PET/CT based transformer model for multi-outcome prediction in oropharyngeal cancer.

BACKGROUND AND PURPOSE: To optimize our previously proposed TransRP, a model integrating CNN (convolutional neural network) and ViT (Vision Transformer) designed for recurrence-free survival prediction in oropharyngeal cancer and to extend its application to the prediction of multiple clinical outcomes, including locoregional control (LRC), Distant metastasis-free survival (DMFS) and overall survival (OS). MATERIALS AND METHODS: Data was collected from 400 patients (300 for training and 100 for testing) diagnosed with oropharyngeal squamous cell carcinoma (OPSCC) who underwent (chemo)radiotherapy at University Medical Center Groningen. Each patient's data comprised pre-treatment PET/CT scans, clinical parameters, and clinical outcome endpoints, namely LRC, DMFS and OS. The prediction performance of TransRP was compared with CNNs when inputting image data only. Additionally, three distinct methods (m1-3) of incorporating clinical predictors into TransRP training and one method (m4) that uses TransRP prediction as one parameter in a clinical Cox model were compared. RESULTS: TransRP achieved higher test C-index values of 0.61, 0.84 and 0.70 than CNNs for LRC, DMFS and OS, respectively. Furthermore, when incorporating TransRP's prediction into a clinical Cox model (m4), a higher C-index of 0.77 for OS was obtained. Compared with a clinical routine risk stratification model of OS, our model, using clinical variables, radiomics and TransRP prediction as predictors, achieved larger separations of survival curves between low, intermediate and high risk groups. CONCLUSION: TransRP outperformed CNN models for all endpoints. Combining clinical data and TransRP prediction in a Cox model achieved better OS prediction.

Effect of emphysema on AI software and human reader performance in lung nodule detection from low-dose chest CT.

BACKGROUND: Emphysema influences the appearance of lung tissue in computed tomography (CT). We evaluated whether this affects lung nodule detection by artificial intelligence (AI) and human readers (HR). METHODS: Individuals were selected from the "Lifelines" cohort who had undergone low-dose chest CT. Nodules in individuals without emphysema were matched to similar-sized nodules in individuals with at least moderate emphysema. AI results for nodular findings of 30-100 mm3 and 101-300 mm3 were compared to those of HR; two expert radiologists blindly reviewed discrepancies. Sensitivity and false positives (FPs)/scan were compared for emphysema and non-emphysema groups. RESULTS: Thirty-nine participants with and 82 without emphysema were included (n = 121, aged 61 ± 8 years (mean ± standard deviation), 58/121 males (47.9%)). AI and HR detected 196 and 206 nodular findings, respectively, yielding 109 concordant nodules and 184 discrepancies, including 118 true nodules. For AI, sensitivity was 0.68 (95% confidence interval 0.57-0.77) in emphysema versus 0.71 (0.62-0.78) in non-emphysema, with FPs/scan 0.51 and 0.22, respectively (p = 0.028). For HR, sensitivity was 0.76 (0.65-0.84) and 0.80 (0.72-0.86), with FPs/scan of 0.15 and 0.27 (p = 0.230). Overall sensitivity was slightly higher for HR than for AI, but this difference disappeared after the exclusion of benign lymph nodes. FPs/scan were higher for AI in emphysema than in non-emphysema (p = 0.028), while FPs/scan for HR were higher than AI for 30-100 mm3 nodules in non-emphysema (p = 0.009). CONCLUSIONS: AI resulted in more FPs/scan in emphysema compared to non-emphysema, a difference not observed for HR. RELEVANCE STATEMENT: In the creation of a benchmark dataset to validate AI software for lung nodule detection, the inclusion of emphysema cases is important due to the additional number of FPs. KEY POINTS: • The sensitivity of nodule detection by AI was similar in emphysema and non-emphysema. • AI had more FPs/scan in emphysema compared to non-emphysema. • Sensitivity and FPs/scan by the human reader were comparable for emphysema and non-emphysema. • Emphysema and non-emphysema representation in benchmark dataset is important for validating AI.

Localization of contrast-enhanced breast lesions in ultrafast screening MRI using deep convolutional neural networks.

