Grevekim4860

BACKGROUND AND PURPOSE Manual delineation of clinical target volumes (CTVs) and organs at risk (OARs) is time-consuming, and automatic contouring tools lack clinical validation. We aimed to construct and validate the use of convolutional neural networks (CNNs) to set better contouring standards for rectal cancer radiotherapy. MATERIALS AND METHODS We retrospectively collected and evaluated computed tomography (CT) scans of 199 rectal cancer patients treated at our hospital from February 2018 to April 2019. Two CNNs-DeepLabv3+ for extracting high-level semantic information and ResUNet for extracting low-level visual features-were used for the CTV and small intestine contouring, and bladder and femoral head contouring, respectively. Contouring quality was compared using the paired t test. Five-point objective grading was performed independently by two experienced radiation oncologists and verified by a third. The CNN manual correction time was recorded. RESULTS CTVs calculated using DeepLabv3+ (CTVDeepLabv3+) had significant quantitative parameter advantages over CTVResUNet (volumetric Dice coefficient, 0.88 vs 0.87, P = 0.0005; surface Dice coefficient, 0.79 vs 0.78, P = 0.008). Among 315 graded cases, DeepLabv3+ obtained the highest scores with 284 cases, consistent with the objective criteria, whereas CTVResUNet had the minimum mean manual correction time (7.29 min). DeepLabv3+ performed better than ResUNet for small intestine contouring and ResUNet performed better for bladder and femoral head contouring. The manual correction time for OARs was less then 4 min for both models. CONCLUSION CNNs at various feature resolution levels well delineate rectal cancer CTVs and OARs, displaying high quality and requiring shorter computation and manual correction time. BACKGROUND & PURPOSE Stereotactic ablative radiation therapy (SABR) is an emerging treatment option for patients with pulmonary metastases; identifying patients who would benefit from SABR can improve outcomes. MATERIALS & METHODS We retrospectively analyzed local failure (LF), distant failure (DF), overall survival (OS), and toxicity in 317 patients with 406 pulmonary metastases treated with SABR in January 2006-September 2017 at a tertiary cancer center. RESULTS Median follow-up time was 23 months. Primary adrenal, colorectal, sarcoma, or pancreatic ("less responsive") tumors led to high rates of LF. LF rates for patients with less responsive vs. responsive tumors were 4.6% vs. 1.6% at 12 months and 12.8% vs. 3.9% at 24 months (hazard ratio [HR] 0.29, 95% confidence interval [CI] 0.11-0.73; Log-Rank P = 0.0087). A nomogram for 24-month local control was created using Cox multivariate factors (surgical history, planning target volume, primary disease site, lung lobe location). Treating patients with ≤3 pulmonary metastases vs. >3 pulmonary metastases was associated with improved 24-month (74.2% vs. 59.3%) and 48-month (47.7% vs. 35.1%) OS (HR 0.66, 95% CI 0.47-0.95; Log-Rank P = 0.043), and reduced 12-month (22.5% vs. 50.8%) and 24-month (31.8% vs. 61.4%) intrathoracic DF (HR 0.53, 95% CI 0.38-0.74; Log-Rank P  less then  0.0001). The most common toxicity was asymptomatic pneumonitis (14.8%). Six patients had grade 3 events (5 pneumonitis, 1 brachial plexus). CONCLUSIONS SABR for pulmonary metastases was effective and well tolerated. Irradiating limited intrathoracic sites of disease led to improved OS and intrathoracic DM. Higher SABR doses or surgery could be considered for less radio-responsive primary tumors. AIMS To report oncologic and functional outcomes in terms of tumor control and toxicity of carbon ion radiotherapy (CIRT) in reirradiation setting for recurrent salivary gland tumors at CNAO. METHODS From November 2013 to September 2016, 51 consecutive patients with inoperable recurrent salivary gland tumors were retreated with CIRT in the frame of the phase II protocol CNAO S14/2012C for recurrent head and neck tumors. RESULTS Majority of pts (74.5%) had adenoid cystic carcinoma, mainly rcT4a (51%) and rcT4b (37%). Median dose of prior photon based radiotherapy was 60 Gy. Median dose of CIRT was 60 Gy [RBE] at a mean of 3 Gy [RBE] per fraction. During reirradiation, 19 patients (37.3%) experienced grade G1 toxicity, 19 pts (37.3%) had G2 and 2 pts (3.9%) had G3. Median follow up time was 19 months. Twenty one (41.2%) patients had stable disease and 30 (58.8%) tumor progression at the time of last follow up. Furthermore, 9 (18%) patients had G1 late toxicity, 19 (37%) had G2 and 9 (17. A1874 5%) had G3. Using the Kaplan Meier method, progression free survival (actuarial) at one and two years were 71.7% and 52.2% respectively. Estimated overall survival (actuarial) at one and two years were 90.2% and 64%, respectively. CONCLUSIONS CIRT is a good option for retreatment of inoperable recurrent salivary gland tumors with acceptable rates of acute and late toxicity. Longer follow up time is needed to assess the effectiveness of CIRT in reirradiation setting of salivary gland tumors. Preferences towards unfamiliar drink brands may be influenced through subliminal conditioning. This can involve associating unfamiliar brands (CS) with positively valenced attributes (US) under constrained visual conditions to prevent the former's conscious detection. According to learning theory, CS associated with positive US should become increasingly preferred as the latter's positive valences generalizes (transfer) across associated CS. Similarly, correlating CS with negative US should reduce CS-associated preferences. There is some evidence that CS-associated preferences can be reliably influenced through subliminal conditioning (Elgendi et al., 2018). Conversely, there is also evidence that subliminal conditioning does not effectively alter evaluations of CS valence (Heycke et al., 2018). Those works suggest CS preferences may be more susceptible to subliminal valence transfer relative to CS evaluations. We explored this hypothesis presently, where four pairs of supraliminal/visible and subliminal trigrams (CS) were respectively associated with four US categories varied along aggregate valence (100% positive, 80% positive, 20% positive, 0% positive). CS evaluations and preferences were recorded before and after conditioning. Bayesian analyses revealed US valence manipulations were likely to shift preferences, but not evaluations, of subliminal CS. Across supraliminal CS, Bayesian and frequentist analyses indicated US valence was significant and likely to shift preferences and evaluations. The present study demonstrates preferences may be influenced through subliminal conditioning even as evaluations are not.