Radiological age assessment based on clavicle ossification in CT: enhanced accuracy through deep learning.

BACKGROUND: Radiological age assessment using reference studies is inherently limited in accuracy due to a finite number of assignable skeletal maturation stages. To overcome this limitation, we present a deep learning approach for continuous age assessment based on clavicle ossification in computed tomography (CT). METHODS: Thoracic CT scans were retrospectively collected from the picture archiving and communication system. Individuals aged 15.0 to 30.0 years examined in routine clinical practice were included. All scans were automatically cropped around the medial clavicular epiphyseal cartilages. A deep learning model was trained to predict a person's chronological age based on these scans. Performance was evaluated using mean absolute error (MAE). Model performance was compared to an optimistic human reader performance estimate for an established reference study method. RESULTS: The deep learning model was trained on 4,400 scans of 1,935 patients (training set: mean age = 24.2 years ± 4.0, 1132 female) and evaluated on 300 scans of 300 patients with a balanced age and sex distribution (test set: mean age = 22.5 years ± 4.4, 150 female). Model MAE was 1.65 years, and the highest absolute error was 6.40 years for females and 7.32 years for males. However, performance could be attributed to norm-variants or pathologic disorders. Human reader estimate MAE was 1.84 years and the highest absolute error was 3.40 years for females and 3.78 years for males. CONCLUSIONS: We present a deep learning approach for continuous age predictions using CT volumes highlighting the medial clavicular epiphyseal cartilage with performance comparable to the human reader estimate.

Integrated intravoxel incoherent motion tensor and diffusion tensor brain MRI in a single fast acquisition.

The acquisition of intravoxel incoherent motion (IVIM) data and diffusion tensor imaging (DTI) data from the brain can be integrated into a single measurement, which offers the possibility to determine orientation-dependent (tensorial) perfusion parameters in addition to established IVIM and DTI parameters. The purpose of this study was to evaluate the feasibility of such a protocol with a clinically feasible scan time below 6 min and to use a model-selection approach to find a set of DTI and IVIM tensor parameters that most adequately describes the acquired data. Diffusion-weighted images of the brain were acquired at 3 T in 20 elderly participants with cerebral small vessel disease using a multiband echoplanar imaging sequence with 15 b-values between 0 and 1000 s/mm2 and six non-collinear diffusion gradient directions for each b-value. Seven different IVIM-diffusion models with 4 to 14 parameters were implemented, which modeled diffusion and pseudo-diffusion as scalar or tensor quantities. The models were compared with respect to their fitting performance based on the goodness of fit (sum of squared fit residuals, chi2 ) and their Akaike weights (calculated from the corrected Akaike information criterion). Lowest chi2 values were found using the model with the largest number of model parameters. However, significantly highest Akaike weights indicating the most appropriate models for the acquired data were found with a nine-parameter IVIM-DTI model (with isotropic perfusion modeling) in normal-appearing white matter (NAWM), and with an 11-parameter model (IVIM-DTI with additional pseudo-diffusion anisotropy) in white matter with hyperintensities (WMH) and in gray matter (GM). The latter model allowed for the additional calculation of the fractional anisotropy of the pseudo-diffusion tensor (with a median value of 0.45 in NAWM, 0.23 in WMH, and 0.36 in GM), which is not accessible with the usually performed IVIM acquisitions based on three orthogonal diffusion-gradient directions.

ChatGPT makes medicine easy to swallow: an exploratory case study on simplified radiology reports.