OBJECTIVES: To develop a deep learning-based method for contrast-enhanced breast lesion detection in ultrafast screening MRI. MATERIALS AND METHODS: A total of 837 breast MRI exams of 488 consecutive patients were included. Lesion's location was independently annotated in the maximum intensity projection (MIP) image of the last time-resolved angiography with stochastic trajectories (TWIST) sequence for each individual breast, resulting in 265 lesions (190 benign, 75 malignant) in 163 breasts (133 women). YOLOv5 models were fine-tuned using training sets containing the same number of MIP images with and without lesions. A long short-term memory (LSTM) network was employed to help reduce false positive predictions. The integrated system was then evaluated on test sets containing enriched uninvolved breasts during cross-validation to mimic the performance in a screening scenario. RESULTS: In five-fold cross-validation, the YOLOv5x model showed a sensitivity of 0.95, 0.97, 0.98, and 0.99, with 0.125, 0.25, 0.5, and 1 false positive per breast, respectively. The LSTM network reduced 15.5% of the false positive prediction from the YOLO model, and the positive predictive value was increased from 0.22 to 0.25. CONCLUSIONS: A fine-tuned YOLOv5x model can detect breast lesions on ultrafast MRI with high sensitivity in a screening population, and the output of the model could be further refined by an LSTM network to reduce the amount of false positive predictions. CLINICAL RELEVANCE STATEMENT: The proposed integrated system would make the ultrafast MRI screening process more effective by assisting radiologists in prioritizing suspicious examinations and supporting the diagnostic workup. KEY POINTS: • Deep convolutional neural networks could be utilized to automatically pinpoint breast lesions in screening MRI with high sensitivity. • False positive predictions significantly increased when the detection models were tested on highly unbalanced test sets with more normal scans. • Dynamic enhancement patterns of breast lesions during contrast inflow learned by the long short-term memory networks helped to reduce false positive predictions.

Masked Conditional Variational Autoencoders for Chromosome Straightening.

Karyotyping is of importance for detecting chromosomal aberrations in human disease. However, chromosomes easily appear curved in microscopic images, which prevents cytogeneticists from analyzing chromosome types. To address this issue, we propose a framework for chromosome straightening, which comprises a preliminary processing algorithm and a generative model called masked conditional variational autoencoders (MC-VAE). The processing method utilizes patch rearrangement to address the difficulty in erasing low degrees of curvature, providing reasonable preliminary results for the MC-VAE. The MC-VAE further straightens the results by leveraging chromosome patches conditioned on their curvatures to learn the mapping between banding patterns and conditions. During model training, we apply a masking strategy with a high masking ratio to train the MC-VAE with eliminated redundancy. This yields a non-trivial reconstruction task, allowing the model to effectively preserve chromosome banding patterns and structure details in the reconstructed results. Extensive experiments on three public datasets with two stain styles show that our framework surpasses the performance of state-of-the-art methods in retaining banding patterns and structure details. Compared to using real-world bent chromosomes, the use of high-quality straightened chromosomes generated by our proposed method can improve the performance of various deep learning models for chromosome classification by a large margin. Such a straightening approach has the potential to be combined with other karyotyping systems to assist cytogeneticists in chromosome analysis.

Deep learning-based outcome prediction using PET/CT and automatically predicted probability maps of primary tumor in patients with oropharyngeal cancer.

BACKGROUND AND OBJECTIVE: Recently, deep learning (DL) algorithms showed to be promising in predicting outcomes such as distant metastasis-free survival (DMFS) and overall survival (OS) using pre-treatment imaging in head and neck cancer. Gross Tumor Volume of the primary tumor (GTVp) segmentation is used as an additional channel in the input to DL algorithms to improve model performance. However, the binary segmentation mask of the GTVp directs the focus of the network to the defined tumor region only and uniformly. DL models trained for tumor segmentation have also been used to generate predicted tumor probability maps (TPM) where each pixel value corresponds to the degree of certainty of that pixel to be classified as tumor. The aim of this study was to explore the effect of using TPM as an extra input channel of CT- and PET-based DL prediction models for oropharyngeal cancer (OPC) patients in terms of local control (LC), regional control (RC), DMFS and OS. METHODS: We included 399 OPC patients from our institute that were treated with definitive (chemo)radiation. For each patient, CT and PET scans and GTVp contours, used for radiotherapy treatment planning, were collected. We first trained a previously developed 2.5D DL framework for tumor probability prediction by 5-fold cross validation using 131 patients. Then, a 3D ResNet18 was trained for outcome prediction using the 3D TPM as one of the possible inputs. The endpoints were LC, RC, DMFS, and OS. We performed 3-fold cross validation on 168 patients for each endpoint using different combinations of image modalities as input. The final prediction in the test set (100) was obtained by averaging the predictions of the 3-fold models. The C-index was used to evaluate the discriminative performance of the models. RESULTS: The models trained replacing the GTVp contours with the TPM achieved the highest C-indexes for LC (0.74) and RC (0.60) prediction. For OS, using the TPM or the GTVp as additional image modality resulted in comparable C-indexes (0.72 and 0.74). CONCLUSIONS: Adding predicted TPMs instead of GTVp contours as an additional input channel for DL-based outcome prediction models improved model performance for LC and RC.

