CBCT images for fusion with the SGRT. The CTV-PTV margins, OAR doses, and dose coverage (V95%) of PTV and CTV was calculated based on SGRT setup errors with different regions of interest (ROIs). Dose correlations between these modalities were investigated using Pearson and Spearman's methods. Patient-specific parameters were recorded to assess their impact on dose. RESULTS: The CTV-PTV margins decreased with increasing CBCT frequencies and were close to 5 mm for CBCTijkl and CBCT12345. For the ipsilateral breast ROI, SGRT errors were larger in the AP direction, and target doses were higher in all modes than in the whole breast ROI (P < 0.05). In the ipsilateral ROI, the target dose correlations between all modes increased with increasing CBCT time intervals, decreased, and then increased with increasing CBCT frequencies, with the inflection point being CBCT participation at week 5. The dose deviations in CBCT123, CBCT124, CBCT125, CBCTijkl, and CBCT12345 were minimal and did not differ significantly (P > 0.05). There was excellent agreement between CBCT124 and CBCT1234, and between (CBCTijkl, CBCT12345) and CBCT125 in determining the classification for the percentage of PTV deviation (Kappa = 0.704-0.901). In addition, there were weak correlations between the patient's Dips_b (ipsilateral breast diameter with bolus) and CTV doses in modes with CBCT participation at week 4 (R = 0.270 to 0.480). CONCLUSIONS: Based on weekly CBCT, these modes with ipsilateral ROI and a combination of daily SGRT and a CBCT frequency of ≥ 3 were recommended, and CBCT was required at weeks 1 and 2 for CBCTijk.

A novel dose fall-off index and preliminary application in brain and lung stereotactic radiotherapy.

BACKGROUND: Stereotactic radiotherapy (SRT) has been widely used for the treatment of brain metastases and early stage non-small-cell lung cancer (NSCLC). Excellent SRT plans are characterized by steep dose fall-off, making it critical to accurately and comprehensively predict and evaluate dose fall-off. PURPOSE: A novel dose fall-off index was proposed to ensure high-quality SRT planning. METHODS: The novel gradient index (NGI) had two different modes: NGIx V for three-dimensions and NGIx r for one-dimension. NGIx V and NGIx r were defined as the ratios of the decreased percentage dose (x%) to the corresponding isodose volume and equivalent sphere radii, respectively. A total of 243 SRT plans at our institution between April 2020 and March 2022 were enrolled, including 126 brain and 117 lung SRT plans. Measurement-based verifications were performed using SRS MapCHECK. Ten plan complexity indexes were calculated. Dosimetric parameters related to radiation injuries were also extracted, including the normal brain volume exposed to 12 Gy (V12 ) and 18 Gy (V18 ) during single-fraction SRT (SF-SRT) and multi-fraction SRT (MF-SRT), respectively, and the normal lung volume exposed to 12 Gy (V12 ). The performance of NGI and other common dose fall-off indexes, gradient index (GI), R50% and D2cm were evaluated using Spearman correlation analysis to explore their correlations with the PTV size, gamma passing rate (GPR), plan complexity indexes, and dosimetric parameters. RESULTS: There were statistically significant correlations between NGI and PTV size (r = -0.98, P < 0.01 for NGI50 V and r = -0.93, P < 0.01 for NGI50 r), which were the strongest correlations compared with GI (r = 0.11, P = 0.13), R50% (r = -0.08, P = 0.19) and D2cm (r = 0.84, P < 0.01). The fitted formulas of NGI50 V = 23.86V-1.00 and NGI50 r = 113.5r-1.05 were established. The GPRs of enrolled SRT plans were 98.6 ± 1.7%, 94.2 ± 4.7% and 97.1 ± 3.1% using the criteria of 3%/2 mm, 3%/1 mm, and 2%/2 mm, respectively. NGI50 V achieved the strongest correlations with various plan complexity indexes (|r| ranged from 0.67 to 0.91, P < 0.01). NGI50 V also showed the highest r values with V12 (r = -0.93, P < 0.01) and V18 (r = -0.96, P < 0.01) of the normal brain during SF-SRT and MF-SRT, respectively, and V12 (r = -0.86, P < 0.01) of the normal lung during lung SRT. CONCLUSIONS: Compared with GI, R50% and D2cm , the proposed dose fall-off index, NGI, had the strongest correlations with the PTV size, plan complexity and V12 /V18 of the normal tissues. These correlations established on NGI are more helpful and reliable for SRT planning, quality control, and reducing the risk of radiation injuries.