OBJECTIVES: To assess the quality of simplified radiology reports generated with the large language model (LLM) ChatGPT and to discuss challenges and chances of ChatGPT-like LLMs for medical text simplification. METHODS: In this exploratory case study, a radiologist created three fictitious radiology reports which we simplified by prompting ChatGPT with "Explain this medical report to a child using simple language." In a questionnaire, we tasked 15 radiologists to rate the quality of the simplified radiology reports with respect to their factual correctness, completeness, and potential harm for patients. We used Likert scale analysis and inductive free-text categorization to assess the quality of the simplified reports. RESULTS: Most radiologists agreed that the simplified reports were factually correct, complete, and not potentially harmful to the patient. Nevertheless, instances of incorrect statements, missed relevant medical information, and potentially harmful passages were reported. CONCLUSION: While we see a need for further adaption to the medical field, the initial insights of this study indicate a tremendous potential in using LLMs like ChatGPT to improve patient-centered care in radiology and other medical domains. CLINICAL RELEVANCE STATEMENT: Patients have started to use ChatGPT to simplify and explain their medical reports, which is expected to affect patient-doctor interaction. This phenomenon raises several opportunities and challenges for clinical routine. KEY POINTS: • Patients have started to use ChatGPT to simplify their medical reports, but their quality was unknown. • In a questionnaire, most participating radiologists overall asserted good quality to radiology reports simplified with ChatGPT. However, they also highlighted a notable presence of errors, potentially leading patients to draw harmful conclusions. • Large language models such as ChatGPT have vast potential to enhance patient-centered care in radiology and other medical domains. To realize this potential while minimizing harm, they need supervision by medical experts and adaption to the medical field.

Automated localization of the medial clavicular epiphyseal cartilages using an object detection network: a step towards deep learning-based forensic age assessment.

BACKGROUND: Deep learning is a promising technique to improve radiological age assessment. However, expensive manual annotation by experts poses a bottleneck for creating large datasets to appropriately train deep neural networks. We propose an object detection approach to automatically annotate the medial clavicular epiphyseal cartilages in computed tomography (CT) scans. METHODS: The sternoclavicular joints were selected as structure-of-interest (SOI) in chest CT scans and served as an easy-to-identify proxy for the actual medial clavicular epiphyseal cartilages. CT slices containing the SOI were manually annotated with bounding boxes around the SOI. All slices in the training set were used to train the object detection network RetinaNet. Afterwards, the network was applied individually to all slices of the test scans for SOI detection. Bounding box and slice position of the detection with the highest classification score were used as the location estimate for the medial clavicular epiphyseal cartilages inside the CT scan. RESULTS: From 100 CT scans of 82 patients, 29,656 slices were used for training and 30,846 slices from 110 CT scans of 110 different patients for testing the object detection network. The location estimate from the deep learning approach for the SOI was in a correct slice in 97/110 (88%), misplaced by one slice in 5/110 (5%), and missing in 8/110 (7%) test scans. No estimate was misplaced by more than one slice. CONCLUSIONS: We demonstrated a robust automated approach for annotating the medial clavicular epiphyseal cartilages. This enables training and testing of deep neural networks for age assessment.

Prognostic value of baseline imaging and clinical features in patients with advanced hepatocellular carcinoma.

AIMS: To investigate the prognostic value of baseline imaging features for overall survival (OS) and liver decompensation (LD) in patients with hepatocellular carcinoma (HCC). DESIGN: Patients with advanced HCC from the SORAMIC trial were evaluated in this post hoc analysis. Several radiological imaging features were collected from baseline computed tomography (CT) and magnetic resonance imaging (MRI) imaging, besides clinical values. The prognostic value of these features for OS and LD (grade 2 bilirubin increase) was quantified with univariate Cox proportional hazard models and multivariate Least Absolute Shrinkage and Selection Operator (LASSO) regression. RESULTS: Three hundred and seventy-six patients were included in this study. The treatment arm was not correlated with OS. LASSO showed satellite lesions, atypical HCC, peritumoral arterial enhancement, larger tumour size, higher albumin-bilirubin (ALBI) score, liver-spleen ratio <1.5, ascites, pleural effusion and higher bilirubin values were predictors of worse OS, and higher relative liver enhancement, smooth margin and capsule were associated with better OS. LASSO analysis for LD showed satellite lesions, peritumoral hypointensity in hepatobiliary phase, high ALBI score, higher bilirubin values and ascites were predictors of LD, while randomisation to sorafenib arm was associated with lower LD. CONCLUSIONS: Imaging features showing aggressive tumour biology and poor liver function, in addition to clinical parameters, can serve as imaging biomarkers for OS and LD in patients receiving sorafenib and selective internal radiation therapy for HCC.