Comparison of computed tomography image features extracted by radiomics, self-supervised learning and end-to-end deep learning for outcome prediction of oropharyngeal cancer.

Background and purpose: To compare the prediction performance of image features of computed tomography (CT) images extracted by radiomics, self-supervised learning and end-to-end deep learning for local control (LC), regional control (RC), locoregional control (LRC), distant metastasis-free survival (DMFS), tumor-specific survival (TSS), overall survival (OS) and disease-free survival (DFS) of oropharyngeal squamous cell carcinoma (OPSCC) patients after (chemo)radiotherapy. Methods and materials: The OPC-Radiomics dataset was used for model development and independent internal testing and the UMCG-OPC set for external testing. Image features were extracted from the Gross Tumor Volume contours of the primary tumor (GTVt) regions in CT scans when using radiomics or a self-supervised learning-based method (autoencoder). Clinical and combined (radiomics, autoencoder or end-to-end) models were built using multivariable Cox proportional-hazard analysis with clinical features only and both clinical and image features for LC, RC, LRC, DMFS, TSS, OS and DFS prediction, respectively. Results: In the internal test set, combined autoencoder models performed better than clinical models and combined radiomics models for LC, RC, LRC, DMFS, TSS and DFS prediction (largest improvements in C-index: 0.91 vs. 0.76 in RC and 0.74 vs. 0.60 in DMFS). In the external test set, combined radiomics models performed better than clinical and combined autoencoder models for all endpoints (largest improvements in LC, 0.82 vs. 0.71). Furthermore, combined models performed better in risk stratification than clinical models and showed good calibration for most endpoints. Conclusions: Image features extracted using self-supervised learning showed best internal prediction performance while radiomics features have better external generalizability.

Automated Breast Density Assessment in MRI Using Deep Learning and Radiomics: Strategies for Reducing Inter-Observer Variability.

BACKGROUND: Accurate breast density evaluation allows for more precise risk estimation but suffers from high inter-observer variability. PURPOSE: To evaluate the feasibility of reducing inter-observer variability of breast density assessment through artificial intelligence (AI) assisted interpretation. STUDY TYPE: Retrospective. POPULATION: Six hundred and twenty-one patients without breast prosthesis or reconstructions were randomly divided into training (N = 377), validation (N = 98), and independent test (N = 146) datasets. FIELD STRENGTH/SEQUENCE: 1.5 T and 3.0 T; T1-weighted spectral attenuated inversion recovery. ASSESSMENT: Five radiologists independently assessed each scan in the independent test set to establish the inter-observer variability baseline and to reach a reference standard. Deep learning and three radiomics models were developed for three classification tasks: (i) four Breast Imaging-Reporting and Data System (BI-RADS) breast composition categories (A-D), (ii) dense (categories C, D) vs. non-dense (categories A, B), and (iii) extremely dense (category D) vs. moderately dense (categories A-C). The models were tested against the reference standard on the independent test set. AI-assisted interpretation was performed by majority voting between the models and each radiologist's assessment. STATISTICAL TESTS: Inter-observer variability was assessed using linear-weighted kappa (κ) statistics. Kappa statistics, accuracy, and area under the receiver operating characteristic curve (AUC) were used to assess models against reference standard. RESULTS: In the independent test set, five readers showed an overall substantial agreement on tasks (i) and (ii), but moderate agreement for task (iii). The best-performing model showed substantial agreement with reference standard for tasks (i) and (ii), but moderate agreement for task (iii). With the assistance of the AI models, almost perfect inter-observer variability was obtained for tasks (i) (mean κ = 0.86), (ii) (mean κ = 0.94), and (iii) (mean κ = 0.94). DATA CONCLUSION: Deep learning and radiomics models have the potential to help reduce inter-observer variability of breast density assessment. LEVEL OF EVIDENCE: 3 TECHNICAL EFFICACY: Stage 1.

CT-based deep multi-label learning prediction model for outcome in patients with oropharyngeal squamous cell carcinoma.

