Consecutive Scans Research Articles

Criteria for Verification and Replanning Based on the Adaptive Radiotherapy Protocol "Best for Adaptive Radiotherapy" in Head and Neck Cancer.

No clear criteria have yet been established to guide decision-making for patient selection and the optimal timing of adaptive radiotherapy (ART) based on image-guided radiotherapy (IGRT). We have developed a novel protocol—the Best for Adaptive Radiotherapy (B-ART) protocol—to guide patient selection for ART. The aim of the present study is to describe this protocol, to evaluate its validity in patients with head and neck (HN) cancer, and to identify the anatomical and clinical predictors of the need for replanning. We retrospectively evaluated 82 patients with HN cancer who underwent helical tomotherapy (HT) and subsequently required replanning due to soft tissue changes upon daily MVCT. Under the proposed criteria, patients with anatomical changes >3 mm on three to four consecutive scans are candidates for ART. We compared the volumes on the initial CT scan (iCT) and the replanning CT (rCT) scan for the clinical target volumes (CTV1, referring to primary tumor or tumor bed and CTV2, metastatic lymph nodes) and for the parotid glands (PG) and body contour (B-body). The patients were stratified by primary tumor localization, clinical stage, and treatment scheme. The main reasons for replanning were: (1) a planning target volume (PTV) outside the body contour (n = 70; 85.4%), (2) PG shrinkage (n = 69; 84.1%), (3) B-body deviations (n = 69; 84.1%), and (4) setup deviations (n = 40; 48.8%). The replanning decision was made, on average, during the fourth week of treatment (n = 47; 57.3%). The mean reductions in the size of the right and left PG volumes were 6.31 cc (20.9%) and 5.98 cc (20.5%), respectively (p < 0.001). The reduction in PG volume was ≥30% in 30 patients (36.6%). The volume reduction in all of the anatomical structures was statistically significant. Four variables—advanced stage disease (T3–T4), chemoradiation, increased weight loss, and oropharyngeal localization—were significantly associated with the need for ART. The B-ART protocol provides clear criteria to eliminate random errors, and to allow for an early response to relevant changes in target volumes.

A high-resolution 4D terrestrial laser scan dataset of the Kijkduin beach-dune system, The Netherlands

Sandy coasts form the interface between land and sea and their morphologies are highly dynamic. A combination of human and natural forcing results in morphologic changes affecting both nature values and coastal safety. Terrestrial laser scanning (TLS) is a technique enabling near-continuous monitoring of the changing morphology of a sandy beach-dune system with centimetre-order accuracy. In Kijkduin, The Netherlands, a laser scanner sampled one kilometre of coast at hourly intervals for about six months. This resulted in over 4,000 consecutive topographic scans of around one million points each, at decimetre-order point spacing. Analysis of the resulting dataset will offer new insights into the morphological behaviour of the beach-dune system at hourly to monthly time scales, ultimately increasing our fundamental scientific understanding of these complex geographic systems. It further provides the basis for developing novel algorithms to extract morphodynamic and geodetic information from this unique 4D spatiotemporal dataset. Finally, experiences from this TLS setup support the development of improved near-continuous 3D observation of both natural and anthropogenic scenes in general.

Accuracy Report on a Handheld 3D Ultrasound Scanner Prototype Based on a Standard Ultrasound Machine and a Spatial Pose Reading Sensor.

The aim of this study was to develop and evaluate a 3D ultrasound scanning method. The main requirements were the freehand architecture of the scanner and high accuracy of the reconstructions. A quantitative evaluation of a freehand 3D ultrasound scanner prototype was performed, comparing the ultrasonographic reconstructions with the CAD (computer-aided design) model of the scanned object, to determine the accuracy of the result. For six consecutive scans, the 3D ultrasonographic reconstructions were scaled and aligned with the model. The mean distance between the 3D objects ranged between 0.019 and 0.05 mm and the standard deviation between 0.287 mm and 0.565 mm. Despite some inherent limitations of our study, the quantitative evaluation of the 3D ultrasonographic reconstructions showed comparable results to other studies performed on smaller areas of the scanned objects, demonstrating the future potential of the developed prototype.

