Response Assessment
Recently Published Documents





2021 ◽  
Vol 11 ◽  
Jungheum Cho ◽  
Young Jae Kim ◽  
Leonard Sunwoo ◽  
Gi Pyo Lee ◽  
Toan Quang Nguyen ◽  

BackgroundAlthough accurate treatment response assessment for brain metastases (BMs) is crucial, it is highly labor intensive. This retrospective study aimed to develop a computer-aided detection (CAD) system for automated BM detection and treatment response evaluation using deep learning.MethodsWe included 214 consecutive MRI examinations of 147 patients with BM obtained between January 2015 and August 2016. These were divided into the training (174 MR images from 127 patients) and test datasets according to temporal separation (temporal test set #1; 40 MR images from 20 patients). For external validation, 24 patients with BM and 11 patients without BM from other institutions were included (geographic test set). In addition, we included 12 MRIs from BM patients obtained between August 2017 and March 2020 (temporal test set #2). Detection sensitivity, dice similarity coefficient (DSC) for segmentation, and agreements in one-dimensional and volumetric Response Assessment in Neuro-Oncology Brain Metastases (RANO-BM) criteria between CAD and radiologists were assessed.ResultsIn the temporal test set #1, the sensitivity was 75.1% (95% confidence interval [CI]: 69.6%, 79.9%), mean DSC was 0.69 ± 0.22, and false-positive (FP) rate per scan was 0.8 for BM ≥ 5 mm. Agreements in the RANO-BM criteria were moderate (κ, 0.52) and substantial (κ, 0.68) for one-dimensional and volumetric, respectively. In the geographic test set, sensitivity was 87.7% (95% CI: 77.2%, 94.5%), mean DSC was 0.68 ± 0.20, and FP rate per scan was 1.9 for BM ≥ 5 mm. In the temporal test set #2, sensitivity was 94.7% (95% CI: 74.0%, 99.9%), mean DSC was 0.82 ± 0.20, and FP per scan was 0.5 (6/12) for BM ≥ 5 mm.ConclusionsOur CAD showed potential for automated treatment response assessment of BM ≥ 5 mm.

2021 ◽  
Vol 6 ◽  
Xiaoming Zhai ◽  
Kevin C. Haudek ◽  
Christopher Wilson ◽  
Molly Stuhlsatz

Estimating and monitoring the construct-irrelevant variance (CIV) is of significant importance to validity, especially for constructed response assessments with rich contextualized information. To examine CIV in contextualized constructed response assessments, we developed a framework including a model accounting for CIV and a measurement that could differentiate the CIV. Specifically, the model includes CIV due to three factors: the variability of assessment item scenarios, judging severity, and rater scoring sensitivity to the scenarios in tasks. We proposed using the many-facet Rasch measurement (MFRM) to examine the CIV because this measurement model can compare different CIV factors on a shared scale. To demonstrate how to apply this framework, we applied the framework to a video-based science teacher pedagogical content knowledge (PCK) assessment, including two tasks, each with three scenarios. Results for task I, which assessed teachers’ analysis of student thinking, indicate that the CIV due to the variability of the scenarios was substantial, while the CIV due to judging severity and rater scoring sensitivity of the scenarios in teacher responses was not. For task II, which assessed teachers’ analysis of responsive teaching, results showed that the CIV due to the three proposed factors was all substantial. We discuss the conceptual and methodological contributions, and how the results inform item development.

2021 ◽  
pp. canres.1499.2021
Ramona Woitek ◽  
Mary A McLean ◽  
Stephan Ursprung ◽  
Oscar M Rueda ◽  
Raquel Manzano Garcia ◽  

2021 ◽  
Vol 23 (Supplement_4) ◽  
pp. iv22-iv23
Markand Patel ◽  
Dilina Rajapakse ◽  
Jian Ping Jen ◽  
Sara Meade ◽  
Helen Benghiat ◽  