BACKGROUND: Personalized treatment is increasingly required for oropharyngeal squamous cell carcinoma (OPSCC) patients due to emerging new cancer subtypes and treatment options. Outcome prediction model can help identify low or high-risk patients who may be suitable to receive de-escalation or intensified treatment approaches. PURPOSE: To develop a deep learning (DL)-based model for predicting multiple and associated efficacy endpoints in OPSCC patients based on computed tomography (CT). METHODS: Two patient cohorts were used in this study: a development cohort consisting of 524 OPSCC patients (70% for training and 30% for independent testing) and an external test cohort of 396 patients. Pre-treatment CT-scans with the gross primary tumor volume contours (GTVt) and clinical parameters were available to predict endpoints, including 2-year local control (LC), regional control (RC), locoregional control (LRC), distant metastasis-free survival (DMFS), disease-specific survival (DSS), overall survival (OS), and disease-free survival (DFS). We proposed DL outcome prediction models with the multi-label learning (MLL) strategy that integrates the associations of different endpoints based on clinical factors and CT-scans. RESULTS: The multi-label learning models outperformed the models that were developed based on a single endpoint for all endpoints especially with high AUCs ≥ 0.80 for 2-year RC, DMFS, DSS, OS, and DFS in the internal independent test set and for all endpoints except 2-year LRC in the external test set. Furthermore, with the models developed, patients could be stratified into high and low-risk groups that were significantly different for all endpoints in the internal test set and for all endpoints except DMFS in the external test set. CONCLUSION: MLL models demonstrated better discriminative ability for all 2-year efficacy endpoints than single outcome models in the internal test and for all endpoints except LRC in the external set.

Deep learning aided oropharyngeal cancer segmentation with adaptive thresholding for predicted tumor probability in FDG PET and CT images.

Objective. Tumor segmentation is a fundamental step for radiotherapy treatment planning. To define an accurate segmentation of the primary tumor (GTVp) of oropharyngeal cancer patients (OPC) each image volume is explored slice-by-slice from different orientations on different image modalities. However, the manual fixed boundary of segmentation neglects the spatial uncertainty known to occur in tumor delineation. This study proposes a novel deep learning-based method that generates probability maps which capture the model uncertainty in the segmentation task.Approach. We included 138 OPC patients treated with (chemo)radiation in our institute. Sequences of 3 consecutive 2D slices of concatenated FDG-PET/CT images and GTVp contours were used as input. Our framework exploits inter and intra-slice context using attention mechanisms and bi-directional long short term memory (Bi-LSTM). Each slice resulted in three predictions that were averaged. A 3-fold cross validation was performed on sequences extracted from the axial, sagittal, and coronal plane. 3D volumes were reconstructed and single- and multi-view ensembling were performed to obtain final results. The output is a tumor probability map determined by averaging multiple predictions.Main Results. Model performance was assessed on 25 patients at different probability thresholds. Predictions were the closest to the GTVp at a threshold of 0.9 (mean surface DSC of 0.81, median HD95of 3.906 mm).Significance. The promising results of the proposed method show that is it possible to offer the probability maps to radiation oncologists to guide them in a in a slice-by-slice adaptive GTVp segmentation.

Survival prediction for stage I-IIIA non-small cell lung cancer using deep learning.

BACKGROUND AND PURPOSE: The aim of this study was to develop and evaluate a prediction model for 2-year overall survival (OS) in stage I-IIIA non-small cell lung cancer (NSCLC) patients who received definitive radiotherapy by considering clinical variables and image features from pre-treatment CT-scans. MATERIALS AND METHODS: NSCLC patients who received stereotactic radiotherapy were prospectively collected at the UMCG and split into a training and a hold out test set including 189 and 81 patients, respectively. External validation was performed on 228 NSCLC patients who were treated with radiation or concurrent chemoradiation at the Maastro clinic (Lung1 dataset). A hybrid model that integrated both image and clinical features was implemented using deep learning. Image features were learned from cubic patches containing lung tumours extracted from pre-treatment CT scans. Relevant clinical variables were selected by univariable and multivariable analyses. RESULTS: Multivariable analysis showed that age and clinical stage were significant prognostic clinical factors for 2-year OS. Using these two clinical variables in combination with image features from pre-treatment CT scans, the hybrid model achieved a median AUC of 0.76 [95 % CI: 0.65-0.86] and 0.64 [95 % CI: 0.58-0.70] on the complete UMCG and Maastro test sets, respectively. The Kaplan-Meier survival curves showed significant separation between low and high mortality risk groups on these two test sets (log-rank test: p-value < 0.001, p-value = 0.012, respectively) CONCLUSION: We demonstrated that a hybrid model could achieve reasonable performance by utilizing both clinical and image features for 2-year OS prediction. Such a model has the potential to identify patients with high mortality risk and guide clinical decision making.