Scanning Distance Influence on the Intraoral Scanning Accuracy-An In Vitro Study.

Intraoral scanners (IOS) have reached a point where their impact in the dental office cannot be denied. The distance between the tip of the IOS and the preparation may have implications on the accuracy of the digital model. The objective of this study was to evaluate the differences in accuracy between digital impressions in the scenario of different scanning distances. Twenty consecutive scans were performed at five predetermined distances: 5 mm, 10 mm, 15 mm, 20 mm and 23 mm by a single operator. The scanning distance of 10 mm displayed the best accuracy with an overall trueness value of 23.05 μm and precision value of 4.2 μm. The drawn conclusion was that increased scanning distances can decrease the accuracy of a digital impression.

Effect of Cardiac Phase on Cardiac Output Index Derived from Dynamic CT Myocardial Perfusion Imaging.

Purpose: The aortic time-enhancement curve obtained from dynamic CT myocardial perfusion imaging can be used to derive the cardiac output (CO) index based on the indicator dilution principle. The objective of this study was to investigate the effect of cardiac phase at which CT myocardial perfusion imaging is triggered on the CO index measurement with this approach. Methods: Electrocardiogram (ECG) gated myocardial perfusion imaging was performed on farm pigs with consecutive cardiac axial scans using a large-coverage CT scanner (Revolution, GE Healthcare) after intravenous contrast administration. Multiple sets of dynamic contrast-enhanced (DCE) cardiac images were reconstructed retrospectively from 30% to 80% R-R intervals with a 5% phase increment. The time-enhancement curve sampled from above the aortic orifice in each DCE image set was fitted with a modified gamma variate function (MGVF). The fitted curve was then normalized to the baseline data point unaffected by the streak artifact emanating from the contrast solution in the right heart chamber. The Stewart–Hamilton equation was used to calculate the CO index based on the integral of the fitted normalized aortic curve, and the results were compared among different cardiac phases. Results: The aortic time-enhancement curves sampled at different cardiac phases were different from each other, especially in the baseline portion of the curve where the effect of streak artifact was prominent. After properly normalizing and denoising with a MGVF, the integrals of the aortic curve were minimally different among cardiac phases (0.228 ± 0.001 Hounsfield Unit × second). The corresponding mean CO index was 4.031 ± 0.028 L/min. There were no statistical differences in either the integral of the aortic curve or CO index among different cardiac phases (p > 0.05 for all phases).

No correlation between diffuse idiopathic skeletal hyperostosis and coronary artery disease on computed tomography using two different scoring systems.

An association between diffuse idiopathic skeletal hyperostosis (DISH) and a history of coronary artery disease (CAD) was previously reported. To investigate the association between DISH and CAD as assessed using the coronary artery calcification score (CACS) and the CAD-Reporting and Data System (CAD-RADS) score in patients with symptomatic chest pain. Consecutive cardiac CT scans performed before and after IV contrast administration were evaluated for CACS (Agatston method), CAD-RADS, and the presence of DISH. The association of DISH with the presence and extent of CACS/CAD-RADS scores was analyzed with and without adjustment for known atherosclerotic risk factors. The study cohort included 268 individuals (157 men, 111 women; median age = 54 years). DISH was present in 65 (24.3%) individuals. CACS was significantly higher in the DISH group compared to the non-DISH group in the univariate analysis (median CACS DISH = 2, range = 0-80.5 vs. median CACS non-DISH = 0, range = 0-11; P < 0.005) but this association did not persist on multivariate analysis. There was a positive trend toward higher CAD-RADS scores in the DISH group (P = 0.03) but after adjustment for age, male sex, and family history, this tendency was not significant. No independent association was found between the presence of DISH and CACS and CAD-RADS scores. Our findings suggest a more complex and possibly non-causal relationship between coronary artery disease and DISH.

