Background: The dynamic tracking of tumors with radiation beams in radiation therapy requires the prediction of real-time target locations prior to beam delivery, as treatment involving radiation beams and gating tracking results in time latency. Objective: In this study, a deep learning model that was based on a temporal convolutional neural network was developed to predict internal target locations by using multiple external markers. Methods: Respiratory signals from 69 treatment fractions of 21 patients with cancer who were treated with the CyberKnife Synchrony device (Accuray Incorporated) were used to train and test the model. The reported model’s performance was evaluated by comparing the model to a long short-term memory model in terms of the root mean square errors (RMSEs) of real and predicted respiratory signals. The effect of the number of external markers was also investigated. Results: The average RMSEs of predicted (ahead time=400 ms) respiratory motion in the superior-inferior, anterior-posterior, and left-right directions and in 3D space were 0.49 mm, 0.28 mm, 0.25 mm, and 0.67 mm, respectively. Conclusions: The experiment results demonstrated that the temporal convolutional neural network–based respiratory prediction model could predict respiratory signals with submillimeter accuracy.

The objective of this study was to evaluate the discriminative capabilities of radiomics signatures derived from three distinct machine learning algorithms and to identify a robust radiomics signature capable of predicting pathological complete response (pCR) after neoadjuvant chemoradiotherapy in patients diagnosed with locally advanced rectal cancer (LARC). In a retrospective study, 211 LARC patients were consecutively enrolled and divided into a training cohort (n = 148) and a validation cohort (n = 63). From pretreatment contrast-enhanced planning CT images, a total of 851 radiomics features were extracted. Feature selection and radiomics score (Radscore) construction were performed using three different machine learning methods: least absolute shrinkage and selection operator (LASSO), random forest (RF) and support vector machine (SVM). The SVM-derived Radscore demonstrated a strong correlation with the pCR status, yielding area under the receiver operating characteristic curves (AUCs) of 0.880 and 0.830 in the training and validation cohorts, respectively, outperforming the RF and LASSO methods. Based on this, a nomogram was developed by combining the SVM-based Radscore with clinical indicators to predict pCR after neoadjuvant chemoradiotherapy. The nomogram exhibited superior predictive power, achieving AUCs of 0.910 and 0.866 in the training and validation cohorts, respectively. Calibration curves and decision curve analyses confirmed its appropriateness. The SVM-based Radscore demonstrated promising performance in predicting pCR for LARC patients. The machine learning-driven nomogram, which integrates the Radscore and clinical indicators, represents a valuable tool for predicting pCR in LARC patients.

To predict real-time 3D deformation field maps (DFMs) using Volumetric Cine MRI (VC-MRI) and adaptive boosting and multi-layer perceptron neural network (ADMLP-NN) for 4D target tracking. One phase of a prior 4D-MRI is set as the prior phase, MRI _prior . Principal component analysis (PCA) is used to extract three major respiratory deformation modes from the DFMs generated between the prior and remaining phases. VC-MRI at each time-step is considered a deformation of MRI _prior , where the DFM is represented as a weighted linear combination of the PCA components. The PCA weightings are solved by minimizing the differences between on-board 2D cine MRI and its corresponding VC-MRI slice. The PCA weightings solved during the initial training period are used to train an ADMLP-NN to predict PCA weightings ahead of time during the prediction period. The predicted PCA weightings are used to build predicted 3D DFM and ultimately, predicted VC-MRIs for 4D target tracking. The method was evaluated using a 4D computerized phantom (XCAT) with patient breathing curves and MRI data from a real liver cancer patient. Effects of breathing amplitude change and ADMLP-NN parameter variations were assessed. The accuracy of the PCA curve prediction was evaluated. The predicted real-time 3D tumor was evaluated against the ground-truth using volume dice coefficient (VDC), center-of-mass-shift (COMS), and target tracking errors. For the XCAT study, the average VDC and COMS for the predicted tumor were 0.92 ± 0.02 and 1.06 ± 0.40 mm, respectively, across all predicted time-steps. The correlation coefficients between predicted and actual PCA curves generated through VC-MRI estimation for the 1st/2nd principal components were 0.98/0.89 and 0.99/0.57 in the SI and AP directions, respectively. The optimal number of input neurons, hidden neurons, and MLP-NN for ADMLP-NN PCA weighting coefficient prediction were determined to be 7, 4, and 10, respectively. The optimal cost function threshold was determined to be 0.05. PCA weighting coefficient and VC-MRI accuracy was reduced for increased prediction-step size. Accurate PCA weighting coefficient prediction correlated with accurate VC-MRI prediction. For the patient study, the predicted 4D tumor tracking errors in superior–inferior, anterior–posterior and lateral directions were 0.50 ± 0.47 mm, 0.40 ± 0.55 mm, and 0.28 ± 0.12 mm, respectively. Preliminary studies demonstrated the feasibility to use VC-MRI and artificial neural networks to predict real-time 3D DFMs of the tumor for 4D target tracking.