Standardization of Artificial Intelligence Development in Radiotherapy.

Application of Artificial Intelligence (AI) tools has recently gained interest in the fields of medical imaging and radiotherapy. Even though there have been many papers published in these domains in the last few years, clinical assessment of the proposed AI methods is limited due to the lack of standardized protocols that can be used to validate the performance of the developed tools. Moreover, each stakeholder uses their own methods, tools, and evaluation criteria. Communication between different stakeholders is limited or absent, which makes it hard to easily exchange models between different clinics. These issues are not limited to radiotherapy but exist in every AI application domain. To deal with these issues, methods like the Machine Learning Canvas, Datasets for Datasheets, and Model cards have been developed. They aim to provide information of the whole creation pipeline of AI solutions, of the datasets used to develop AI, along with their biases, as well as to facilitate easier collaboration/communication between different stakeholders and facilitate the clinical introduction of AI. This work introduces the concepts of these 3 open-source solutions including the author's experiences applying them to AI applications for radiotherapy.

Pre-screening to guide coronary artery calcium scoring for early identification of high-risk individuals in the general population.

AIMS: To evaluate the ability of Systematic COronary Risk Estimation 2 (SCORE2) and other pre-screening methods to identify individuals with high coronary artery calcium score (CACS) in the general population. METHODS AND RESULTS: Computed tomography-based CACS quantification was performed in 6530 individuals aged 45 years or older from the general population. Various pre-screening methods to guide referral for CACS were evaluated. Miss rates for high CACS (CACS ≥300 and ≥100) were evaluated for various pre-screening methods: moderate (≥5%) and high (≥10%) SCORE2 risk, any traditional coronary artery disease (CAD) risk factor, any Risk Or Benefit IN Screening for CArdiovascular Disease (ROBINSCA) risk factor, and moderately (>3 mg/24 h) increased urine albumin excretion (UAE). Out of 6530 participants, 643 (9.8%) had CACS ≥300 and 1236 (18.9%) had CACS ≥100. For CACS ≥300 and CACS ≥100, miss rate was 32 and 41% for pre-screening by moderate (≥5%) SCORE2 risk and 81 and 87% for high (≥10%) SCORE2 risk, respectively. For CACS ≥300 and CACS ≥100, miss rate was 8 and 11% for pre-screening by at least one CAD risk factor, 24 and 25% for at least one ROBINSCA risk factor, and 67 and 67% for moderately increased UAE, respectively. CONCLUSION: Many individuals with high CACS in the general population are left unidentified when only performing CACS in case of at least moderate (≥5%) SCORE2, which closely resembles current clinical practice. Less stringent pre-screening by presence of at least one CAD risk factor to guide CACS identifies more individuals with high CACS and could improve CAD prevention.

2D Gait Skeleton Data Normalization for Quantitative Assessment of Movement Disorders from Freehand Single Camera Video Recordings.

Overlapping phenotypic features between Early Onset Ataxia (EOA) and Developmental Coordination Disorder (DCD) can complicate the clinical distinction of these disorders. Clinical rating scales are a common way to quantify movement disorders but in children these scales also rely on the observer's assessment and interpretation. Despite the introduction of inertial measurement units for objective and more precise evaluation, special hardware is still required, restricting their widespread application. Gait video recordings of movement disorder patients are frequently captured in routine clinical settings, but there is presently no suitable quantitative analysis method for these recordings. Owing to advancements in computer vision technology, deep learning pose estimation techniques may soon be ready for convenient and low-cost clinical usage. This study presents a framework based on 2D video recording in the coronal plane and pose estimation for the quantitative assessment of gait in movement disorders. To allow the calculation of distance-based features, seven different methods to normalize 2D skeleton keypoint data derived from pose estimation using deep neural networks applied to freehand video recording of gait were evaluated. In our experiments, 15 children (five EOA, five DCD and five healthy controls) were asked to walk naturally while being videotaped by a single camera in 1280 × 720 resolution at 25 frames per second. The high likelihood of the prediction of keypoint locations (mean = 0.889, standard deviation = 0.02) demonstrates the potential for distance-based features derived from routine video recordings to assist in the clinical evaluation of movement in EOA and DCD. By comparison of mean absolute angle error and mean variance of distance, the normalization methods using the Euclidean (2D) distance of left shoulder and right hip, or the average distance from left shoulder to right hip and from right shoulder to left hip were found to better perform for deriving distance-based features and further quantitative assessment of movement disorders.