Abstract Aims Following stereotactic radiosurgery (SRS), brain metastases can increase in size in up to a third of cases. Conventional magnetic resonance imaging (MRI) has a limited role to distinguish between tumour recurrence and SRS-induced changes, which can impact patient management. Delayed contrast MRI treatment response assessment maps (TRAM) use the principle of contrast clearance seen in other tumours, where high vascularity shows a rapid rise in contrast as well as rapid clearance, whereas areas of damaged or low vascularity show accumulation of contrast. We aimed to assess the ability of delayed contrast MRI and multiparametric MRI techniques of diffusion-weighted imaging (DWI), perfusion-weighted imaging (PWI) and MR spectroscopy (MRS) to distinguish between radiation-related effects and tumour tissue, as these techniques assess tissue physiological and metabolic information. Method A retrospective review was performed on 23 patients who had delayed contrast and multiparametric MRI between October 2018 to April 2020. Studies were restricted to cases with brain metastases enlarging post-SRS with uncertainty at the MDT meeting regarding progression or treatment-related change, impacting the patient’s management. MRI was performed at 3T including DWI, PWI, MRS with short and intermediate echo times, and 3D T1 MPRAGE at 3-5, 20-30 and 70-90 minutes after administration of intravenous contrast. Contrast clearance analysis was performed by selecting an enhancing region of interest (ROI), measuring signal intensities at the three different timepoints and taking apparent diffusion coefficient (ADC) and relative cerebral blood volume (rCBV) values from the ROI. Choline/Creatine values were calculated from a single-voxel (10 mm isotropic) encompassing the entire contrast-enhancing lesion. Outcome was established from MRI follow-up at 6 months, with a stable or responding lesion considered treatment-related changes and increase considered progression. Results Across 23 patients, 24 metastases were assessed. Two patients were excluded as appropriate follow-up was not available. Sites of primary tumours included breast (n=8), lung (n=6), melanoma (n=4), neuroendocrine tumour from the lung (n=2) and renal cell carcinoma (n=2). Mean age was 56 years and 50% were female. In this cohort, 59% (n=13) were classified as having radiation-related changes on follow-up. Delayed MRI contrast clearance between the 3-5 and 70-90 minute imaging was significantly higher in cases of progression (23.6% vs. 2.5% decrease, p<0.05), as were the rCBV and Cho/Cr ratio (rCBV 3.1 vs. 1.5 and Cho/Cr ratio 2.3 vs. 1.4, p<0.05). Accuracy, sensitivity and specificity of using TRAM alone (contrast clearance decrease of >0%) for progression was 63%/100%/38%, PWI alone (rCBV cut-off 2.0) yielded results of 77%/75%/79% and for both Cho/Cr ratio alone (cut-off 1.8) and combined with TRAM, it was 90%/88%/92%. Neuroradiologist assessment of all techniques was 95%/100%/92%. Conclusion This study shows the effectiveness of delayed contrast and multiparametric MRI for treatment response assessment in patients with brain metastases treated by SRS in clinical practice. Although a delayed contrast MRI study is a very sensitive tool for detecting tumour progression, it lacks specificity. The accuracy of differentiating between tumour and treatment-related effects increases when delayed contrast MRI is used in combination with other advanced techniques such as MRS. By combining all these techniques, neuroradiologists had the highest accuracy, sensitivity and specificity for detecting progression in post-SRS brain metastases.

2021 ◽  
Vol 23 (Supplement_4) ◽  
pp. iv1-iv1
Markand Patel ◽  
Jinfeng Zhan ◽  
Kal Natarajan ◽  
Robert Flintham ◽  
Nigel Davies ◽  

