Search

Simple Summary Due to the fact that they take inter-slice information into account, 3D- and 2.5D-convolutional neural networks (CNNs) potentially perform better in tumor detection tasks than 2D-CNNs. However, this potential benefit is at the expense of increased computational power and the need for segmentations as an input. Therefore, in this study we aimed to detect vestibular schwannomas (VSs) in individual magnetic resonance imaging (MRI) slices by using a 2D-CNN. We retrained (539 patients) and internally validated (94 patients) a pretrained CNN using contrast-enhanced MRI slices from one institution. Furthermore, we externally validated the CNN using contrast-enhanced MRI slices from another institution. This resulted in an accuracy of 0.949 (95% CI 0.935-0.963) and 0.912 (95% CI 0.866-0.958) for the internal and external validation, respectively. Our findings indicate that 2D-CNNs might be a promising alternative to 2.5-/3D-CNNs for certain tasks thanks to the decreased requirement for computational power and the fact that there is no need for segmentations. In this study. we aimed to detect vestibular schwannomas (VSs) in individual magnetic resonance imaging (MRI) slices by using a 2D-CNN. A pretrained CNN (ResNet-34) was retrained and internally validated using contrast-enhanced T1-weighted (T1c) MRI slices from one institution. In a second step, the model was externally validated using T1c- and T1-weighted (T1) slices from a different institution. As a substitute, bisected slices were used with and without tumors originating from whole transversal slices that contained part of the unilateral VS. The model predictions were assessed based on the categorical accuracy and confusion matrices. A total of 539, 94, and 74 patients were included for training, internal validation, and external T1c validation, respectively. This resulted in an accuracy of 0.949 (95% CI 0.935-0.963) for the internal validation and 0.912 (95% CI 0.866-0.958) for the external T1c valid

Introduction: Many proposed algorithms for tumor detection rely on 2.5/3D convolutional neural networks (CNNs) and the input of segmentations for training. The purpose of this study is therefore to assess the performance of tumor detection on single MRI slices containing vestibular schwannomas (VS) as a computationally inexpensive alternative that does not require the creation of segmentations. Methods: A total of 2992 T1-weighted contrast-enhanced axial slices containing VS from the MRIs of 633 patients were labeled according to tumor location, of which 2538 slices from 539 patients were used for training a CNN (ResNet-34) to classify them according to the side of the tumor as a surrogate for detection and 454 slices from 94 patients were used for internal validation. The model was then externally validated on contrast-enhanced and non-contrast-enhanced slices from a different institution. Categorical accuracy was noted, and the results of the predictions for the validation set are provided with confusion matrices. Results: The model achieved an accuracy of 0.928 (95% CI: 0.869-0.987) on contrast-enhanced slices and 0.795 (95% CI: 0.702-0.888) on non-contrast-enhanced slices from the external validation cohorts. The implementation of Gradient-weighted Class Activation Mapping (Grad-CAM) revealed that the focus of the model was not limited to the contrast-enhancing tumor but to a larger area of the cerebellum and the cerebellopontine angle. Conclusions: Single-slice predictions might constitute a computationally inexpensive alternative to training 2.5/3D-CNNs for certain detection tasks in medical imaging even without the use of segmentations. Head-to-head comparisons between 2D and more sophisticated architectures could help to determine the difference in accuracy, especially for more difficult tasks.

The role of postoperative radiation therapy (PORT) in stage III N2 NSCLC is controversial. We analyzed decision-making for PORT among European radiation oncology experts in lung cancer. Twenty-two experts were asked before and after presentation of the results of the LungART trial to describe their decision criteria for PORT in the management of pN+ NSCLC patients. Treatment strategies were subsequently converted into decision trees and analyzed. Following decision criteria were identified: extracapsular nodal extension, incomplete lymph node resection, multistation lymph nodes, high nodal tumor load, poor response to induction chemotherapy, ineligibility to receive adjuvant chemotherapy, performance status, resection margin, lung function and cardiopulmonary comorbidities. The LungART results had impact on decision-making and reduced the number of recommendations for PORT. The only clear indication for PORT was a R1/2 resection. Six experts out of ten who initially recommended PORT for all R0 resected pN2 patients no longer used PORT routinely for these patients, while four still recommended PORT for all patients with pN2. Fourteen experts used PORT only for patients with risk factors, compared to eleven before the presentation of the LungART trial. Four experts stated that PORT was never recommended in R0 resected pN2 patients regardless of risk factors. After presentation of the LungART trial results at ESMO 2020, 82% of our experts still used PORT for stage III pN2 NSCLC patients with risk factors. The recommendation for PORT decreased, especially for patients without risk factors. Cardiopulmonary comorbidities became more relevant in the decision-making for PORT.

BACKGROUND: Whole brain radiotherapy (WBRT) is a common treatment option for brain metastases secondary to non-small cell lung cancer (NSCLC). Data from the QUARTZ trial suggest that WBRT can be omitted in selected patients and treated with optimal supportive care alone. Nevertheless, WBRT is still widely used to treat brain metastases secondary to NSCLC. We analysed decision criteria influencing the selection for WBRT among European radiation oncology experts. METHODS: Twenty-two European radiation oncology experts in lung cancer as selected by the European Society for Therapeutic Radiation Oncology (ESTRO) for previous projects and by the Advisory Committee on Radiation Oncology Practice (ACROP) for lung cancer were asked to describe their strategies in the management of brain metastases of NSCLC. Treatment strategies were subsequently converted into decision trees and analysed for agreement and discrepancies. RESULTS: Eight decision criteria (suitability for SRS, performance status, symptoms, eligibility for targeted therapy, extra-cranial tumour control, age, prognostic scores and "Zugzwang" (the compulsion to treat)) were identified. WBRT was recommended by a majority of the European experts for symptomatic patients not suitable for radiosurgery or fractionated stereotactic radiotherapy. There was also a tendency to use WBRT in the ALK/EGFR/ROS1 negative NSCLC setting. CONCLUSION: Despite the results of the QUARTZ trial WBRT is still widely used among European radiation oncology experts.