Background: Treatment monitoring in metastatic colorectal cancer (mCRC) relies on imaging to evaluate the tumour burden. Response Evaluation Criteria in Solid Tumors provide a framework on... Show moreBackground: Treatment monitoring in metastatic colorectal cancer (mCRC) relies on imaging to evaluate the tumour burden. Response Evaluation Criteria in Solid Tumors provide a framework on reporting and interpretation of imaging findings yet offer no guidance on a standardised imaging protocol tailored to patients with mCRC. Imaging protocol hetero-geneity remains a challenge for the reproducibility of conventional imaging end-points and is an obstacle for research on novel imaging end-points. Patients and methods: Acknowledging the recently highlighted potential of radiomics and arti-ficial intelligence tools as decision support for patient care in mCRC, a multidisciplinary, international and expert panel of imaging specialists was formed to find consensus on mCRC imaging protocols using the Delphi method. Results: Under the guidance of the European Organisation for Research and Treatment of Cancer (EORTC) Imaging and Gastrointestinal Tract Cancer Groups, the European Society of Oncologic Imaging (ESOI) and the European Society of Gastrointestinal and Abdominal Radiology (ESGAR), the EORTC-ESOI-ESGAR core imaging protocol was identified. Conclusion: This consensus protocol attempts to promote standardisation and to diminish variations in patient preparation, scan acquisition and scan reconstruction. We anticipate that this standardisation will increase reproducibility of radiomics and artificial intelligence studies and serve as a catalyst for future research on imaging end-points. For ongoing and future mCRC trials, we encourage principal investigators to support the dissemination of these im-aging standards across recruiting centres. (c) 2022 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). Show less
A chronic subdural hematoma (CSDH) is an intracranial bleeding between the outer two meninges of the brain due to rupture of cerebral veins or an inflammatory response in the subdural space.... Show moreA chronic subdural hematoma (CSDH) is an intracranial bleeding between the outer two meninges of the brain due to rupture of cerebral veins or an inflammatory response in the subdural space. Elderly patients using anti-thrombotic therapy are at higher risk for hematoma development. A rise in CSDH incidence is expected because of the aging population and increase in anti-thrombotic therapy use due to cardiovascular disease. To date, no treatment guideline exists regarding optimal CSDH treatment. Surgery with subdural drainage is the mainstay treatment. However, due to relevant surgical complications, a recurrence risk up to 30% and increased mortality in this vulnerable patient population, corticosteroid therapy is being administered as an alternative or adjuvant treatment modality.In this thesis we have shown that surgical treatment results in significantly better treatment outcome than medicinal approach by the corticosteroid dexamethasone in a retrospective study (chapter 2) as well as in a randomized controlled trial (chapter 3 and 4). Furthermore, we revealed radiological markers that are of prognostic value to predict treatment outcome after surgery (chapter 5) and dexamethasone therapy (chapter 6) in symptomatic CSDH patients. Show less
Purpose To characterize the prevalence of missed pancreatic masses and pancreatic ductal adenocarcinoma (PDAC)-related findings on CT and MRI between pre-diagnostic patients and healthy individuals... Show morePurpose To characterize the prevalence of missed pancreatic masses and pancreatic ductal adenocarcinoma (PDAC)-related findings on CT and MRI between pre-diagnostic patients and healthy individuals.Materials and methods Patients diagnosed with PDAC (2010-2016) were retrospectively reviewed for abdominal CT- or MRI-examinations 1 month-3 years prior to their diagnosis, and subsequently matched to controls in a 1:4 ratio. Two blinded radiologists scored each imaging exam on the presence of a pancreatic mass and secondary features of PDAC. Additionally, original radiology reports were graded based on the revised RADPEER criteria.Results The cohort of 595 PDAC patients contained 60 patients with a pre-diagnostic CT and 27 with an MRI. A pancreatic mass was suspected in hindsight on CT in 51.7% and 50% of cases and in 1.3% and 0.9% of controls