Abstract Aims Treatment response assessment in glioblastoma is challenging. Patients routinely undergo conventional magnetic resonance imaging (MRI), but it has a low diagnostic accuracy for distinguishing between true progression (tPD) and pseudoprogression (psPD) in the early post-chemoradiotherapy time period due to similar imaging appearances. The aim of this study was to use artificial intelligence (AI) on imaging data, clinical characteristics and molecular information within machine learning models, to distinguish between and predict early tPD from psPD in patients with glioblastoma. Method The study involved retrospective analysis of patients with newly-diagnosed glioblastoma over a 3.5 year period (n=340), undergoing surgery and standard chemoradiotherapy treatment, with an increase in contrast-enhancing disease on the baseline MRI study 4-6 weeks post-chemoradiotherapy. Studies had contrast-enhanced T1-weighted imaging (CE-T1WI), T2-weighted imaging (T2WI) and apparent diffusion coefficient (ADC) sequences, acquired at 1.5 Tesla with 6-months follow-up to determine the reference standard outcome. 76 patients (mean age 55 years, range 18-76 years, 39% female, 46 tPD, 30 psPD) were included. Machine learning models utilised information from clinical characteristics (age, gender, resection extent, performance status), O6-methylguanine-DNA methyltransferase (MGMT) promoter methylation status and 307 quantitative imaging features; extracted from baseline study CE-T1WI/ADC and T2WI sequences using semi-automatically segmented enhancing disease and perilesional oedema masks respectively. Feature selection was performed within bootstrapped cross-validated recursive feature elimination with a random forest algorithm and Naïve Bayes five-fold cross-validation to validate the final model. Results Treatment response assessment based on the standard-of-care reports by clinical neuroradiologists showed an accuracy of 33% (sensitivity/specificity 52%/3%) to distinguish between tPD and psPD from the early post-treatment MRI study at 4-6 weeks. Machine learning-based models based on clinical and molecular features alone demonstrated an AUC of 0.66 and models using radiomic features alone from the early post-treatment MRI demonstrated an AUC of 0.46-0.69 depending on the feature and mask subset. A combined clinico-radiomic model utilising top common features demonstrated an AUC of 0.80 and an accuracy of 74% (sensitivity/specificity 78%/67%). The features in the final model were age, MGMT promoter methylation status, two shape-based features from the enhancing disease mask (elongation and sphericity), three radiomic features from the enhancing disease mask on ADC (kurtosis, correlation, contrast) and one radiomic feature from the perilesional oedema mask on T2WI (dependence entropy). Conclusion Current standard-of-care glioblastoma treatment response assessment imaging has limitations. In this study, the use of AI through a machine learning-based approach incorporating clinical characteristics and MGMT promoter methylation status with quantitative radiomic features from standard MRI sequences at early 4-6 weeks post-treatment imaging showed the best model performance and a higher accuracy to distinguish between tPD and psPD for early prediction of glioblastoma treatment response.

2021 ◽  
Vol 3 (Supplement_4) ◽  
pp. iv7-iv7
Jana Ivanidze ◽  
Sean Kim ◽  
Michelle Roytman ◽  
Rohan Ramakrishna ◽  
Susan Pannullo ◽  

Abstract PURPOSE Postoperative PET/MRI with [68Ga]-DOTATATE can differentiate residual meningioma from postsurgical change, aid in target delineation, and portend a more favorable dosimetry with decreased PTV and organ-at-risk dose. Our purpose was to demonstrate utility of DOTATATE PET/MR for radiosurgical treatment (RT) response assessment in meningiomas. METHODS Patients underwent postoperative radiation treatment planning using DOTATATE PET/MRI as part of our IRB-approved prospective trial. Both DOTATATE PET and gadolinium-enhanced T1 weighted MR imaging were incorporated in RT-planning. All patients underwent follow-up DOTATATE PET/MRI at 6-12 months following completion of radiosurgery. Maximum absolute standardized uptake value (SUV) and SUV ratio (SUVR) of lesion/ superior sagittal sinus SUV were obtained. RANO criteria were applied to determine significance of change in size. Statistical analyses were performed using paired t-tests. RESULTS 13 patients (15% WHO-I, 54% WHO-II, 23% WHO-III, 8% WHO grade unknown) were followed postoperatively with pre- and post-RT DOTATATE PET/MRI. 29 meningiomas were treated. 46% (6/13) of subjects received SBRT and 54% (7/13) received SRS. Post-RT DOTATATE PET/MRI demonstrated a 46.4% SUV decrease (p-value = 0.0001) and a 60.8% SUVR decrease (p-value < 0.0001). Of 21 measurable lesions, the size product decreased by 21%; while this decrease was statistically significant (p-value = 0.0008), it was below the 25% decrease defined as clinically significant by RANO guidelines. To date, all patients remain stable radiographically without evidence of recurrence (mean follow-up post RT: 14 months; range: 6-24 months). CONCLUSIONS DOTATATE PET SUV and SUVR demonstrated marked, significant decrease post radiosurgery. Lesion size decrease was statistically significant but not clinically significant by RANO criteria. DOTATATE PET/MR thus represents a promising approach to aid in response assessment for radiosurgically treated meningiomas. Longer-term follow-up is needed to determine the correlation between the degree of post-RT SUV and/or SUVR decrease and progression-free-survival.

Sign in / Sign up

Export Citation Format

Share Document