Using deep learning to safely exclude lesions with only ultrafast breast MRI to shorten acquisition and reading time.

OBJECTIVES: To investigate the feasibility of automatically identifying normal scans in ultrafast breast MRI with artificial intelligence (AI) to increase efficiency and reduce workload. METHODS: In this retrospective analysis, 837 breast MRI examinations performed on 438 women from April 2016 to October 2019 were included. The left and right breasts in each examination were labelled normal (without suspicious lesions) or abnormal (with suspicious lesions) based on final interpretation. Maximum intensity projection (MIP) images of each breast were then used to train a deep learning model. A high sensitivity threshold was calculated based on the detection trade - off (DET) curve on the validation set. The performance of the model was evaluated by receiver operating characteristic analysis of the independent test set. The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) with the high sensitivity threshold were calculated. RESULTS: The independent test set consisted of 178 examinations of 149 patients (mean age, 44 years ± 14 [standard deviation]). The trained model achieved an AUC of 0.81 (95% CI: 0.75-0.88) on the independent test set. Applying a threshold of 0.25 yielded a sensitivity of 98% (95% CI: 90%; 100%), an NPV of 98% (95% CI: 89%; 100%), a workload reduction of 15.7%, and a scan time reduction of 16.6%. CONCLUSION: This deep learning model has a high potential to help identify normal scans in ultrafast breast MRI and thereby reduce radiologists' workload and scan time. KEY POINTS: • Deep learning in TWIST may eliminate the necessity of additional sequences for identifying normal breasts during MRI screening. • Workload and scanning time reductions of 15.7% and 16.6%, respectively, could be achieved with the cost of 1 (1 of 55) false negative prediction.

Computer 3D modeling of radiofrequency ablation of atypical cartilaginous tumours in long bones using finite element methods and real patient anatomy.

BACKGROUND: Radiofrequency ablation (RFA) is a minimally invasive technique used for the treatment of neoplasms, with a growing interest in the treatment of bone tumours. However, the lack of data concerning the size of the resulting ablation zones in RFA of bone tumours makes prospective planning challenging, needed for safe and effective treatment. METHODS: Using retrospective computed tomography and magnetic resonance imaging data from patients treated with RFA of atypical cartilaginous tumours (ACTs), the bone, tumours, and final position of the RFA electrode were segmented from the medical images and used in finite element models to simulate RFA. Tissue parameters were optimised, and boundary conditions were defined to mimic the clinical scenario. The resulting ablation diameters from postoperative images were then measured and compared to the ones from the simulations, and the error between them was calculated. RESULTS: Seven cases had all the information required to create the finite element models. The resulting median error (in all three directions) was -1 mm, with interquartile ranges from -3 to 3 mm. The three-dimensional models showed that the thermal damage concentrates close to the cortical wall in the first minutes and then becomes more evenly distributed. CONCLUSIONS: Computer simulations can predict the ablation diameters with acceptable accuracy and may thus be utilised for patient planning. This could allow interventional radiologists to accurately define the time, electrode length, and position required to treat ACTs with RFA and make adjustments as needed to guarantee total tumour destruction while sparing as much healthy tissue as possible.

Coronary calcium scoring as first-line test to detect and exclude coronary artery disease in patients presenting to the general practitioner with stable chest pain: protocol of the cluster-randomised CONCRETE trial.

INTRODUCTION: Identifying and excluding coronary artery disease (CAD) in patients with atypical angina pectoris (AP) and non-specific thoracic complaints is a challenge for general practitioners (GPs). A diagnostic and prognostic tool could help GPs in determining the likelihood of CAD and guide patient management. Studies in outpatient settings have shown that the CT-based coronary calcium score (CCS) has high accuracy for diagnosis and exclusion of CAD. However, the CT CCS test has not been tested in a primary care setting. In the COroNary Calcium scoring as fiRst-linE Test to dEtect and exclude coronary artery disease in GPs patients with stable chest pain (CONCRETE) study, the impact of direct access of GPs to CT CCS will be investigated. We hypothesise that this will allow for early diagnosis of CAD and treatment, more efficient referral to the cardiologist and a reduction of healthcare-related costs. METHODS AND ANALYSIS: CONCRETE is a pragmatic multicentre trial with a cluster randomised design, in which direct GP access to the CT CCS test is compared with standard of care. In both arms, at least 40 GP offices, and circa 800 patients with atypical AP and non-specific thoracic complaints will be included. To determine the increase in detection and treatment rate of CAD in GP offices, the CVRM registration rate is derived from the GPs electronic registration system. Individual patients' data regarding cardiovascular risk factors, expressed chest pain complaints, quality of life, downstream testing and CAD diagnosis will be collected through questionnaires and the electronic GP dossier. ETHICS AND DISSEMINATION: CONCRETE has been approved by the Medical Ethical Committee of the University Medical Center of Groningen. TRIAL REGISTRATION NUMBER: NTR 7475; Pre-results.

