A feasibility study to predict 3D dose delivery accuracy for IMRT using DenseNet with log files.

OBJECTIVE: This study aims to explore the feasibility of DenseNet in the establishment of a three-dimensional (3D) gamma prediction model of IMRT based on the actual parameters recorded in the log files during delivery. METHODS: A total of 55 IMRT plans (including 367 fields) were randomly selected. The gamma analysis was performed using gamma criteria of 3% /3 mm (Dose Difference/Distance to Agreement), 3% /2 mm, 2% /3 mm, and 2% /2 mm with a 10% dose threshold. In addition, the log files that recorded the gantry angle, monitor units (MU), multi-leaf collimator (MLC), and jaws position during delivery were collected. These log files were then converted to MU-weighted fluence maps as the input of DenseNet, gamma passing rates (GPRs) under four different gamma criteria as the output, and mean square errors (MSEs) as the loss function of this model. RESULTS: Under different gamma criteria, the accuracy of a 3D GPR prediction model decreased with the implementation of stricter gamma criteria. In the test set, the mean absolute error (MAE) of the prediction model under the gamma criteria of 3% /3 mm, 2% /3 mm, 3% /2 mm, and 2% /2 mm was 1.41, 1.44, 3.29, and 3.54, respectively; the root mean square error (RMSE) was 1.91, 1.85, 4.27, and 4.40, respectively; the Sr was 0.487, 0.554, 0.573, and 0.506, respectively. There was a correlation between predicted and measured GPRs (P <  0.01). Additionally, there was no significant difference in the accuracy between the validation set and the test set. The accuracy in the high GPR group was high, and the MAE in the high GPR group was smaller than that in the low GPR group under four different gamma criteria. CONCLUSIONS: In this study, a 3D GPR prediction model of patient-specific QA using DenseNet was established based on log files. As an auxiliary tool for 3D dose verification in IMRT, this model is expected to improve the accuracy and efficiency of dose validation.

An independent Monte Carlo-based IMRT QA tool for a 0.35 T MRI-guided linear accelerator.

PURPOSE: To develop an independent log file-based intensity-modulated radiation therapy (IMRT) quality assurance (QA) tool for the 0.35 T magnetic resonance-linac (MR-linac) and investigate the ability of various IMRT plan complexity metrics to predict the QA results. Complexity metrics related to tissue heterogeneity were also introduced. METHODS: The tool for particle simulation (TOPAS) Monte Carlo code was utilized with a previously validated linac head model. A cohort of 29 treatment plans was selected for IMRT QA using the developed QA tool and the vendor-supplied adaptive QA (AQA) tool. For 27 independent patient cases, various IMRT plan complexity metrics were calculated to assess the deliverability of these plans. A correlation between the gamma pass rates (GPRs) from the AQA results and calculated IMRT complexity metrics was determined using the Pearson correlation coefficients. Tissue heterogeneity complexity metrics were calculated based on the gradient of the Hounsfield units. RESULTS: The median and interquartile range for the TOPAS GPRs (3%/3 mm criteria) were 97.24% and 3.75%, respectively, and were 99.54% and 0.36% for the AQA tool, respectively. The computational time for TOPAS ranged from 4 to 8 h to achieve a statistical uncertainty of <1.5%, whereas the AQA tool had an average calculation time of a few minutes. Of the 23 calculated IMRT plan complexity metrics, the AQA GPRs had correlations with 7 out of 23 of the calculated metrics. Strong correlations (|r| > 0.7) were found between the GPRs and the heterogeneity complexity metrics introduced in this work. CONCLUSIONS: An independent MC and log file-based IMRT QA tool was successfully developed and can be clinically deployed for offline QA. The complexity metrics will supplement QA reports and provide information regarding plan complexity.

Insensitivity of machine log files to MLC leaf backlash and effect of MLC backlash on clinical dynamic MLC motion: An experimental investigation.