by reviewer 1 (p < .001) and reviewer 2 (p < .001), respectively. On MRI, a mass was suspected in 70.4% and 55.6% of cases and 2.9% and 0% of the controls by reviewer 1 (p < .001) and reviewer 2 (p < .001), respectively. Pancreatic duct dilation, duct interruption, focal atrophy, and features of acute pancreatitis is strongly associated with PDAC (p < .001). In cases, a RADPEER-score of 2 or 3 was assigned to 56.3% of the CT-reports and 71.4% of MRI-reports.Conclusion Radiological features as pancreatic duct dilation and interruption, and focal atrophy are common first signs of PDAC and are often missed or unrecognized. Further investigation with dedicated pancreas imaging is warranted in patients with PDAC-related radiological findings. Show less
With the rising global health burden of aortic valve diseas and growing awareness of the consequences of severe AS, more patients are expected to be referred for aortic valve replacement.... Show moreWith the rising global health burden of aortic valve diseas and growing awareness of the consequences of severe AS, more patients are expected to be referred for aortic valve replacement. Multimodality imaging remains of paramount importance for proper patient selection, determining optimal timing of intervention and in the evaluation of therapy. For risk stratification and defining optimal timing of intervention, conventional and advanced echocardiography and computed tomography are crucial. Conventional echocardiography can be used to assess extra-aortic valvular cardiac damage in patients with severe AS. On top of conventional echocardiography, advanced echocardiography can provide additional insights. Left ventricular GLS by speckle tracking echocardiography has been suggested as a more sensitive marker of LV systolic dysfunction. LV GLS may be of help to define more optimal timing of intervention in asymptomatic patients with severe AS. Also, LV GLS may be of incremental value on top of conventional echocardiographic parameters for the assessment of cardiac injury. For the evaluation of prothesis function and durability after aortic valve implantation and detection of possible (late) complications, echocardiography is the mainstay imaging modality. Also, both conventional and advanced echocardiography can provide additional information on the effects of therapy on LV function and remodelling. Show less
Objectives Thrombus computed tomography (CT) characteristics might be used to assess histopathologic thrombus composition in patients treated with endovascular thrombectomy (EVT) for acute ischemic... Show moreObjectives Thrombus computed tomography (CT) characteristics might be used to assess histopathologic thrombus composition in patients treated with endovascular thrombectomy (EVT) for acute ischemic stroke (AIS). We aimed to assess the variability in thrombus composition that could be predicted with combined thrombus CT characteristics. Methods Thrombi of patients enrolled in the MR CLEAN Registry between March 2014 and June 2016 were histologically analyzed with hematoxylin-eosin staining and quantified for percentages of red blood cells (RBCs) and fibrin/platelets. We estimated the association between general qualitative characteristics (hyperdense artery sign [HAS], occlusion location, clot burden score [CBS]) and thrombus composition with linear regression, and quantified RBC variability that could be explained with individual and combined characteristics with R-2. For patients with available thin-slice (<= 2.5 mm) imaging, we performed similar analyses for general and quantitative characteristics (HAS, occlusion location, CBS, [relative] thrombus density, thrombus length, perviousness, distance from ICA-terminus). Results In 332 included patients, the presence of HAS (a beta 7.8 [95% CI 3.9-11.7]) and shift towards a more proximal occlusion location (a beta 3.9 [95% CI 0.6-7.1]) were independently associated with increased RBC and decreased fibrin/platelet content. With general characteristics, 12% of RBC variability could be explained; HAS was the strongest predictor. In 94 patients with available thin-slice imaging, 30% of RBC variability could be explained; thrombus density and thrombus length were the strongest predictors. Conclusions Quantitative thrombus CT characteristics on thin-slice admission CT improve prediction of thrombus composition and might be used to further guide clinical decision-making in patients treated with EVT for AIS in the future. Show less