Deep learning in CT colonography: differentiating premalignant from benign colorectal polyps.

OBJECTIVES: To investigate the differentiation of premalignant from benign colorectal polyps detected by CT colonography using deep learning. METHODS: In this retrospective analysis of an average risk colorectal cancer screening sample, polyps of all size categories and morphologies were manually segmented on supine and prone CT colonography images and classified as premalignant (adenoma) or benign (hyperplastic polyp or regular mucosa) according to histopathology. Two deep learning models SEG and noSEG were trained on 3D CT colonography image subvolumes to predict polyp class, and model SEG was additionally trained with polyp segmentation masks. Diagnostic performance was validated in an independent external multicentre test sample. Predictions were analysed with the visualisation technique Grad-CAM++. RESULTS: The training set consisted of 107 colorectal polyps in 63 patients (mean age: 63 ± 8 years, 40 men) comprising 169 polyp segmentations. The external test set included 77 polyps in 59 patients comprising 118 polyp segmentations. Model SEG achieved a ROC-AUC of 0.83 and 80% sensitivity at 69% specificity for differentiating premalignant from benign polyps. Model noSEG yielded a ROC-AUC of 0.75, 80% sensitivity at 44% specificity, and an average Grad-CAM++ heatmap score of ≥ 0.25 in 90% of polyp tissue. CONCLUSIONS: In this proof-of-concept study, deep learning enabled the differentiation of premalignant from benign colorectal polyps detected with CT colonography and the visualisation of image regions important for predictions. The approach did not require polyp segmentation and thus has the potential to facilitate the identification of high-risk polyps as an automated second reader. KEY POINTS: • Non-invasive deep learning image analysis may differentiate premalignant from benign colorectal polyps found in CT colonography scans. • Deep learning autonomously learned to focus on polyp tissue for predictions without the need for prior polyp segmentation by experts. • Deep learning potentially improves the diagnostic accuracy of CT colonography in colorectal cancer screening by allowing for a more precise selection of patients who would benefit from endoscopic polypectomy, especially for patients with polyps of 6-9 mm size.

Machine Learning-based Differentiation of Benign and Premalignant Colorectal Polyps Detected with CT Colonography in an Asymptomatic Screening Population: A Proof-of-Concept Study.

Background CT colonography does not enable definite differentiation between benign and premalignant colorectal polyps. Purpose To perform machine learning-based differentiation of benign and premalignant colorectal polyps detected with CT colonography in an average-risk asymptomatic colorectal cancer screening sample with external validation using radiomics. Materials and Methods In this secondary analysis of a prospective trial, colorectal polyps of all size categories and morphologies were manually segmented on CT colonographic images and were classified as benign (hyperplastic polyp or regular mucosa) or premalignant (adenoma) according to the histopathologic reference standard. Quantitative image features characterizing shape (n = 14), gray level histogram statistics (n = 18), and image texture (n = 68) were extracted from segmentations after applying 22 image filters, resulting in 1906 feature-filter combinations. Based on these features, a random forest classification algorithm was trained to predict the individual polyp character. Diagnostic performance was validated in an external test set. Results The random forest model was fitted using a training set consisting of 107 colorectal polyps in 63 patients (mean age, 63 years ± 8 [standard deviation]; 40 men) comprising 169 segmentations on CT colonographic images. The external test set included 77 polyps in 59 patients comprising 118 segmentations. Random forest analysis yielded an area under the receiver operating characteristic curve of 0.91 (95% CI: 0.85, 0.96), a sensitivity of 82% (65 of 79) (95% CI: 74%, 91%), and a specificity of 85% (33 of 39) (95% CI: 72%, 95%) in the external test set. In two subgroup analyses of the external test set, the area under the receiver operating characteristic curve was 0.87 in the size category of 6-9 mm and 0.90 in the size category of 10 mm or larger. The most important image feature for decision making (relative importance of 3.7%) was quantifying first-order gray level histogram statistics. Conclusion In this proof-of-concept study, machine learning-based image analysis enabled noninvasive differentiation of benign and premalignant colorectal polyps with CT colonography. © RSNA, 2021 Online supplemental material is available for this article.