PURPOSE: Multi-leaf-collimator (MLC) leaf position accuracy is important for accurate dynamic radiotherapy treatment plan delivery. Machine log files have become widely utilized for quality assurance (QA) of such dynamic treatments. The primary aim is to test the sensitivity of machine log files in comparison to electronic portal imaging device (EPID)-based measurements to MLC position errors caused by leaf backlash. The secondary aim is to investigate the effect of MLC leaf backlash on MLC leaf motion during clinical dynamic plan delivery. METHODS: The sensitivity of machine log files and two EPID-based measurements were assessed via a controlled experiment, whereby the length of the "T" section of a series of 12 MLC leaf T-nuts in a Varian Millennium MLC for a Trilogy C-series type linac was reduced by sandpapering the top of the "T" to introduce backlash. The built-in machine MLC leaf backlash test as well as measurements for two EPID-based dynamic MLC positional tests along with log files were recorded pre- and post-T-nut modification. All methods were investigated for sensitivity to the T-nut change by assessing the effect on measured MLC leaf positions. A reduced version of the experiment was repeated on a TrueBeam type linac with Millennium MLC. RESULTS: No significant differences before and after T-nut modification were detected in any of the log file data. Both EPID methods demonstrated sensitivity to the introduced change at approximately the expected magnitude with a strong dependence observed with gantry angle. EPID-based data showed MLC positional error in agreement with the micrometer measured T-nut length change to 0.07 ± 0.05 mm (1 SD) using the departmental routine QA test. Backlash results were consistent between linac types. CONCLUSION: Machine log files appear insensitive to MLC position errors caused by MLC leaf backlash introduced via the T-nut. The effect of backlash on clinical MLC motions is heavily gantry angle dependent.

Machine learning models to predict the delivered positions of Elekta multileaf collimator leaves for volumetric modulated arc therapy.

PURPOSE: Accurate positioning of multileaf collimator (MLC) leaves during volumetric modulated arc therapy (VMAT) is essential for accurate treatment delivery. We developed a linear regression, support vector machine, random forest, extreme gradient boosting (XGBoost), and an artificial neural network (ANN) for predicting the delivered leaf positions for VMAT plans. METHODS: For this study, 160 MLC log files from 80 VMAT plans were obtained from a single institution treated on 3 Elekta Versa HD linear accelerators. The gravity vector, X1 and X2 jaw positions, leaf gap, leaf position, leaf velocity, and leaf acceleration were extracted and used as model inputs. The models were trained using 70% of the log files and tested on the remaining 30%. Mean absolute error (MAE), root mean square error (RMSE), the coefficient of determination R2 , and fitted line plots showing the relationship between delivered and predicted leaf positions were used to evaluate model performance. RESULTS: The models achieved the following errors: linear regression (MAE = 0.158 mm, RMSE = 0.225 mm), support vector machine (MAE = 0.141 mm, RMSE = 0.199 mm), random forest (MAE = 0.161 mm, RMSE = 0.229 mm), XGBoost (MAE = 0.185 mm, RMSE = 0.273 mm), and ANN (MAE = 0.361 mm, RMSE = 0.521 mm). A significant correlation between a plan's gamma passing rate (GPR) and the prediction errors of linear regression, support vector machine, and random forest is seen (p < 0.045). CONCLUSIONS: We examined various models to predict the delivered MLC positions for VMAT plans treated with Elekta linacs. Linear regression, support vector machine, random forest, and XGBoost achieved lower errors than ANN. Models that can accurately predict the individual leaf positions during treatment can help identify leaves that are deviating from the planned position, which can improve a plan's GPR.

Differential Effects of Outpatient Portal User Status on Inpatient Portal Use: Observational Study.