Zanden, V. van der; Soolingen, N.J. van; Viddeleer, A.R.; Trum, J.W.; Amant, F.; Mourits, M.J.E.; ... ; Munster, B.C. van 2022
Objective: To assess the association between loss of lumbar skeletal muscle mass and density during neoadjuvant chemotherapy (NACT) and postoperative complications after interval cytoreductive... Show moreObjective: To assess the association between loss of lumbar skeletal muscle mass and density during neoadjuvant chemotherapy (NACT) and postoperative complications after interval cytoreductive surgery (CRS) in older patients with ovarian cancer. Materials and methods: This multicenter, retrospective cohort study included patients aged 70 years and older with primary advanced stage ovarian cancer (International Federation of Gynecology and Ob-stetrics stage III-IV), treated with NACT and interval CRS. Skeletal muscle mass and density were retrospectively assessed using Skeletal Muscle Index (SMI) and Muscle Attenuation (MA) on routinely made Computed Tomography scans before and after NACT. Loss of skeletal muscle mass or density was defined as >2% decrease per 100 days in SMI or MA during NACT. Results: In total, 111 patients were included. Loss of skeletal muscle density during NACT was associated with developing any postoperative complication <30 days after interval CRS both in univariable (Odds Ratio (OR) 3.69; 95% Confidence Interval (CI) 1.57-8.68) and in multivariable analysis adjusted for functional impairment and WHO performance status (OR 3.62; 95%CI 1.27-10.25). Loss of skeletal muscle density was also associated with infectious complications (OR 3.67; 95%CI 1.42-9.52) and unintended discontinuation of adjuvant chemotherapy (OR 5.07; 95%CI 1.41-18.19). Unlike loss of skeletal muscle density, loss of skeletal muscle mass showed no association with postoperative outcomes. Conclusion: In older patients with ovarian cancer, loss of skeletal muscle density during NACT is asso-ciated with worse postoperative outcomes. These results could add to perioperative risk assessment, guiding the decision to undergo surgery or the need for perioperative interventions. Show less
Rosendael, A.R. van; Smit, J.M.; El'Mahdiui, M.; Rosendael, P.J. van; Leung, M.; Delgado, V.; Bax, J.J. 2022
Aims: Left atrial (LA) volume and LA epicardial fat are both substrates for atrial fibrillation (AF), but may relate with AF at different (early vs. late) stages in the AF disease process. We... Show moreAims: Left atrial (LA) volume and LA epicardial fat are both substrates for atrial fibrillation (AF), but may relate with AF at different (early vs. late) stages in the AF disease process. We evaluated associations between LA epicardial fat and LA volume in patients with sinus rhythm (SR), paroxysmal AF (PAF), and persistent/permanent AF. Methods and results: In total, 300 patients (100 with SR, 100 with PAF, and 100 with persistent/permanent AF) who underwent cardiac computed tomography angiography (CTA) were included. The epicardial fat mass posterior to the LA and the LA volume were quantified from CTA and compared between patients with SR, PAF, and persistent/permanent AF. Furthermore, four groups were created by classifying LA epicardial fat and LA volume into large or small according to their median. The mean age of the population was 58.9 +/- 10.5 years and 69.7% was male. Left atrial epicardial fat mass was larger in patients with PAF compared with SR, but did not further increase from PAF to persistent/permanent AF. Left atrial volume increased significantly from SR to PAF and to persistent/permanent AF. Left atrial epicardial fat and LA volume were both concordantly large or small in 184 (61%) patients, and discordant in 116 (39%). When both were small, 65.2% of the patients had SR, 23.9% PAF, and 10.9% persistent/permanent AF. When the LA epicardial fat mass was large and the LA volume small (compared with both being small), patients were significantly more often in PAF (55.2 vs. 23.9, P < 0.05), less frequently in SR (32.8% vs. 65.2%, P < 0.05) but showed comparable rates of persistent/permanent AF (12.0% vs. 10.9%, P < 0.05). When the LA volume was large, most patients had persistent/permanent AF. Conclusion: Left atrial epicardial fat mass was larger in PAF vs. SR, possibly indicating a marker of early disease, while large LA volumes were associated with a high prevalence of persistent/permanent AF. Elevated LA epicardial fat mass without large LA volume may reflect the early AF disease process. Show less