Motion-corrected 23 Na MRI of the human brain using interleaved 1 H 3D navigator images.

To evaluate the feasibility of motion correction for sodium (23 Na) MRI based on interleaved acquired 3D proton (1 H) navigator images. A 3D radial density-adapted sequence for interleaved 23 Na/1 H MRI was implemented on a 7 Tesla whole-body MRI system. The 1 H data obtained during the 23 Na acquisition were used to reconstruct 140 navigator image volumes with a nominal spatial resolution of (2.5 mm)3 and a temporal resolution of 6 s. The motion information received from co-registration was then used to correct the 23 Na image dataset, which also had a nominal spatial resolution of (2.5 mm)3 . The approach was evaluated on six healthy volunteers, whose motion during the scans had different intensities and characteristics. Interleaved acquisition of two nuclei did not show any relevant influence on image quality (SNR of 13.0 for interleaved versus 13.2 for standard 23 Na MRI and 176.4 for interleaved versus 178.0 for standard 1 H MRI). The applied motion correction increased the consistency between two consecutive scans for all examined volunteers and improved the image quality for all kinds of motion. The SD of the differences ranged between 2.30% and 6.96% for the uncorrected and between 2.13% and 2.67% for the corrected images. The feasibility of interleaved acquired 1 H navigator images to be used for retrospective motion correction of 23 Na images was successfully demonstrated. The approach neither affected the 23 Na image quality nor elongated the scan time and can therefore be an important tool to improve the accuracy of quantitative 23 Na MRI.

Scalable Scene Flow From Point Clouds in the Real World

Autonomous vehicles operate in highly dynamic environments necessitating an accurate assessment of which aspects of a scene are moving and where they are moving to. A popular approach to 3D motion estimation, termed scene flow, is to employ 3D point cloud data from consecutive LiDAR scans, although such approaches have been limited by the small size of real-world, annotated LiDAR data. In this work, we introduce a new large-scale dataset for scene flow estimation derived from corresponding tracked 3D objects, which is <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\sim 1,000\times$</tex-math></inline-formula> larger than previous real-world datasets in terms of the number of annotated frames. We demonstrate how previous works were bounded by the amount of real LiDAR data available, suggesting that larger datasets are required to achieve state-of-the-art predictive performance. Furthermore, we show how previous heuristics such as down-sampling heavily degrade performance, motivating a new class of models that are tractable on the full point cloud. To address this issue, we introduce the FastFlow3D architecture which provides real time inference on the full point cloud. Additionally, we design human-interpretable metrics that better capture real world aspects by accounting for ego-motion and providing breakdowns per object type. We hope that this dataset may provide new opportunities for developing real world scene flow systems.

Prognostic utility of longitudinal quantification of PET myocardial blood flow early post heart transplantation

BackgroundMyocardial blood flow (MBF) quantification by Rubidium-82 positron emission tomography (PET) has shown promise for cardiac allograft vasculopathy (CAV) surveillance and risk stratification post heart transplantation. The objective was to determine the prognostic value of serial PET performed early post transplantation. Methods and ResultHeart transplant (HT) recipients at the University of Ottawa Heart Institute with 2 PET examinations (PET1 = baseline, PET2 = follow-up) within 6 years of transplant were included in the study. Evaluation of PET flow quantification included stress MBF, coronary vascular resistance (CVR), and myocardial flow reserve (MFR). The primary composite outcome was all-cause death, re-transplant, myocardial infarction, revascularization, allograft dysfunction, cardiac allograft vasculopathy (CAV), or heart failure hospitalization. A total of 121 patients were evaluated (79% male, mean age 56 ± 11 years) with consecutive scans performed at mean 1.4 ± 0.7 and 2.6 ± 1.0 years post HT for PET1 and PET2, respectively. Over a mean follow-up of 3.0 (IQR 1.8, 4.6) years, 26 (22%) patients developed the primary outcome: 1 death, 11 new or progressive angiographic CAV, 2 percutaneous coronary interventions, 12 allograft dysfunction. Unadjusted Cox analysis showed a significant reduction in event-free survival in patients with PET1 stress MBF < 2.1 (HR: 2.43, 95% CI 1.11-5.29 P = 0.047) and persistent abnormal PET1 to PET2 CVR > 76 (HR: 2.19, 95% CI 0.87-5.51 P = 0.045). There was no association between MFR and outcomes. ConclusionLow-stress MBF and persistent increased CVR on serial PET imaging early post HT are associated with adverse cardiovascular outcomes. Early post-transplant and longitudinal assessment by PET may identify at-risk patients for increased surveillance post HT.