Purpose: To enhance the accuracy of real-time four-dimensional cone beam CT (4D-CBCT) imaging by incorporating spatiotemporal correlation from the sequential projection image into the single projection-based 4D-CBCT estimation process.Methods: We first derived 4D deformation vector fields (DVFs) from patient 4D-CT. Principal component analysis (PCA) was then employed to extract distinctive feature labels for each DVF, focusing on the first three PCA coefficients. To simulate a wide range of respiratory motion, we expanded the motion amplitude and used random sampling to generate approximately 900 sets of PCA labels. These labels were used to produce 900 simulated 4D-DVFs, which in turn deformed the 0% phase 4D-CT to obtain 900 CBCT volumes with continuous motion amplitudes. Following this, the forward projection was performed at one angle to get all of the digital reconstructed radiographs (DRRs). These DRRs and the PCA labels were used as the training data set. To capture the spatiotemporal correlation in the projections, we propose to use the convolutional LSTM (ConvLSTM) network for PCA coefficient estimation. For network testing, when several online CBCT projections (with different motion amplitudes that cover the full respiration range) are acquired and sent into the network, the corresponding 4D-PCA coefficients will be obtained and finally lead to a full online 4D-CBCT prediction. A phantom experiment is first performed with the XCAT phantom; then, a pilot clinical evaluation is further conducted.Results: Results on the XCAT phantom and the patient data show that the proposed approach outperformed other networks in terms of visual inspection and quantitative metrics. For the XCAT phantom experiment, ConvLSTM achieves the highest quantification accuracy with MAPE(Mean Absolute Percentage Error), PSNR (Peak Signal-to-Noise Ratio), and RMSE(Root Mean Squared Error) of 0.0459, 64.6742, and 0.0011, respectively. For the patient pilot clinical experiment, ConvLSTM also achieves the best quantification accuracy with that of 0.0934, 63.7294, and 0.0019, respectively. The quantification evaluation labels that we used are 1) the Mean Absolute Error (MAE), 2) the Normalized Cross Correlation (NCC), 3)the Structural Similarity Index Measurement(SSIM), 4)the Peak Signal-to-Noise Ratio (PSNR), 5)the Root Mean Squared Error(RMSE), and 6) the Absolute Percentage Error (MAPE).Conclusion: The spatiotemporal correlation-based respiration motion modeling supplied a potential solution for accurate real-time 4D-CBCT reconstruction.

BACKGROUND The dynamic tracking of tumors with radiation beams in radiation therapy requires the prediction of real-time target locations prior to beam delivery, as treatment involving radiation beams and gating tracking results in time latency. OBJECTIVE In this study, a deep learning model that was based on a temporal convolutional neural network was developed to predict internal target locations by using multiple external markers. METHODS Respiratory signals from 69 treatment fractions of 21 patients with cancer who were treated with the CyberKnife Synchrony device (Accuray Incorporated) were used to train and test the model. The reported model’s performance was evaluated by comparing the model to a long short-term memory model in terms of the root mean square errors (RMSEs) of real and predicted respiratory signals. The effect of the number of external markers was also investigated. RESULTS The average RMSEs of predicted (ahead time=400 ms) respiratory motion in the superior-inferior, anterior-posterior, and left-right directions and in 3D space were 0.49 mm, 0.28 mm, 0.25 mm, and 0.67 mm, respectively. CONCLUSIONS The experiment results demonstrated that the temporal convolutional neural network–based respiratory prediction model could predict respiratory signals with submillimeter accuracy.

Personal protective equipment (PPE) guidelines serve to protect healthcare providers and patients from harmful biohazards. With the rise of the 2019 SARS-CoV-2 disease (COVID-19), many institutions have mandated strictly enforced endoscopic PPE guidelines. We currently do not know how current practitioners perceive these mandates or how they will influence their practice long-term. We aimed to survey the PPE practices among endoscopists across Canada and compare their perceived differences in practice between the pre- and post-COVID-19 pandemic eras. We hypothesize that the PPE guidelines during the pandemic will influence changes in PPE practices in endoscopy. A 74-item questionnaire was emailed from June 2020 to September 2020 to all members of the Canadian Association of Gastroenterologists and the Canadian Association of General Surgeons through newsletters. The survey was created by expert consensus and distributed using the REDCap software. Survey questions collected basic demographics and differences between PPE practices pre- and post-COVID-19 pandemic eras. PPE practices were categorized into four endoscopic procedure types: diagnostic or therapeutic, and upper or lower gastrointestinal endoscopy. Individual outcomes were reported as rates, or ranges when evaluating for all procedure types. A total of 77 respondents completed the survey with the majority of respondents aged 40–49 (34 [44.2%]) and identifying as Gastroenterologists (54 [70.1%]). Gender was evenly split (38 females [49.4%] versus 39 males [50.6%]). In the pre-pandemic era, the majority of endoscopists wore gowns (91.0–93.9%) and all endoscopists wore gloves (100%). However, the majority of endoscopists did not wear surgical masks (20.9%-31.3%), N95 respirators (1.5%-3.2%), face shields (13.4%-33.9%), eye protection (13.4%-21.3%), or hair protection (11.1%-12.5%). In the post-pandemic era, endoscopists reported a plan to dramatically change their pre-pandemic practices and adopt current PPE mandates. All endoscopists reported a plan to fully gown and glove (100%) with the majority reporting they will continue wearing surgical masks (87.7%-90.5%), face shields (57.8%-75.0%), and hair protection (50.8%-53.8%). However, the majority reported a plan to decrease universal use of N95 respirators (6.5%-23.7%) or eye protection (36.5%-40.0%). Over half of the respondents reported changing their practice from no masking pre-pandemic to implementing routine masking. The COVID-19 pandemic has changed the attitudes of many endoscopists regarding future PPE use in routine endoscopy. Ongoing studies comparing the rates of transmission of hospital-acquired infections in the setting of endoscopy are needed to develop a new post-pandemic PPE consensus. None