The aim of this thesis is to determine diagnostic performance of machine learning in differentiating between atypical cartilaginous tumor (ACT) and high-grade chondrosarcoma (CS) based on radiomic... Show moreThe aim of this thesis is to determine diagnostic performance of machine learning in differentiating between atypical cartilaginous tumor (ACT) and high-grade chondrosarcoma (CS) based on radiomic features derived from magnetic resonance imaging (MRI) and computed tomography (CT). In chapter 2, the concept of radiomics of musculoskeletal sarcomas is introduced and a systematic review on radiomic feature reproducibility and validation strategies is conducted. In chapter 3, a preliminary study is performed to investigate the performance of MRI radiomics-based machine learning in discriminating ACT from high-grade CS, using a single-center cohort, in comparison with an expert radiologist. In chapter 4, the influence of interobserver segmentation variability on the reproducibility of CT and MRI radiomic features of cartilaginous bone tumors is assessed. In chapter 5, the performance of CT radiomics-based machine learning in discriminating ACT from high-grade CS of long bones is determined and validated using independent data from a multicenter cohort, compared to an expert radiologist. In chapter 6, the performance of MRI radiomics-based machine learning in differentiating between ACT and grade II CS of long bones is determined and validated using independent data from a multicenter cohort, in comparison with an expert radiologist. Finally, in chapter 7, the main results and implications of this thesis are summarized and discussed. Show less
Venous thromboembolism (VTE) encompasses pulmonary embolism (PE) and deep vein thrombosis (DVT). DVT most commonly occurs in the deep veins of the lower extremity but can also occur in the veins of... Show moreVenous thromboembolism (VTE) encompasses pulmonary embolism (PE) and deep vein thrombosis (DVT). DVT most commonly occurs in the deep veins of the lower extremity but can also occur in the veins of upper extremity, abdomen and cerebrum. As symptoms of VTE are nonspecific, the diagnosis of VTE is based on diagnostic tests, including clinical decision rules (CDR), D-dimer tests and imaging. Although the diagnostic management of VTE has greatly advanced in recent years with the introduction of novel CDRs and high-sensitive D-dimer tests, the diagnosis may still be challenging in certain settings. The latter is mainly caused by the indirect way of thrombus visualisation by current imaging tests, such as by showing incompressibility with compression ultrasonography (CUS) or a filling defect on contrast-enhanced computed tomography (CT) or magnetic resonance imaging (MRI).This thesis focuses on challenging settings for diagnosing VTE, including suspected recurrent ipsilateral DVT, upper extremity DVT, cerebral vein thrombosis and portal vein thrombosis. We studied a novel imaging technique called Magnetic Resonance Non-Contrast Thrombus Imaging (MR-NCTI) and its application in these different VTE settings. Show less
Whitehouse, D.P.; Monteiro, M.; Czeiter, E.; Vande Vyvere, T.; Valerio, F.; Ye, Z.; ... ; CENTER-TBI Participants Investigat 2022
Background We aimed to understand the relationship between serum biomarker concentration and lesion type and volume found on computed tomography (CT) following all severities of TBI.Methods... Show moreBackground We aimed to understand the relationship between serum biomarker concentration and lesion type and volume found on computed tomography (CT) following all severities of TBI.Methods Concentrations of six serum biomarkers (GFAP, NFL, NSE, S100B, t-tau and UCH-L1) were measured in samples obtained <24 hours post-injury from 2869 patients with all severities of TBI, enrolled in the CENTER-TBI prospective cohort study (NCT02210221). Imaging phenotypes were defined as intraparenchymal haemorrhage (IPH), oedema, subdural haematoma (SDH), extradural haematoma (EDH), traumatic subarachnoid haemorrhage (tSAH), diffuse axonal injury (DAI), and intraventricular haemorrhage (IVH). Multivariable polynomial regression was performed to examine the association between biomarker levels and both distinct lesion types and lesion volumes. Hierarchical clustering was used to explore imaging phenotypes; and principal component analysis and k-means clustering