A Robust, Fully Automatic Detection Method and Calculation Technique of Midline Shift in Intracranial Hemorrhage and Its Clinical Application.

A midline shift (MLS) is an important clinical indicator for intracranial hemorrhage. In this study, we proposed a robust, fully automatic neural network-based model for the detection of MLS and compared it with MLSs drawn by clinicians; we also evaluated the clinical applications of the fully automatic model. We recruited 300 consecutive non-contrast CT scans consisting of 7269 slices in this study. Six different types of hemorrhage were included. The automatic detection of MLS was based on modified Keypoint R-CNN with keypoint detection followed by training on the ResNet-FPN-50 backbone. The results were further compared with manually drawn outcomes and manually defined keypoint calculations. Clinical parameters, including Glasgow coma scale (GCS), Glasgow outcome scale (GOS), and 30-day mortality, were also analyzed. The mean absolute error for the automatic detection of an MLS was 0.936 mm compared with the ground truth. The interclass correlation was 0.9899 between the automatic method and MLS drawn by different clinicians. There was high sensitivity and specificity in the detection of MLS at 2 mm (91.7%, 80%) and 5 mm (87.5%, 96.7%) and MLSs greater than 10 mm (85.7%, 97.7%). MLS showed a significant association with initial poor GCS and GCS on day 7 and was inversely correlated with poor 30-day GOS (p < 0.001). In conclusion, automatic detection and calculation of MLS can provide an accurate, robust method for MLS measurement that is clinically comparable to the manually drawn method.

Pharmacological and biopharmaceutical studies of paracetamol and N-acetyl-D-glucosamine combination as an analgetic drug

The aim. Experimental selection of paracetamol and N-acetyl-D-glucosamine (NAG) ratio and combined dose and bioavailability study of the pharmaceutical composition and active pharmaceutical ingredient (API) in the model of cell biomembranes.&#x0D; Materials and methods. The following substances were used: paracetamol, Actimask Acetaminophen (gelatin-coated paracetamol), and NAG. For pharmacological studies were used laboratory rats, which received test objects intragastrically. The study was performed on the model of inflammatory hyperalgesia according to the Randall-Selitto method. The values of pain threshold before and after pathology induction were established. For biopharmaceutical studies, dry L-α-dimyristoylphosphatidylcholine purchased from Avanti Polar Lipids (purity 99.9 %) was used, hydrated with an appropriate amount of double-distilled water to obtain 70 % wt./wt. water dispersion. Differential scanning calorimetry (DSC) studies were performed employing DSC 1 microcalorimeter (Mettler Toledo). Pharmaceuticals investigated were placed on the bottom of a crucible, then a proper amount of the lipid membrane was added and this moment was taken as the time reference point (t=0). Then a crucible was sealed with a lid and a sample was undergone consecutive temperature scans, heating from 0 to 35 ºC at a scanning rate of 2 ºC/min. The procedure was repeated until no more changes in DSC profiles were observed, i.e. system equilibrium was reached.&#x0D; Results. It has been experimentally proven that the combination of paracetamol and NAG in the ratio of 4:1 showed better analgesic efficacy. The dose of active ingredients was determined to be 50 mg/kg by the sum of API. The active interaction of paracetamol with the bilayer of biomembranes was established and it was determined that Actimask has a worse rate of penetration into the membrane due to the coating of paracetamol with a gelatin shell. NAG didn't significantly affect the rate of penetration of Actimask through the bilayer of membranes, but the auxiliary components of the tablet mixture significantly improved the rate and completeness of penetration of paracetamol through the bilayer of biomembranes.&#x0D; Conclusions. The study found the most effective ratio between paracetamol and NAG in the composition, which is 4:1. The next step was to determine the dose of API, which is 50 mg/kg of the sum of active substances. It has been determined that paracetamol has good permeability through the bilayer of biomembranes, and the tablet mass significantly improves the permeability of paracetamol