BACKGROUND: The decision to use patient portals can be influenced by multiple factors, including individuals' perceptions of the tool, which are based on both their personal skills and experiences. Prior experience with one type of portal may make individuals more comfortable with using newer portal technologies. Experienced outpatient portal users in particular may have confidence in their ability to use inpatient portals that have similar functionality. In practice, the use of both outpatient and inpatient portal technologies can provide patients with continuity of access to their health information across care settings, but the influence of one type of portal use on the use of other portals has not been studied. OBJECTIVE: This study aims to understand how patients' use of an inpatient portal is influenced by outpatient portal use. METHODS: This study included patients from an academic medical center who were provided access to an inpatient portal during their hospital stays between 2016 and 2018 (N=1571). We analyzed inpatient portal log files to investigate how inpatient portal use varied by using 3 categories of outpatient portal users: prior users, new users, and nonusers. RESULTS: Compared with prior users (695/1571, 44.24%) of an outpatient portal, new users (214/1571, 13.62%) had higher use of a select set of inpatient portal functions (messaging function: incidence rate ratio [IRR] 1.33, 95% CI 1.06-1.67; function that provides access to the outpatient portal through the inpatient portal: IRR 1.34, 95% CI 1.13-1.58). Nonusers (662/1571, 42.14%), compared with prior users, had lower overall inpatient portal use (all active functions: IRR 0.68, 95% CI 0.60-0.78) and lower use of specific functions, which included the function to review vitals and laboratory results (IRR 0.51, 95% CI 0.36-0.73) and the function to access the outpatient portal (IRR 0.53, 95% CI 0.45-0.62). In comparison with prior users, nonusers also had lower odds of being comprehensive users (defined as using 8 or more unique portal functions; odds ratio [OR] 0.57, 95% CI 0.45-0.73) or composite users (defined as comprehensive users who initiated a 75th or greater percentile of portal sessions) of the inpatient portal (OR 0.42, 95% CI 0.29-0.60). CONCLUSIONS: Patients' use of an inpatient portal during their hospital stay appeared to be influenced by a combination of factors, including prior outpatient portal use. For new users, hospitalization itself, a major event that can motivate behavioral changes, may have influenced portal use. In contrast, nonusers might have lower self-efficacy in their ability to use technology to manage their health, contributing to their lower portal use. Understanding the relationship between the use of outpatient and inpatient portals can help direct targeted implementation strategies that encourage individuals to use these tools to better manage their health across care settings.

Cross verification of independent dose recalculation, log files based, and phantom measurement-based pretreatment quality assurance for volumetric modulated arc therapy.

Independent treatment planning system (TPS) check with Mobius3D software, log files based quality assurance (QA) with MobiusFX, and phantom measurement-based QA with ArcCHECK were performed and cross verified for head-and-neck (17 patients), chest (16 patients), and abdominal (19 patients) cancer patients who underwent volumetric modulated arc therapy (VMAT). Dosimetric differences and percentage gamma passing rates (%GPs) were evaluated and compared for this cross verification. For the dosimetric differences in planning target volume (PTV) coverage, there was no significant difference among TPS vs. Mobius3D, TPS vs. MobiusFX, and TPS vs. ArcCHECK. For the dosimetric differences in organs at risks (OARs), the number of metrics with an average dosimetric differences higher than ±3% for TPS vs Mobius3D, TPS vs MobiusFX, and TPS vs ArcCHECK were 1, 1, 7; 2, 1, 4; 1, 1, 5 for the patients with head-and-neck, abdomen, and chest cancer, respectively. The %GPs of global gamma indices for Mobius3D and MobiousFX were above 97%, while it ranged from 92% to 96% for ArcCHECK. The %GPs of individual volume-based gamma indices were around 98% for Mobius3D and MobiousFX, except for γPTV for chest and abdominal cancer (88.9% to 92%); while it ranged from 86% to 99% for ArcCHECK. In conclusion, some differences in dosimetric metrics and gamma passing rates were observed with ArcCHECK measurement-based QA in comparison with independent dosecheck and log files based QA. Care must be taken when considering replacing phantom measurement-based IMRT/VMAT QA.