of acute biomarker concentrations to explore patterns of biomarker clustering.Findings 2869 patient were included, 68% (n=1946) male with a median age of 49 years (range 2-96). All severities of TBI (mild, moderate and severe) were included for analysis with majority (n=1946, 68%) having a mild injury (GCS 13-15). Patients with severe diffuse injury (Marshall III/IV) showed significantly higher levels of all measured biomarkers, with the exception of NFL, than patients with focal mass lesions (Marshall grades V/VI). Patients with either DAI+IVH or SDH+IPH+tSAH, had significantly higher biomarker concentrations than patients with EDH. Higher biomarker concentrations were associated with greater volume of IPH (GFAP, S100B, t-tau;adj r2 range:0.48-0.49; p<0.05), oedema (GFAP, NFL, NSE, t-tau, UCH-L1;adj r2 range:0. 44-0.44; p<0.01), IVH (S100B;adj r2 range:0.48-0.49; p<0.05), Unsupervised k-means biomarker clustering revealed two clusters explaining 83.9% of variance, with phenotyping characteristics related to clinical injury severity.Interpretation Interpretation: Biomarker concentration within 24 hours of TBI is primarily related to severity of injury and intracranial disease burden, rather than pathoanatomical type of injury. Copyright (C) 2021 The Author(s). Published by Elsevier B.V. Show less
Zanden, V. van der; Soolingen, N.J. van; Viddeleer, A.R.; Trum, J.W.; Amant, F.; Mourits, M.J.E.; ... ; Souwer, E.T.D. 2021
Objective. To determine the predictive value of lumbar skeletal muscle mass and density for postoperative outcomes in older women with advanced stage ovarian cancer.Methods. A multicenter,... Show moreObjective. To determine the predictive value of lumbar skeletal muscle mass and density for postoperative outcomes in older women with advanced stage ovarian cancer.Methods. A multicenter, retrospective cohort study was performed in women >= 70 years old receiving surgery for primary, advanced stage ovarian cancer. Skeletal muscle mass and density were assessed in axial CT slices on level L3. Low skeletal muscle mass was defined as skeletal muscle index < 38.50 cm(2)/m(2). Low skeletal muscle density was defined as one standard deviation below the mean (muscle attenuation < 22.55 Hounsfield Units). The primary outcome was any postoperative complication <-30 days after surgery. Secondary outcomes included severe complications, infections, delirium, prolonged hospital stay, discharge destination, discontinua-tion of adjuvant chemotherapy and mortality.Results. In analysis of 213 patients, preoperative low skeletal muscle density was associated with postopera-tive complications <-30 days after surgery (Odds Ratio (OR) 2.83; 95% Confidence Interval (CI) 1.41-5.67), severe complications (OR 3.01; 95%CI 1.09-8.33), infectious complications (OR 2.79; 95%CI 1.30-5.99) and discharge to a care facility (OR 3.04; 95%CI 1.16-7.93). Preoperative low skeletal muscle mass was only associated with infec-tious complications (OR 2.32; 95%CI 1.09-4.92). In a multivariable model, low skeletal muscle density was of added predictive value for postoperative complications (OR 2.57; 95%CI 1.21-5.45) to the strongest existing pre-dictor functional impairment (KATZ-ADL >= 2).Conclusion. Low skeletal muscle density, as a proxy of muscle quality, is associated with poor postoperative outcomes in older patients with advanced stage ovarian cancer. These findings can contribute to postoperative risk assessment and clinical decision making. (C) 2021 The Author(s). Published by Elsevier Inc. Show less
Kim, M.; Lee, S.P.; Kwak, S.; Yang, S.; Kim, Y.J.; Andreini, D.; ... ; Chang, H.J. 2021
Background: The association of age with coronary plaque dynamics is not well characterized by coronary computed tomography angiography (CCTA).Methods: From a multinational registry of patients who... Show moreBackground: The association of age with coronary plaque dynamics is not well characterized by coronary computed tomography angiography (CCTA).Methods: From a multinational registry of patients who underwent serial CCTA, 1153 subjects (61 +/- 5 years old, 61.1% male) were analyzed. Annualized volume changes of total, fibrous, fibrofatty, necrotic core, and dense calcification plaque components of the whole heart were compared by age quartile groups. Clinical events, a composite of all-cause death, acute coronary syndrome, and any revascularization after 30 days of the initial CCTA, were also analyzed. Random forest analysis