Cortical step sign in spinal clearance on trauma computed tomography – Indicator of acute thoracolumbar compression fracture

Introduction Differentiating an acute from chronic compression fracture of the thoracolumbar (TL) spine can pose a dilemma for radiologists interpreting spinal imaging following trauma. Mild wedging of the vertebrae can be due to spondylosis or osteoporosis, whilst acute simple compression fractures may not always be associated with loss of body height. In this observational study, we hypothesize that the presence of a vertebral body cortical step is a reliable sign of an acute compression fracture on Computed Tomography (CT) scans. Methods In a retrospective review of thoracolumbar CT scans following trauma, two observers analysed for the presence of a cortical step at the anterior or posterior vertebral body cortex, fracture morphology and associated injuries. A ‘cortical step’ is defined as a break of hyperdense cortex on CT scans, intervening non-sclerosed trabecular bone, and sharp overlap of the underlying cortex. MRI of the spine was used as gold standard. Results 187 consecutive CT scans over 2 years were assessed. Sensitivity, specificity and accuracy of cortical step sign were 100%, 90.2% and 97% in diagnosing an acute thoracolumbar compression fracture, respectively. The interobserver reliability was high (kappa = 0.97). False positive cortical step was seen in Kummel’s disease and large Schmorl’s nodule. Conclusion Our results demonstrate high sensitivity and specificity of ‘cortical step sign’ in diagnosing acute vertebral body compression fractures of TL spine on CT scans in patients with trauma. This sign can be useful to radiologists for safe clearance of the thoracolumbar spine following trauma, helping distinguish acute trauma from chronic causes of vertebral body height loss.

Influence of antidepressant use on 123I-MIBG heart and lung uptakes in the diagnosis of Lewy body disease

ObjectiveThe clinical significance of decreased physiological lung uptake of 123I-metaiodobenzylguanidine (MIBG) has not been well investigated. This study aimed to elucidate the association between a decrease in lung MIBG uptake with antidepressant intake and the myocardial MIBG uptake in patients who were clinically diagnosed with Lewy body disease (LBD) and patients who were diagnosed as not having LBD.MethodsWe retrospectively reviewed the heart and lung uptakes on 167 consecutive MIBG scans, antidepressant status, and clinical diagnosis of LBD. The images were visually classified into two groups: decreased lung uptake and preserved lung uptake. A semi-quantitative analysis was performed using the heart-to-mediastinum ratio (H/M), lung-to-mediastinum ratio (L/M), and myocardial washout rate (WR).ResultsAll 17 patients with decreased lung uptake were on treated with antidepressants, while none of the 150 patients with preserved lung uptake were treated with any antidepressants. Of the 17 patients with decreased lung uptake, 6 patients were clinically diagnosed as LBD and other 11 were clinically diagnosed as non-LBD. There was not significant difference in early H/M, delayed H/M, and myocardial WR between the 11 non-LBD patients with decreased lung uptake and 83 non-LBD patients with preserved lung uptake (2.87 ± 0.69 vs. 2.89 ± 0.44, 3.09 ± 0.48 vs. 2.98 ± 0.59, and 21.8 ± 11.3% vs. 21.1 ± 12.5%, respectively). Moreover, in LBD patients, there were no significant differences in those values between six patients with decreased lung uptake and 67 patients with preserved lung uptake (1.68 ± 0.32 vs. 1.73 ± 0.42, 1.34 ± 0.21 vs. 1.54 ± 0.57, 46.2 ± 22.8% vs. 42.8 ± 21.3%, respectively).ConclusionsAntidepressants probably blocked MIBG uptake in the lungs, and a decreased lung uptake was not significantly associated with heart uptake. A remarkable decrease in lung uptake can be a signal to check a patient’s medication status.

