Using Intent-to-Attend to Predict Attendance in Community-Based Relationship Education.

Government-supported relationship education has provided resources for inclusion of economically vulnerable and ethnically diverse participants; however, many grantees and programs struggled to retain couples in longitudinal studies, which has likely influenced study effects and threatened internal validity. In the present study, we assessed 1,056 couples' baseline relationship satisfaction and intent-to-attend their next scheduled visit while participating in a randomized controlled trial of relationship education and evaluated the predictive ability of their responses to remain in the six-month study. We conducted actor-partner interdependence models for couples, using a probit cross-lagged regression with a structural equation modeling framework, to test the dyadic influence of intent-to-attend on future couple attendance. We also examined the influence of higher or lower baseline relationship satisfaction between partners and group assignment (treatment or wait-list control) on attendance. Intent-to-attend scores were associated with attendance for couples at the one-month follow-up, and early attendance was the biggest predictor of later attendance. Additionally, baseline intent-to-attend scores predicted later intent-to-attend scores for all follow-up time points. However, we found no partner effects, and no effects for the influence of baseline relationship satisfaction or group assignment. We discuss practical suggestions for including intent-to-attend in future studies, relationship education programming, and general therapy practice.

La capacitación en relaciones financiada por el gobierno ha facilitado recursos para la inclusión de participantes económicamente vulnerables y de distintas etnias; sin embargo, a muchos beneficiarios y programas les costó mantener a las parejas en los estudios longitudinales, lo cual probablemente haya influido en los efectos de los estudios y amenazado su validez interna. En el presente estudio, evaluamos la satisfacción con la relación en el momento basal de 1056 parejas y la intención de asistir a su próxima visita programada mientras participaban en un ensayo controlado aleatorizado de capacitación en relaciones, y evaluamos la capacidad predictiva de sus respuestas para permanecer en el estudio de seis meses. Implementamos modelos de interdependencia actor-pareja para las parejas usando un modelo Probit de regresión y retardo cruzado con un marco de modelos de ecuaciones estructurales con el fin de evaluar la influencia diádica de la intención de asistir en la asistencia futura de la pareja. También analizamos la influencia del nivel más bajo o más alto de satisfacción con la relación en el momento basal entre los integrantes de la pareja y la distribución a un grupo (de tratamiento o de control en lista de espera) en la asistencia. Los puntajes de la intención de asistir estuvieron asociados con la asistencia de las parejas en el seguimiento de un mes, y la asistencia inicial fue la mayor predictora de la asistencia posterior. Además, los puntajes de la intención de asistir en el momento basal predijeron los puntajes posteriores de la intención de asistir de todos los momentos de seguimiento. Sin embargo, no hallamos efectos de la pareja ni efectos de la influencia de la satisfacción con la relación o la distribución a un grupo en el momento basal. Comentamos sugerencias prácticas para incluir la intención de asistir en estudios futuros, en programas de capacitación en relaciones y en la práctica de la terapia general.

Evaluation of Deep Learning Clinical Target Volumes Auto-Contouring for Magnetic Resonance Imaging-Guided Online Adaptive Treatment of Rectal Cancer.

Purpose: Segmentation of clinical target volumes (CTV) on medical images can be time-consuming and is prone to interobserver variation (IOV). This is a problem for online adaptive radiation therapy, where CTV segmentation must be performed every treatment fraction, leading to longer treatment times and logistic challenges. Deep learning (DL)-based auto-contouring has the potential to speed up CTV contouring, but its current clinical use is limited. One reason for this is that it can be time-consuming to verify the accuracy of CTV contours produced using auto-contouring, and there is a risk of bias being introduced. To be accepted by clinicians, auto-contouring must be trustworthy. Therefore, there is a need for a comprehensive commissioning framework when introducing DL-based auto-contouring in clinical practice. We present such a framework and apply it to an in-house developed DL model for auto-contouring of the CTV in rectal cancer patients treated with MRI-guided online adaptive radiation therapy. Methods and Materials: The framework for evaluating DL-based auto-contouring consisted of 3 steps: (1) Quantitative evaluation of the model's performance and comparison with IOV; (2) Expert observations and corrections; and (3) Evaluation of the impact on expected volumetric target coverage. These steps were performed on independent data sets. The framework was applied to an in-house trained nnU-Net model, using the data of 44 rectal cancer patients treated at our institution. Results: The framework established that the model's performance after expert corrections was comparable to IOV, and although the model introduced a bias, this had no relevant impact on clinical practice. Additionally, we found a substantial time gain without reducing quality as determined by volumetric target coverage. Conclusions: Our framework provides a comprehensive evaluation of the performance and clinical usability of target auto-contouring models. Based on the results, we conclude that the model is eligible for clinical use.

A network score-based metric to optimize the quality assurance of automatic radiotherapy target segmentations.

Background and purpose: Existing methods for quality assurance of the radiotherapy auto-segmentations focus on the correlation between the average model entropy and the Dice Similarity Coefficient (DSC) only. We identified a metric directly derived from the output of the network and correlated it with clinically relevant metrics for contour accuracy. Materials and Methods: Magnetic Resonance Imaging auto-segmentations were available for the gross tumor volume for cervical cancer brachytherapy (106 segmentations) and for the clinical target volume for rectal cancer external-beam radiotherapy (77 segmentations). The nnU-Net's output before binarization was taken as a score map. We defined a metric as the mean of the voxels in the score map above a threshold (λ). Comparisons were made with the mean and standard deviation over the score map and with the mean over the entropy map. The DSC, the 95th Hausdorff distance, the mean surface distance (MSD) and the surface DSC were computed for segmentation quality. Correlations between the studied metrics and model quality were assessed with the Pearson correlation coefficient (r). The area under the curve (AUC) was determined for detecting segmentations that require reviewing. Results: For both tasks, our metric (λ = 0.30) correlated more strongly with the segmentation quality than the mean over the entropy map (for surface DSC, r > 0.65 vs. r < 0.60). The AUC was above 0.84 for detecting MSD values above 2 mm. Conclusions: Our metric correlated strongly with clinically relevant segmentation metrics and detected segmentations that required reviewing, indicating its potential for automatic quality assurance of radiotherapy target auto-segmentations.