was used to define the relative importance of age on plaque progression.Results: With a 3.3-years' median interval between the two CCTA, the median annual volume changes of total plaque in each age quartile group was 7.8, 10.5, 10.8, and 12.1 mm(3)/year and for dense calcification, 2.5, 4.6, 5.4, and 7.1 mm(3)/year, both of which demonstrated a tendency to increase by age (p-for-trend = 0.001 and < 0.001, respectively). However, this tendency was not observed in any other plaque components. The annual volume changes of total plaque and dense calcification were also significantly different in the propensity score-matched lowest age quartile group versus the other age groups as was the composite clinical event (log-rank p = 0.003). In random forest analysis, age had comparable importance in the total plaque volume progression as other traditional factors.Conclusions: The rate of whole-heart plaque progression and dense calcification increases depending on age. Age is a significant factor in plaque growth, the importance of which is comparable to other traditional risk factors. Show less
Background: Pancreatic neuroendocrine tumors (pNETs) have a high prevalence in patients with multiple endocrine neoplasia type 1 (MEN1) and are the leading cause of death. Tumor size is still... Show moreBackground: Pancreatic neuroendocrine tumors (pNETs) have a high prevalence in patients with multiple endocrine neoplasia type 1 (MEN1) and are the leading cause of death. Tumor size is still regarded as the main prognostic factor and therefore used for surgical decision-making. We assessed reliability and agreement of radiological and pathological tumor size in a population-based cohort of patients with MEN1-related pNETs. Methods: Patients were selected from the Dutch MEN1 database if they had undergone a resection for a pNET between 2003 and 2018. Radiological (MRI, CT, and endoscopic ultrasonography [EUS]) and pathological tumor size were collected from patient records. Measures of agreement (Bland-Altman plots with limits of agreement [LoA] and absolute agreement) and reliability (intraclass correlation coefficients [ICC] and unweighted kappa) were calculated for continuous and categorized (< or >= 2 cm) pNET size. Results: In 73 included patients, the median radiological and pathological tumor sizes measured were 22 (3-160) and 21 (4-200) mm, respectively. Mean bias between radiological and pathological tumor size was -0.2 mm and LoA ranged from -12.9 to 12.6 mm. For the subgroups of MRI, CT, and EUS, LoA of radiological and pathological tumor size ranged from -9.6 to 10.9, -15.9 to 15.8, and -13.9 to 11.0, respectively. ICCs for the overall cohort, MRI, CT, and EUS were 0.80, 0.86, 0.75, and 0.76, respectively. Based on the 2 cm criterion, agreement was 81.5%; hence, 12 patients (18.5%) were classified differently between imaging and pathology. Absolute agreement and kappa values of MRI, CT, and EUS were 88.6, 85.7, and 75.0%, and 0.77, 0.71, and 0.50, respectively. Conclusion: Within a population-based cohort, MEN1-related pNET size was not systematically over- or underestimated on preoperative imaging. Based on agreement and reliability measures, MRI is the preferred imaging modality. Show less
The objectives of this thesis were to elucidate the pathogenesis of metabolic heart disease, evaluate the associated changes in myocardial structure and contractile function, and determine the long... Show moreThe objectives of this thesis were to elucidate the pathogenesis of metabolic heart disease, evaluate the associated changes in myocardial structure and contractile function, and determine the long-term prognostic implications of subclinical myocardial dysfunction on all-cause mortality. Show less
Manual or automatic delineation of the esophageal tumor in CT images is known to be very challenging. This is due to the low contrast between the tumor and adjacent tissues, the anatomical... Show moreManual or automatic delineation of the esophageal tumor in CT images is known to be very challenging. This is due to the low contrast between the tumor and adjacent tissues, the anatomical variation of the esophagus, as well as the occasional presence of foreign bodies (e.g. feeding tubes). Physicians therefore usually exploit additional knowledge such as endoscopic findings, clinical history, additional imaging modalities like PET scans. Achieving his additional information is time-consuming, while the results