To wallow or nurse: Sows housed outdoors have distinctive approaches to thermoregulation in gestation and lactation

The objective of this study was to determine which behaviours may be important for thermoregulation when gestating and lactating sows are housed outdoors in Québec, Canada during the summer. We kept six groups of four Yorkshire-Landrace sows, from July to September 2018, in outdoor paddocks outfitted with a wallow, shade structure, individual farrowing huts and a pasture plot. From week 15 of gestation to week 3 of lactation, we recorded the location of each sow during five daily observation periods (5 days/week), each consisting of 15 consecutive scan samples at 1-min intervals. Simultaneously, we monitored environmental conditions with temperature and humidity loggers. We analyzed the impact of day relative to farrowing and the temperature humidity index (THI) on the percent of daily observations in the wallow, shade, and farrowing hut with polynomial regression models, with day and THI as covariates and the sow as the subject of repeated measures. In gestation, sows were mainly observed in the wallow (40 ± 2% of observations), while in lactation, sows were primarily observed in the huts (54 ± 17%) and were only in the wallow for 6 ± 4% of observations. THI was associated with the percent of observations in the wallow (P < 0.001) during both gestation and lactation. Day also had a significant association with wallowing; gestating sows were observed in the wallow more often as farrowing drew nearer (P = 0.029), while lactating sows were observed in the wallow more often as the number of days since farrowing increased (P < 0.0001). In gestation, THI was negatively associated with the percent of observations in the farrowing hut (P < 0.0001); however, in lactation, only day predicted the use of the farrowing hut in a non-linear fashion (P < 0.01). In gestation, the sows’ wallow use was associated with THI. However, lactating sows exhibited low wallow usage even when THI was high, instead prioritizing the farrowing hut, especially in the first few days after farrowing. The differences between gestation and lactation in the timing and frequency of the sows’ use of the wallow and farrowing hut may indicate a conflict of motivation between thermoregulatory and maternal behaviours when sows are nursing their piglets.

Sphenomaxillary Plate: Landmine or Landmark?

Sphenomaxillary plate (SMP) is an inconstant bony septum separating the maxillary and sphenoid sinuses. Neglecting the SMP during endoscopic sinus surgery may lead to mistaking the sphenoid sinus for the posterior ethmoid cell with potentially dangerous consequences. However, its proper identification may allow planned sphenoidotomy through the posterior wall of the maxillary sinus or to enlarge transnasal or transethmoidal sphenoidotomy. The aim of this study was to evaluate (1) the incidence and morphology of the SMP and (2) possibility and safety of trans-SMP sphenoidotomy. In the radioanatomical part of the study multiplanar reconstruction analysis of 117 consecutive paranasal sinus CT scans was conducted. In the clinical part of the research, trans-SMP sphenoidotomy was performed in consecutive patients operated on for chronic rhinosinusitis. The SMP was found in 25% of sides (34% of patients). It was constantly located superolateral to the posterior insertion of the middle turbinate. The SMP formed the prominence on the posterior wall of the maxillary sinus that could be identified using volume rendering in 11% of sides (19% of patients). 30 trans-SMP sphenoidotomies were performed in 18 patients. The SMP prominence was identified in 15 sides. In absence of SMP prominence, identification of the SMP was still possible using other anatomical landmarks. The SMP is present in over 30% Caucasian subjects at least on one side. Sphenoidotomy through the SMP is feasible and safe.

Inter-Observer Agreement between Low-Dose and Standard-Dose CT with Soft and Sharp Convolution Kernels in COVID-19 Pneumonia.