Pseudo-siamese network combined with dosimetric and clinical factors, radiomics features, CT images and 3D dose distribution for the prediction of radiation pneumonitis: A feasibility study.

Purpose: Radiation pneumonitis (RP)(grade ≥ 2) can have a considerable impact on patient quality-of life. In previous studies, the traditional method commonly used radiomics and clinical factors for RP prediction. This study aims to develop and evaluate a novel pseudo-siamese network (PSN) to assist radiologists predict RP before radiotherapy based on combination of dosimetric and clinical factors, radiomics features, CT (computed tomography) images, and dose distribution (hybrid model). Method: One hundred and ten patients with lung cancer (19 RP ≥ 2) who received radiotherapy between 2016 and 2020 were retrospectively enrolled in this study. Dosimetric factors were calculated from DVH (dose-volume histogram), such as lung mean dose, lung V5, and prescription dose. Clinical characteristics were recorded, such as age, sex, smoking status, TN stage, and overall stage. A total of 1419 radiomics features were extracted. Cluster analysis was used for detecting radiomics features that associated with RP. Patients were randomly split into a training set (90 %, 85 non-RP, and 14 RP) and a validation set (10 %, 6 non-RP, and 5 RP). A PSN architecture was designed for combining 1D (dosimetric and clinical factors, radiomics) and 3D (CT images, 3D dose distribution) features. 5-fold cross-validation procedure for estimating the skill of the model on new data. Results: For cluster analysis, totally of 106 radiomics features with high correlation were selected. The accuracy was 0.727, 0.636, 0.545, and 0.727 for input dosimetric and clinical factors, dose distribution, CT images, and radiomics features, respectively. The accuracy of hybrid model was 0.818. The sensitivity of hybrid model was 0.800 (95 % confidence interval (CI) [0.299, 0.989]), and specificity was 0.833(95 % CI [0.364, 0.991]). The areas under the receiver operating characteristic curves (AUCs) result in 5-fold cross-validation was 0.77-0.90(mean AUC ± std was 0.85 ± 0.05). Conclusion: This study firstly propose method that the combination of high dimensional and low dimensional features for RP prediction. The results confirm the feasibility of multi-dimensional features predict RP.

Clinical Application Effect of Cluster Management in Noninvasive Ventilator Nursing Care of Patients with Severe Heart Failure.

Objective: To elucidate the clinical application effect of cluster management in noninvasive ventilator nursing care of patients with severe heart failure (HF). Methods: A total of 116 severe patients with HF who received treatment in the our hospital between October 2018 and December 2019 were included, including 50 cases (control group) treated with routine nursing and 66 cases (research group) treated with cluster management. The treatment-related indexes (mechanical ventilation time and hospitalization expenses), symptom resolution (dyspnea, insomnia, nausea, and upper abdominal pain), systolic/diastolic blood pressure (SBP/DBP), heart rate (HR), and prognosis (mortality and disability rate) were observed and compared between the two groups. Results: Statistically shorter time of mechanical ventilation and symptom (dyspnea, insomnia, nausea, and upper abdominal pain) resolution were found in the research group compared with the control group. In addition, the research group showed significantly lower hospitalization expenses, SBP, DBP, and HR than the control group. Moreover, lower mortality and disability rates were determined in the research group, yet with no statistical significance between the two cohorts. Conclusion: The above results indicate the remarkable clinical application effect of cluster management in noninvasive ventilator nursing of severe HF, which can enhance the treatment efficacy, blood pressure and HR of patients, and facilitate their recovery.