are error-prone and might lead to non-deterministic results. In this paper we aim to investigate if and to what extent a simplified clinical workflow based on CT alone, allows one to automatically segment the esophageal tumor with sufficient quality. For this purpose, we present a fully automatic end-to-end esophageal tumor segmentation method based on convolutional neural networks (CNNs). The proposed network, called Dilated Dense Attention Unet (DDAUnet), leverages spatial and channel attention gates in each dense block to selectively concentrate on determinant feature maps and regions. Dilated convolutional layers are used to manage GPU memory and increase the network receptive field. We collected a dataset of 792 scans from 288 distinct patients including varying anatomies with air pockets, feeding tubes and proximal tumors. Repeatability and reproducibility studies were conducted for three distinct splits of training and validation sets. The proposed network achieved a DSC value of 0.79 +/- 0.20, a mean surface distance of 5.4 +/- 20.2mm and 95% Hausdorff distance of 14.7 +/- 25.0mm for 287 test scans, demonstrating promising results with a simplified clinical workflow based on CT alone. Our code is publicly available via https://github.com/yousefis/DenseUnet_Esophagus_Segmentation. Show less
In this paper we propose a supervised method to predict registration misalignment using convolutional neural networks (CNNs). This task is casted to a classification problem with multiple classes... Show moreIn this paper we propose a supervised method to predict registration misalignment using convolutional neural networks (CNNs). This task is casted to a classification problem with multiple classes of misalignment: "correct" 0-3 mm, "poor" 3-6 mm and "wrong" over 6 mm. Rather than a direct prediction, we propose a hierarchical approach, where the prediction is gradually refined from coarse to fine. Our solution is based on a convolutional Long Short-Term Memory (LSTM), using hierarchical misalignment predictions on three resolutions of the image pair, leveraging the intrinsic strengths of an LSTM for this problem. The convolutional LSTM is trained on a set of artificially generated image pairs obtained from artificial displacement vector fields (DVFs). Results on chest CT scans show that incorporating multi-resolution information, and the hierarchical use via an LSTM for this, leads to overall better F1 scores, with fewer misclassifications in a well-tuned registration setup. The final system yields an accuracy of 87.1%, and an average F1 score of 66.4% aggregated in two independent chest CT scan studies. Show less
This thesis showed that cardiac computed tomography may provide additional geometrical insights in valvular heart disease and cardiac remodeling. Further, computed tomography provides accurate... Show moreThis thesis showed that cardiac computed tomography may provide additional geometrical insights in valvular heart disease and cardiac remodeling. Further, computed tomography provides accurate anatomical images of the heart that may assist the planning of minimal invasive transcatheter cardiac procedures. Show less
Dam, L.F. van; Kroft, L.J.M.; Wal, L.I. van der; Cannegieter, S.C.; Eikenboom, J.; Jonge, E. de; ... ; Klok, F.A. 2020
Introduction: COVID-19 infections are associated with a high prevalence of venous thromboembolism, particularly pulmonary embolism (PE). It is suggested that COVID-19 associated PE represents in... Show moreIntroduction: COVID-19 infections are associated with a high prevalence of venous thromboembolism, particularly pulmonary embolism (PE). It is suggested that COVID-19 associated PE represents in situ immunothrombosis rather than venous thromboembolism, although the origin of thrombotic lesions in COVID-19 patients remains largely unknown.Methods: In this study, we assessed the clinical and computed tomography (CT) characteristics of PE in 23 consecutive patients with COVID-19 pneumonia and compared these to those of 100 consecutive control patients diagnosed with acute PE before the COVID-19 outbreak. Specifically, RV/LV diameter ratio, pulmonary artery trunk diameter and total thrombus load (according to Qanadli score) were measured and compared.Results: We observed that all thrombotic lesions in COVID-19 patients were found to be in lung parenchyma affected by COVID-19. Also, the thrombus load was lower in COVID-19 patients (Qanadli score -8%, 95% confidence interval [95%CI] -16 to -0.36%) as was the prevalence of the most proximal PE in the main/lobar pulmonary artery (17% versus 47%; -30%, 95%CI -44% to -8.2). Moreover, the mean RV/LV ratio (mean difference -0.23, 95%CI -0.39 to -0.07) and the prevalence of RV/LV ratio > 1.0 (prevalence difference -23%, 95%CI -41 to -0.86%) were lower in the COVID-19 patients.Conclusion: Our findings therefore suggest that the phenotype of COVID-19 associated PE indeed differs from PE in patients without COVID-19, fuelling the discussion on its pathophysiology. Show less