Computed tomography (CT) has been an essential diagnostic tool during the COVID-19 pandemic. The study aimed to develop an optimal CT protocol in terms of safety and reliability. For this, we assessed the inter-observer agreement between CT and low-dose CT (LDCT) with soft and sharp kernels using a semi-quantitative severity scale in a prospective study (Moscow, Russia). Two consecutive scans with CT and LDCT were performed in a single visit. Reading was performed by ten radiologists with 3–25 years’ experience. The study included 230 patients, and statistical analysis showed LDCT with a sharp kernel as the most reliable protocol (percentage agreement 74.35 ± 43.77%), but its advantage was marginal. There was no significant correlation between radiologists’ experience and average percentage agreement for all four evaluated protocols. Regarding the radiation exposure, CTDIvol was 3.6 ± 0.64 times lower for LDCT. In conclusion, CT and LDCT with soft and sharp reconstructions are equally reliable for COVID-19 reporting using the “CT 0-4” scale. The LDCT protocol allows for a significant decrease in radiation exposure but may be restricted by body mass index.

Defining disease status in gastroenteropancreatic neuroendocrine tumors: Choi-criteria or RECIST?

Adequate monitoring of changes in tumor load is fundamental for the assessment of the course of disease and response to treatment. There is an ongoing debate on the utility of RECIST v1.1 in gastroenteropancreatic neuroendocrine tumors (GEP-NETs). In this retrospective real-life cohort study, Choi-criteria were compared with RECIST v1.1. The agreement between both criteria and the association with survival endpoints were evaluated. Seventy-five patients were included with a median follow-up of 35months (range 8-53). Median progression-free survival (mPFS) according to RECIST v1.1 was 15months (range 2-50) compared to 14months (range 2-50) in Choi. According to RECIST, 33 (44%) patients were classified as having stable disease (SD), 40 (53%) as progressive disease (PD) and two (3%) patients as partial response (PR), compared to 9 (12%) patients classified as SD, 50 (67%) as PD and 16 (21%) as PR according to Choi-criteria. Overall concordance between the criteria was moderate (Cohen's Kappa = 0.408, p < 0.001) and agreement varied between 57 and 69% at each consecutive scan (p < 0.001). Survival analysis showed significant differences in overall survival (OS) for RECIST v1.1 categories PD and non-PD (log-rank p = 0.02), however, in Choi no significant differences in OS were found (p = 0.27). RECIST v1.1 had a better clinical utility and prognostic value compared to Choi-criteria. Still, RECIST were also not sufficient to adequately predict OS. This outlines the need for new tools that provides accurate information on the disease course and treatment response to support precise prognostication in patients with GEP-NETs.

Development of a deep learning-based method to diagnose pulmonary ground-glass nodules by sequential computed tomography imaging.

BackgroundEarly identification of the malignant propensity of pulmonary ground‐glass nodules (GGNs) can relieve the pressure from tracking lesions and personalized treatment adaptation. The purpose of this study was to develop a deep learning‐based method using sequential computed tomography (CT) imaging for diagnosing pulmonary GGNs.MethodsThis diagnostic study retrospectively enrolled 762 patients with GGNs from West China Hospital of Sichuan University between July 2009 and March 2019. All patients underwent surgical resection and at least two consecutive time‐point CT scans. We developed a deep learning‐based method to identify GGNs using sequential CT imaging on a training set consisting of 1524 CT sections from 508 patients and then evaluated 256 patients in the testing set. Afterwards, an observer study was conducted to compare the diagnostic performance between the deep learning model and two trained radiologists in the testing set. We further performed stratified analysis to further relieve the impact of histological types, nodule size, time interval between two CTs, and the component of GGNs. Receiver operating characteristic (ROC) analysis was used to assess the performance of all models.ResultsThe deep learning model that used integrated DL‐features from initial and follow‐up CT images yielded the best diagnostic performance, with an area under the curve of 0.841. The observer study showed that the accuracies for the deep learning model, junior radiologist, and senior radiologist were 77.17%, 66.89%, and 77.03%, respectively. Stratified analyses showed that the deep learning model and radiologists exhibited higher performance in the subgroup of nodule sizes larger than 10 mm. With a longer time interval between two CTs, the deep learning model yielded higher diagnostic accuracy, but no general rules were yielded for radiologists. Different densities of components did not affect the performance of the deep learning model. In contrast, the radiologists were affected by the nodule component.ConclusionsDeep learning can achieve diagnostic performance on par with or better than radiologists in identifying pulmonary GGNs.