Trajectory log file sensitivity: A critical analysis using DVH and EPID.

AIM: The aim of this study was to investigate the sensitivity of the trajectory log file based quality assurance to detect potential errors such as MLC positioning and gantry positioning by comparing it with EPID measurement using the most commonly used criteria of 3%/3 mm. MATERIALS AND METHODS: An in-house program was used to modified plans using information from log files, which can then be used to recalculate a new dose distribution. The recalculated dose volume histograms (DVH) were compared with the originals to assess differences in target and critical organ dose. The dose according to the differences in DVH was also compared with dosimetry from an electronic portal imaging device. RESULTS: In all organs at risk (OARs) and planning target volumes (PTVs), there was a strong positive linear relationship between MLC positioning and dose error, in both IMRT and VMAT plans. However, gantry positioning errors exhibited little impact in VMAT delivery. For the ten clinical cases, no significant correlations were found between gamma passing rates under the criteria of 3%/3 mm for the composite dose and the mean dose error in DVH (r < 0.3, P > 0.05); however, a significant positive correlation was found between the gamma passing rate of 3%/3 mm (%) averaged over all fields and the mean dose error in the DVH of the VMAT plans (r = 0.59, P < 0.001). CONCLUSIONS: This study has successfully shown the sensitivity of the trajectory log file to detect the impact of systematic MLC errors and random errors in dose delivery and analyzed the correlation of gamma passing rates with DVH.

Meta-T: TetrisⓇ as an experimental paradigm for cognitive skills research.

Studies of human performance in complex tasks using video games are an attractive prospect, but many existing games lack a comprehensive way to modify the game and track performance beyond basic levels of analysis. Meta-T provides experimenters a tool to study behavior in a dynamic task environment with time-stressed decision-making and strong perceptual-motor elements, offering a host of experimental manipulations with a robust and detailed logging system for all user events, system events, and screen objects. Its experimenter-friendly interface provides control over detailed parameters of the task environment without need for programming expertise. Support for eye-tracking and computational cognitive modeling extend the paradigm's scope.

MRI log file analysis for workflow improvement.

Magnetic Resonance Imaging (MRI) is an important diagnostic scanning tool for the detection and monitoring of specific diseases and conditions. However, the equipment cost, maintenance and specialty training of the technologists make the examination expensive. Consequently, unnecessary scanner time caused by poor scheduling, repeated sequences, aborted sequences, scanner idleness, or capture of non-diagnostic or low-value sequences is an opportunity to reduce costs and increase efficiency. This paper analyzes data collected from log files on 29 scanners over several years. 'Wasted' time is defined and key performance indicators (KPIs) are identified. A decrease in exam duration results when actively modifying and monitoring the number of sequences that comprise the exam card for a protocol.

4D Monte Carlo dose reconstructions using surface motion measurements.

PURPOSE: To investigate the feasibility of a 4D Monte Carlo based dose reconstruction method to study the dosimetric impact of respiratory motion using surface motion measurements for patients undergoing VMAT treatments for Non-Small Cell Lung Cancer. METHODS: The 4Ddefdosxyznrc/EGSnrc algorithm was used to reconstruct VMAT doses delivered to the patients using machine log files and respiratory traces measured with the RADPOS 4D dosimetry system. The RADPOS sensor was adhered to the patient's abdomen prior to each treatment fraction and its position was used as a surrogate for tumour motion. Treatment log files were synchronized with the patient respiratory traces. Patient specific respiratory models were generated from deformable registration of the inhale and exhale 4DCT images and the respiratory traces. The reconstructed doses were compared to planned doses calculated with DOSXYZnrc/EGSnrc on the average-intensity and the exhale phase CT images. RESULTS: Respiratory motion measurements and log files were acquired for 2 patients over 5 treatment fractions each. The motion was predominantly along the anterior/posterior direction (A/P). The average respiratory amplitudes were 8.7 ± 2.7 mm and 10.0 ± 1.2 mm for Patient 1 and 2, respectively. Both patients displayed inter- and intra-fractional variations in the baseline position. Small inter-fractional differences were observed in the reconstructed doses for each patient. Differences between the reconstructed and planned doses were attributed to differences in organ volumes. CONCLUSION: The 4D reconstruction method was successfully implemented for the two patients studied. Small differences between the planned and reconstructed doses were observed due to the small tumour motion of these patients.