Cardiovascular (CV) imaging is an important tool in baseline risk assessment and detection of CV disease in oncology patients receiving cardiotoxic cancer therapies. This position statement... Show moreCardiovascular (CV) imaging is an important tool in baseline risk assessment and detection of CV disease in oncology patients receiving cardiotoxic cancer therapies. This position statement examines the role of echocardiography, cardiac magnetic resonance, nuclear cardiac imaging and computed tomography in the management of cancer patients. The Imaging and Cardio-Oncology Study Groups of the Heart Failure Association (HFA) of the European Society of Cardiology (ESC) in collaboration with the European Association of Cardiovascular Imaging (EACVI) and the Cardio-Oncology Council of the ESC have evaluated the current evidence for the value of modern CV imaging in the cardio-oncology field. The most relevant echocardiographic parameters, including global longitudinal strain and three-dimensional ejection fraction, are proposed. The protocol for baseline pre-treatment evaluation and specific surveillance algorithms or pathways for anthracycline chemotherapy, HER2-targeted therapies such as trastuzumab, vascular endothelial growth factor tyrosine kinase inhibitors, BCr-Abl tyrosine kinase inhibitors, proteasome inhibitors and immune checkpoint inhibitors are presented. The indications for CV imaging after completion of oncology treatment are considered. The typical consequences of radiation therapy and the possibility of their identification in the long term are also summarized. Special populations are discussed including female survivors planning pregnancy, patients with carcinoid disease, patients with cardiac tumours and patients with right heart failure. Future directions and ongoing CV imaging research in cardio-oncology are discussed. Show less
Purpose The aim of this study was to evaluate whether the addition of brain CT imaging data to a model incorporating clinical risk factors improves prediction of ischemic stroke recurrence over 5... Show morePurpose The aim of this study was to evaluate whether the addition of brain CT imaging data to a model incorporating clinical risk factors improves prediction of ischemic stroke recurrence over 5 years of follow-up. Methods A total of 638 patients with ischemic stroke from three centers were selected from the Dutch acute stroke study (DUST). CT-derived candidate predictors included findings on non-contrast CT, CT perfusion, and CT angiography. Five-year follow-up data were extracted from medical records. We developed a multivariable Cox regression model containing clinical predictors and an extended model including CT-derived predictors by applying backward elimination. We calculated net reclassification improvement and integrated discrimination improvement indices. Discrimination was evaluated with the optimism-corrected c-statistic and calibration with a calibration plot. Results During 5 years of follow-up, 56 patients (9%) had a recurrence. The c-statistic of the clinical model, which contained male sex, history of hyperlipidemia, and history of stroke or transient ischemic attack, was 0.61. Compared with the clinical model, the extended model, which contained previous cerebral infarcts on non-contrast CT and Alberta Stroke Program Early CT score greater than 7 on mean transit time maps derived from CT perfusion, had higher discriminative performance (c-statistic 0.65,P= 0.01). Inclusion of these CT variables led to a significant improvement in reclassification measures, by using the net reclassification improvement and integrated discrimination improvement indices. Conclusion Data from CT imaging significantly improved the discriminatory performance and reclassification in predicting ischemic stroke recurrence beyond a model incorporating clinical risk factors only. Show less