Sources of systematic error in DCE-MRI estimation of low-level blood-brain barrier leakage.

PURPOSE: Dynamic contrast-enhanced (DCE) -MRI with Patlak model analysis is increasingly used to quantify low-level blood-brain barrier (BBB) leakage in studies of pathophysiology. We aimed to investigate systematic errors due to physiological, experimental, and modeling factors influencing quantification of the permeability-surface area product PS and blood plasma volume vp , and to propose modifications to reduce the errors so that subtle differences in BBB permeability can be accurately measured. METHODS: Simulations were performed to predict the effects of potential sources of systematic error on conventional PS and vp quantification: restricted BBB water exchange, reduced cerebral blood flow, arterial input function (AIF) delay and B1+ error. The impact of targeted modifications to the acquisition and processing were evaluated, including: assumption of fast versus no BBB water exchange, bolus versus slow injection of contrast agent, exclusion of early data from model fitting and B1+ correction. The optimal protocol was applied in a cohort of recent mild ischaemic stroke patients. RESULTS: Simulation results demonstrated substantial systematic errors due to the factors investigated (absolute PS error ≤ 4.48 × 10-4 min-1 ). However, these were reduced (≤0.56 × 10-4 min-1 ) by applying modifications to the acquisition and processing pipeline. Processing modifications also had substantial effects on in-vivo normal-appearing white matter PS estimation (absolute change ≤ 0.45 × 10-4 min-1 ). CONCLUSION: Measuring subtle BBB leakage with DCE-MRI presents unique challenges and is affected by several confounds that should be considered when acquiring or interpreting such data. The evaluated modifications should improve accuracy in studies of neurodegenerative diseases involving subtle BBB breakdown.

Pneumothorax detection in chest radiographs: optimizing artificial intelligence system for accuracy and confounding bias reduction using in-image annotations in algorithm training.

OBJECTIVES: Diagnostic accuracy of artificial intelligence (AI) pneumothorax (PTX) detection in chest radiographs (CXR) is limited by the noisy annotation quality of public training data and confounding thoracic tubes (TT). We hypothesize that in-image annotations of the dehiscent visceral pleura for algorithm training boosts algorithm's performance and suppresses confounders. METHODS: Our single-center evaluation cohort of 3062 supine CXRs includes 760 PTX-positive cases with radiological annotations of PTX size and inserted TTs. Three step-by-step improved algorithms (differing in algorithm architecture, training data from public datasets/clinical sites, and in-image annotations included in algorithm training) were characterized by area under the receiver operating characteristics (AUROC) in detailed subgroup analyses and referenced to the well-established "CheXNet" algorithm. RESULTS: Performances of established algorithms exclusively trained on publicly available data without in-image annotations are limited to AUROCs of 0.778 and strongly biased towards TTs that can completely eliminate algorithm's discriminative power in individual subgroups. Contrarily, our final "algorithm 2" which was trained on a lower number of images but additionally with in-image annotations of the dehiscent pleura achieved an overall AUROC of 0.877 for unilateral PTX detection with a significantly reduced TT-related confounding bias. CONCLUSIONS: We demonstrated strong limitations of an established PTX-detecting AI algorithm that can be significantly reduced by designing an AI system capable of learning to both classify and localize PTX. Our results are aimed at drawing attention to the necessity of high-quality in-image localization in training data to reduce the risks of unintentionally biasing the training process of pathology-detecting AI algorithms. KEY POINTS: • Established pneumothorax-detecting artificial intelligence algorithms trained on public training data are strongly limited and biased by confounding thoracic tubes. • We used high-quality in-image annotated training data to effectively boost algorithm performance and suppress the impact of confounding thoracic tubes. • Based on our results, we hypothesize that even hidden confounders might be effectively addressed by in-image annotations of pathology-related image features.