Comparing log file to measurement-based patient-specific quality assurance.

Recent technological advances have allowed the possibility of performing patient-specific quality assurance (QA) without time-intensive measurements. The objectives of this study are to: (1) compare how well the log file-based Mobius QA system agrees with measurement-based QA methods (ArcCHECK and portal dosimetry, PD) in passing and failing plans, and; (2) evaluate their error sensitivities. To these ends, ten phantom plans and 100 patient plans were measured with ArcCHECK and PD on VitalBeam, while log files were sent to Mobius for dose recalculation. Gamma evaluation was performed using criteria 3%/2 mm, per TG218 recommendations, and non-inferiority of the Mobius recalculation was determined with statistical testing. Ten random plans were edited to include systematic errors, then subjected to QA. Receiver operating characteristic curves were constructed to compare error sensitivities across the QA systems, and clinical significance of the errors was determined by recalculating dose to patients. We found no significant difference between Mobius, ArcCHECK, and PD in passing plans at the TG218 action limit. Mobius showed good sensitivity to collimator and gantry errors but not MLC bank shift errors, but could flag discrepancies in treatment delivery. Systematic errors were clinically significant only at large magnitudes; such unacceptable plans did not pass QA checks at the TG218 tolerance limit. Our results show that Mobius is not inferior to existing measurement-based QA systems, and can supplement existing QA practice by detecting real-time delivery discrepancies. However, it is still important to maintain rigorous routine machine QA to ensure reliability of machine log files.

Investigating learning processes through analysis of navigation behavior using log files.

The empirical study investigates what log files and process mining can contribute to promoting successful learning. We want to show how monitoring and evaluation of learning processes can be implemented in the educational life by analyzing log files and navigation behavior. Thus, we questioned to what extent log file analyses and process mining can predict learning outcomes. This work aims to provide support for learners and instructors regarding efficient learning with computer-based learning environments (CBLEs). We evaluated log file and questionnaire data from students (N = 58) who used a CBLE for two weeks. Results show a significant learning increase after studying with the CBLE with a very high effect size (p < .001, g = 1.71). A cluster analysis revealed two groups with significantly different learning outcomes accompanied by different navigation patterns. The time spent on learning-relevant pages and the interactivity with a CBLE are meaningful indicators for Recall and Transfer performance. Our results show that navigation behaviors indicate both beneficial and detrimental learning processes. Moreover, we could demonstrate that navigation behaviors impact the learning outcome. We present an easy-to-use approach for learners as well as instructors to promote successful learning by tracking the duration spent in a CBLE and the interactivity.

A structured FMEA approach to optimizing combinations of plan-specific quality assurance techniques for IMRT and VMAT QA.