Facilitating standardized COVID-19 suspicion prediction based on computed tomography radiomics in a multi-demographic setting.

OBJECTIVE: To develop an automatic COVID-19 Reporting and Data System (CO-RADS)-based classification in a multi-demographic setting. METHODS: This multi-institutional review boards-approved retrospective study included 2720 chest CT scans (mean age, 58 years [range 18-100 years]) from Italian and Russian patients. Three board-certified radiologists from three countries assessed randomly selected subcohorts from each population and provided CO-RADS-based annotations. CT radiomic features were extracted from the selected subcohorts after preprocessing steps like lung lobe segmentation and automatic noise reduction. We compared three machine learning models, logistic regression (LR), multilayer perceptron (MLP), and random forest (RF) for the automated CO-RADS classification. Model evaluation was carried out in two scenarios, first, training on a mixed multi-demographic subcohort and testing on an independent hold-out dataset. In the second scenario, training was done on a single demography and externally validated on the other demography. RESULTS: The overall inter-observer agreement for the CO-RADS scoring between the radiologists was substantial (k = 0.80). Irrespective of the type of validation test scenario, suspected COVID-19 CT scans were identified with an accuracy of 84%. SHapley Additive exPlanations (SHAP) interpretation showed that the "wavelet_(LH)_GLCM_Imc1" feature had a positive impact on COVID prediction both with and without noise reduction. The application of noise reduction improved the overall performance between the classifiers for all types. CONCLUSION: Using an automated model based on the COVID-19 Reporting and Data System (CO-RADS), we achieved clinically acceptable performance in a multi-demographic setting. This approach can serve as a standardized tool for automated COVID-19 assessment. KEYPOINTS: • Automatic CO-RADS scoring of large-scale multi-demographic chest CTs with mean AUC of 0.93 ± 0.04. • Validation procedure resembles TRIPOD 2b and 3 categories, enhancing the quality of experimental design to test the cross-dataset domain shift between institutions aiding clinical integration. • Identification of COVID-19 pneumonia in the presence of community-acquired pneumonia and other comorbidities with an AUC of 0.92.

Structural similarity analysis of midfacial fractures-a feasibility study.

The structural similarity index metric is used to measure the similarity between two images. The aim here was to study the feasibility of this metric to measure the structural similarity and fracture characteristics of midfacial fractures in computed tomography (CT) datasets following radiation dose reduction, iterative reconstruction (IR) and deep learning reconstruction. Zygomaticomaxillary fractures were inflicted on four human cadaver specimen and scanned with standard and low dose CT protocols. Datasets were reconstructed using varying strengths of IR and the subsequently applying the PixelShine™ deep learning algorithm as post processing. Individual small and non-dislocated fractures were selected for the data analysis. After attenuating the osseous anatomy of interest, registration was performed to superimpose the datasets and subsequently to measure by structural image quality. Changes to the fracture characteristics were measured by comparing each fracture to the mirrored contralateral anatomy. Twelve fracture locations were included in the data analysis. The most structural image quality changes occurred with radiation dose reduction (0.980036±0.011904), whilst the effects of IR strength (0.995399±0.001059) and the deep learning algorithm (0.999996±0.000002) were small. Radiation dose reduction and IR strength tended to affect the fracture characteristics. Both the structural image quality and fracture characteristics were not affected by the use of the deep learning algorithm. In conclusion, evidence is provided for the feasibility of using the structural similarity index metric for the analysis of structural image quality and fracture characteristics.

The Diagnostic and Prognostic Value of Coronary Calcium Scoring in Stable Chest Pain Patients: A Narrative Review.