Application of Continuous Positive Airway Pressure for Thoracic Respiratory Motion Management: An Assessment in a Magnetic Resonance Imaging–Guided Radiation Therapy Environment

PurposePatient tolerability of magnetic resonance (MR)–guided radiation treatment delivery is limited by the need for repeated deep inspiratory breath holds (DIBHs). This volunteer study assessed the feasibility of continuous positive airway pressure (CPAP) with and without DIBH for respiratory motion management during radiation treatment with an MR-linear accelerator (MR-linac). Methods and MaterialsMR imaging safety was first addressed by placing the CPAP device in an MR-safe closet and configuring a tube circuit via waveguide to the magnet bore. Reproducibility and linearity of the final configuration were assessed. Six healthy volunteers underwent thoracic imaging in a 0.35T MR-linac, with one free breathing (FB) and 2 DIBH acquisitions being obtained at 5 pressures from 0 to 15 cm-H2O. Lung and heart volumes and positions were recorded; repeatability was assessed by comparing 2 consecutive DIBH scans. Blinded reviewers graded images for motion artifact using a 3-point grading scale. Participants completed comfort and perception surveys before and after imaging sessions. ResultsCompared with FB alone, FB-10, FB-12, and FB-15 cm H2O significantly increased lung volumes (+23%, +34%, +44%; all P <.05) and inferiorly displaced the heart (0.86 cm, 0.96 cm, 1.18 cm; all P < . 05). Lung volumes were significantly greater with DIBH-0 cm H2O compared with FB-15 cm H2O (+105% vs +44%, P = .01), and DIBH-15 cm H2O yielded additional volume increase (+131% vs +105%, P = .01). Adding CPAP to DIBH decreased lung volume differences between consecutive breath holds (correlation coefficient 0.97 at 15 cm H2O vs 0.00 at 0 cm H2O). The addition of 15 cm H2O CPAP reduced artifact scores (P = .03) compared with FB; all DIBH images (0-15 cm H2O) had less artifact (P < .01). ConclusionsThis work demonstrates the feasibility of integrating CPAP in an MR-linac environment in healthy volunteers. Extending this work to a larger patient cohort is warranted to further establish the role of CPAP as an alternative and concurrent approach to DIBH in MR-guided radiation therapy.

A Possible White Matter Compensating Mechanism in the Brain of Relatives of People Affected by Psychosis Inferred from Repeated Long-Term DTI Scans.

An existing model suggests that some brain features of relatives of people affected by psychosis can be distinguished from both the probands and a control group. Such findings can be interpreted as representing a compensating mechanism. We studied white matter features using diffusion tensor imaging in a cohort of 82 people affected by psychosis, 122 of their first-degree relatives, and 89 control subjects that were scanned between two to three times with an interval of approximately 3 years between consecutive scans. We measured both fractional anisotropy and other standard diffusivity measures such as axial diffusivity. Additionally, we calculated standard connectivity measures such as path length based on probabilistic or deterministic tractography. Finally, by averaging the values of the different measures over the two or three consecutive scans, we studied epoch-averagely the difference between these three groups. For several tracts and several connectivity measures, the relatives showed distinct features from both the probands and the control groups. In those cases, the relatives did not necessarily score between the probands and the control group. An aggregate analysis in the form of a group-dependent score for the different modes of the analysis (e.g., for fractional anisotropy) supported this observation. We interpret these results as evidence supporting a compensation mechanism in the brain of relatives that may be related to resilience that some of them exhibit in the face of the genetic risk they have for being affected by psychosis.