BACKGROUND: Many commercial tools are available for plan-specific quality assurance (QA) of radiotherapy plans, with their functionality assessed in isolation. However, multiple QA tools are required to review the full range of potential errors. It is important to assess their effectiveness in combination with each other to look for ways to both streamline the QA process and to make certain that errors of high impact and/or high occurrence are caught before reaching patient treatment. PURPOSE: To develop a structured method to assess the effective risk reduction of combinations of QA methods for IMRT/VMAT treatments. METHODS: First, a structured prospective risk assessment was performed to establish the major failure modes (FMs) of IMRT/VMAT QA, and assign occurrence (O), severity (S), and baseline detectability (BD) rankings to them. The baseline assumed that chart checks and linear accelerator QA was performed, but no plan-specific secondary dose calculation or measurement was done. Second, the detectability of each FM for two secondary dose calculation methods and four plan measurement methods (point-based dose calculation, Monte-Carlo-based dose calculation, 2D fluence-based measurement, 2.5D phantom-based measurement, log file analysis with dose recalculation, and log file analysis combined with MLC QA) was determined. Third, we used a minimum detectability approach in addition to each FM's occurrence and severity to determine the optimal combination of QA methods. We analyzed the cumulative risk priority number of eight combinations of QA methods. The analysis was done on (1) all FMs, (2) FMs with high severity, (3) FMs with high-risk priority numbers (RPN) of O*S*BD, and (4) on FMs with both high severity and high RPN. RESULTS: Our analysis resulted in 54 FMs, including commissioning, planning, data transfer, and linear accelerator failures. 1D secondary dose calculation plus measurement provided a 19%-22% risk reduction from baseline. 1D/3D secondary dose calculation plus log files created a 25%-32% reduction. 3D secondary dose calculation plus measurement resulted in a 27%-34% reduction. 3D secondary dose calculation plus log files with additional machine QA provided the greatest reduction of 31%-42%. CONCLUSION: This novel structured approach to comparing combinations of QA methods will allow us to optimize our procedures, with the goal of detecting all clinically significant FMs. Our results show that log-file QA with 3D dose recalculation and supplemental machine QA provides better risk reduction than measurement-based QA. This work builds evidence to justify reducing or eliminating measurement-based PSQA with an independent 3D dose verification, log-file measurement, and appropriate supplementation of machine QA. The process also highlights FMs that cannot be caught by pre-treatment QA, prompting us to consider future directions for on-treatment QA.

A Log File Analysis on the Validity of Partial Solutions in Figural Matrices Tests.

As a component of many intelligence test batteries, figural matrices tests are an effective way to assess reasoning, which is considered a core ability of intelligence. Traditionally, the sum of correct items is used as a performance indicator (total solution procedure). However, recent advances in the development of computer-based figural matrices tests allow additional indicators to be considered for scoring. In two studies, we focused on the added value of a partial solution procedure employing log file analyses from a computer-based figural matrices test. In the first study (n = 198), we explored the internal validity of this procedure by applying both an exploratory bottom-up approach (using sequence analyses) and a complementary top-down approach (using rule jumps, an indicator taken from relevant studies). Both approaches confirmed that higher scores in the partial solution procedure were associated with higher structuredness in participants' response behavior. In the second study (n = 169), we examined the external validity by correlating the partial solution procedure in addition to the total solution procedure with a Grade Point Average (GPA) criterion. The partial solution procedure showed an advantage over the total solution procedure in predicting GPA, especially at lower ability levels. The implications of the results and their applicability to other tests are discussed.

Prediction of VMAT delivery accuracy using plan modulation complexity score and log-files analysis.

The purpose of this study was to develop a predictive model based on plan complexity metrics and linac log-files analysis to classify the dosimetric accuracy of VMAT plans. A total of 612 VMAT plans, corresponding to 1224 arcs, were analyzed. All VMAT arcs underwent pre-treatment verification that was performed by means of the dynamic log-files generated by the linac. The comparison of predicted (by TPS) and measured (by log-files) integral fluences was performed usingγ-analysis in terms of the percentage of points withγ-value smaller than one (γ%) and using a stringent 2%(local)/2 mm criteria. Thisγ-analysis was performed by a commercial software LinacWatch. The action limits (AL) were derived from the mean values, standard deviations and the confidence limit (CL) of theγ% distribution. A plan complexity metric, the modulation complexity score (MCS), based on the aperture beam area variability and leaf sequence variability was used as input variable of the model. A binary logistic regression (LR) model was developed to classify QA results as 'pass' (γ% ≥ AL) or 'fail' (γ% < AL). Receiver operator characteristics (ROC) curves were used to determine the optimal MCS threshold to flag 'failed' plans that need to be re-optimized. The model reliability was evaluated stratifying the plans in training, validation and testing groups. The confidence and action limits forγ% were found 20.1% and 79.9%, respectively. The accuracy of the model for the training and testing dataset was 97.4% and 98.0%, respectively. The optimal MCS threshold value for the identification of failed plans was 0.142, providing a true positive rate able to flag the plans failing QA of 91%. In clinical routine, the use of this MCS threshold may allow the prompt identification of overly modulated plans, then reducing the number of QA failures and improving the quality of VMAT plans used for treatment.