BACKGROUND: Non-contrast computed tomography (CT) scanning allows for reliable coronary calcium score (CCS) calculation at a low radiation dose and has been well established as marker to assess the future risk of coronary artery disease (CAD) events in asymptomatic individuals. However, the diagnostic and prognostic value in symptomatic patients remains a matter of debate. This narrative review focuses on the available evidence for CCS in patients with stable chest pain complaints. METHOD: PubMed, Embase, and Web of Science were searched for literature using search terms related to three overarching categories: CT, symptomatic chest pain patients, and coronary calcium. The search resulted in 42 articles fulfilling the inclusion and exclusion criteria: 27 articles (n = 38 137 patients) focused on diagnostic value and 23 articles (n = 44 683 patients) on prognostic value of CCS. Of these, 10 articles (n = 21 208 patients) focused on both the diagnostic and prognostic value of CCS. RESULTS: Between 22 and 10 037 patients were included in the studies on the diagnostic and prognostic value of CCS, including 43 % and 51 % patients with CCS 0. The most evidence is available for patients with a low and intermediate pre-test probability (PTP) of CAD. Overall, the prevalence of obstructive CAD (OCAD, defined as a luminal stenosis of ≥ 50 % in any of the coronary arteries) as determined with CT coronary angiography in CCS 0 patients, was 4.4 % (n = 703/16 074) with a range of 0-26 % in individual studies. The event rate for major adverse cardiac events (MACE) ranged from 0 % to 2.1 % during a follow-up of 1.6 to 6.8 years, resulting in a high negative predictive value for MACE between 98 % and 100 % in CCS 0 patients. At increasing CCS, the OCAD probability and MACE risk increased. OCAD was present in 58.3 % (n = 617/1058) of CCS > 400 patients with percentages ranging from 20 % to 94 % and MACE occurred in 16.7 % (n = 175/1048) of these patients with percentages ranging from 6.9 % to 50 %. CONCLUSION: Accumulating evidence shows that OCAD is unlikely and the MACE risk is very low in symptomatic patients with CCS 0, especially in those with low and intermediate PTPs. This suggests a role of CCS as a gatekeeper for additional diagnostic testing. Increasing CCS is related to an increasing probability of OCAD and risk of cardiac events. Additional research is needed to assess the value of CCS in women and patient management in a primary healthcare setting. KEY POINTS: · A CCS of zero makes OCAD in patients at low-intermediate PTP unlikely. · A CCS of zero is related to a very low risk of MACE. · Categories of increasing CCS are related to increasing rates of OCAD and MACE. · Future studies should focus on the diagnostic and prognostic value of CCS in symptomatic women and the role in primary care. CITATION FORMAT: · Koopman MY, Willemsen RT, van der Harst P et al. The Diagnostic and Prognostic Value of Coronary Calcium Scoring in Stable Chest Pain Patients: A Narrative Review. Fortschr Röntgenstr 2022; 194: 257 - 265.

Performance of a deep learning-based lung nodule detection system as an alternative reader in a Chinese lung cancer screening program.

OBJECTIVE: To evaluate the performance of a deep learning-based computer-aided detection (DL-CAD) system in a Chinese low-dose CT (LDCT) lung cancer screening program. MATERIALS AND METHODS: One-hundred-and-eighty individuals with a lung nodule on their baseline LDCT lung cancer screening scan were randomly mixed with screenees without nodules in a 1:1 ratio (total: 360 individuals). All scans were assessed by double reading and subsequently processed by an academic DL-CAD system. The findings of double reading and the DL-CAD system were then evaluated by two senior radiologists to derive the reference standard. The detection performance was evaluated by the Free Response Operating Characteristic curve, sensitivity and false-positive (FP) rate. The senior radiologists categorized nodules according to nodule diameter, type (solid, part-solid, non-solid) and Lung-RADS. RESULTS: The reference standard consisted of 262 nodules ≥ 4 mm in 196 individuals; 359 findings were considered false positives. The DL-CAD system achieved a sensitivity of 90.1% with 1.0 FP/scan for detection of lung nodules regardless of size or type, whereas double reading had a sensitivity of 76.0% with 0.04 FP/scan (P = 0.001). The sensitivity for detection of nodules ≥ 4 - ≤ 6 mm was significantly higher with DL-CAD than with double reading (86.3% vs. 58.9% respectively; P = 0.001). Sixty-three nodules were only identified by the DL-CAD system, and 27 nodules only found by double reading. The DL-CAD system reached similar performance compared to double reading in Lung-RADS 3 (94.3% vs. 90.0%, P = 0.549) and Lung-RADS 4 nodules (100.0% vs. 97.0%, P = 1.000), but showed a higher sensitivity in Lung-RADS 2 (86.2% vs. 65.4%, P < 0.001). CONCLUSIONS: The DL-CAD system can accurately detect pulmonary nodules on LDCT, with an acceptable false-positive rate of 1 nodule per scan and has higher detection performance than double reading. This DL-CAD system may assist radiologists in nodule detection in LDCT lung cancer screening.