Prediction and classification of VMAT dosimetric accuracy using plan complexity and log-files analysis.

PURPOSE: We presented different machine learning models based on log files analysis and complexity indexes to predict and classify the dosimetric accuracy of VMAT plans. METHODS: A total of 1302 VMAT arcs from 651 treatment plans were analyzed using the modulation complexity score (MCS) and the dynamic log-files generated by the linac. Predicted and measured fluences were compared using γ-analysis in terms of mean γ-values (γmean) and γ-pass rate (γ%). A kernel regression model was developed aiming to predict individual γ% and γmean values. Multinomial logistic regression (LR), Naïve-Bayes (NB) and support vector machine (SVM) models were developed based on MCS values to classify QA results as "pass" (γ%greater than90 % and γmean < 0.5), "control" (80 % < Î³% < 90 % and 0.50 < Î³mean < 0.75) and "fail" (γ% < 80 % and γmean > 0.75). Training, validation and testing groups were used to evaluate the model reliability. A complexity-based traffic light protocol was implemented to flag pass (green light), control (orange light) and failed plans (red light). RESULTS: Prediction accuracy of residuals for γ% was 2.1 % and 2.2 % in the training and testing cohorts, respectively. For 2 %(local)/2mm, both γ% and γmean classification performances reported weighted precision, recall and F1-values greater than 90 % for all machine learning models. The optimal MCS threshold value for the identification of failed plans was 0.130, with a sensibility and specificity of 0.994 and 0.952, respectively. The optimal MCS threshold for reliable plans was 0.270, with a sensitivity and specificity of 0.925 and 0.922, respectively. CONCLUSIONS: Machine learning can accurately predict the dosimetric accuracy of VMAT treatments, representing an efficient tool to assist patient-specific QA.

Deep Learning for Patient-Specific Quality Assurance: Predicting Gamma Passing Rates for IMRT Based on Delivery Fluence Informed by log Files.

Objectives: In this study, we propose a deep learning-based approach to predict Intensity-modulated radiation therapy (IMRT) quality assurance (QA) gamma passing rates using delivery fluence informed by log files. Methods: A total of 112 IMRT plans for chest cancers were planned and measured by portal dosimetry equipped on TrueBeam linac. The convolutional neural network (CNN) based learning model was trained using delivery fluence as inputs and gamma passing rates (GPRs) of 4 different criteria (3%/3 mm, 2%/3 mm, 3%/2 mm, and 2%/2 mm) as outputs. Model performance for both validation and test sets was assessed using mean absolute error (MAE), mean squared error (MSE), root MSE (RMSE), Spearman rank correlation coefficients (Sr), and Determination coefficient (R2) between the measured and predicted GPR values. Results: In the test set, the MAE of the prediction model were 0.402, 0.511, 1.724, and 2.530, the MSE were 0.640, 0.986, 6.654, and 9.508, the RMSE were 0.800, 0.993, 2.580, and 3.083, the Sr were 0.643, 0.684, 0.821, and 0.824 (P < .001) and the R2 were 0.4110, 0.4666, 0.6677, and 0.6769 for 3%/3 mm, 3%/2 mm, 2%/3 mm, and 2%/2 mm, respectively. The MAE and RMSE of the prediction model decreased with stricter gamma criteria while the Sr and R2 between measured and predicted GPR values increased. Conclusions: The CNN prediction model based on delivery fluence informed by log files could accurately predict IMRT QA passing rates for different gamma criteria. It could reduce QA workload and improve efficiency in pretreatment QA. Our results suggest that the CNN prediction model based on delivery fluence informed by log files may be a promising tool for the gamma evaluation of IMRT QA.

Outpatient portal use among pregnant individuals: Cross-sectional, temporal, and cluster analysis of use.

Background: Outpatient portal technology can improve patient engagement. For pregnant individuals, the level of engagement could have important implications for maternal and infant outcomes. Objective: This study: (1) cross-sectionally and temporally characterized the outpatient portal use among pregnant individuals seen at our academic medical center; and (2) identified clusters of the outpatient portal user groups based on the cross-sectional and temporal patterns of use. Methods: We used outpatient portal server-side log files to execute a hierarchical clustering algorithm to group 7663 pregnant individuals based on proportions of outpatient portal function use. Post-hoc analyses were performed to further assess outpatient portal use on key encounter characteristics. Results: The most frequently used functions were MyRecord (access personal health information), Visits (manage appointments), Messaging (send/receive messages), and Billing (view bills, insurance information). Median outpatient portal function use plateaued by the third trimester. Four distinct clusters were identified among all pregnant individuals: "Schedulers," "Resulters," "Intense Digital Engagers," and "Average Users." Post-hoc analyses revealed that the use of the Visits function increased and the use of the MyRecord function decreased over time among clusters. Conclusions: Our identification of distinct cluster groups of outpatient portal users among pregnant individuals underscores the importance of avoiding the use of generalizations when describing how such patients might engage with patient-facing technologies such as an outpatient portal. These results can be used to improve user experience and training with outpatient portal functions and may educate maternal health providers on patient engagement with the outpatient portal.

Outpatient portal use in prenatal care: differential use by race, risk, and area social determinants of health.

To report the relationship of outpatient portal (OPP) use with clinical risk, area social determinants of health (SDoH), and race/ethnicity among pregnant women. Regression models predicting overall and individual portal feature use (main effects and interactions) based on key variables were specified using log files and clinical data. Overall OPP use among non-Hispanic Black women or patients who lived in lower SDoH neighborhoods were significantly less. High-risk pregnancy patients were likely to use the OPP more than those with normal-risk pregnancy. We found similar associations with individual OPP features, like Visit (scheduling) and My Record (test results). We also found significant interactive associations between race/ethnicity, clinical risk, and SDoH. Non-Hispanic Black women and those living in lower SDoH areas used OPP less than non-Hispanic White women from similar or affluent areas. More research must be conducted to learn of OPP use implications for pregnant women with specific clinical diagnoses.

Real-Time Analysis of the Log Files of the HeartWare Continuous-Flow Left Ventricular Assist Device for the Early Diagnosis of Pump Thrombosis: a Step Forward Toward Clinical Translation.

We evaluated the real-time diagnostic capability of a new tool enhancing early diagnosis of pump thrombosis (PT) of the HeartWare left ventricular assist device via time-frequency analysis (TFA) of the log files. We analyzed 173 log files, including 24 (14%) associated with a clinical diagnosis of PT and 149 (86%) controls. The 30-day log file records were discretized into consecutive windows of a 24-h duration, which were iteratively acquired and processed via TFA. This way, we simulated longitudinal acquisition of pump parameters and provided real-time analysis of consecutive data, thus resembling the clinical scenario. Sensitivity and specificity of the tool were 79% and 84%, respectively. Sensitivity against PT events with progressive "build-up" thrombus increased up to 95%, and early signs of a forthcoming PT were identified 10±8 days prior to its clinical manifestation. This study demonstrates high reliability and the potential for effective clinical translation of this prognostic tool.