Artificial Intelligence Algorithm Detecting Lung Infection in Supine Chest Radiographs of Critically Ill Patients With a Diagnostic Accuracy Similar to Board-Certified Radiologists.

OBJECTIVES: Interpretation of lung opacities in ICU supine chest radiographs remains challenging. We evaluated a prototype artificial intelligence algorithm to classify basal lung opacities according to underlying pathologies. DESIGN: Retrospective study. The deep neural network was trained on two publicly available datasets including 297,541 images of 86,876 patients. PATIENTS: One hundred sixty-six patients received both supine chest radiograph and CT scans (reference standard) within 90 minutes without any intervention in between. MEASUREMENTS AND MAIN RESULTS: Algorithm accuracy was referenced to board-certified radiologists who evaluated supine chest radiographs according to side-separate reading scores for pneumonia and effusion (0 = absent, 1 = possible, and 2 = highly suspected). Radiologists were blinded to the supine chest radiograph findings during CT interpretation. Performances of radiologists and the artificial intelligence algorithm were quantified by receiver-operating characteristic curve analysis. Diagnostic metrics (sensitivity, specificity, positive predictive value, negative predictive value, and accuracy) were calculated based on different receiver-operating characteristic operating points. Regarding pneumonia detection, radiologists achieved a maximum diagnostic accuracy of up to 0.87 (95% CI, 0.78-0.93) when considering only the supine chest radiograph reading score 2 as positive for pneumonia. Radiologist's maximum sensitivity up to 0.87 (95% CI, 0.76-0.94) was achieved by additionally rating the supine chest radiograph reading score 1 as positive for pneumonia and taking previous examinations into account. Radiologic assessment essentially achieved nonsignificantly higher results compared with the artificial intelligence algorithm: artificial intelligence-area under the receiver-operating characteristic curve of 0.737 (0.659-0.815) versus radiologist's area under the receiver-operating characteristic curve of 0.779 (0.723-0.836), diagnostic metrics of receiver-operating characteristic operating points did not significantly differ. Regarding the detection of pleural effusions, there was no significant performance difference between radiologist's and artificial intelligence algorithm: artificial intelligence-area under the receiver-operating characteristic curve of 0.740 (0.662-0.817) versus radiologist's area under the receiver-operating characteristic curve of 0.698 (0.646-0.749) with similar diagnostic metrics for receiver-operating characteristic operating points. CONCLUSIONS: Considering the minor level of performance differences between the algorithm and radiologists, we regard artificial intelligence as a promising clinical decision support tool for supine chest radiograph examinations in the clinical routine with high potential to reduce the number of missed findings in an artificial intelligence-assisted reading setting.

MITK-ModelFit: A generic open-source framework for model fits and their exploration in medical imaging - design, implementation and application on the example of DCE-MRI.

BACKGROUND: Many medical imaging techniques utilize fitting approaches for quantitative parameter estimation and analysis. Common examples are pharmacokinetic modeling in dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI)/computed tomography (CT), apparent diffusion coefficient calculations and intravoxel incoherent motion modeling in diffusion-weighted MRI and Z-spectra analysis in chemical exchange saturation transfer MRI. Most available software tools are limited to a special purpose and do not allow for own developments and extensions. Furthermore, they are mostly designed as stand-alone solutions using external frameworks and thus cannot be easily incorporated natively in the analysis workflow. RESULTS: We present a framework for medical image fitting tasks that is included in the Medical Imaging Interaction Toolkit MITK, following a rigorous open-source, well-integrated and operating system independent policy. Software engineering-wise, the local models, the fitting infrastructure and the results representation are abstracted and thus can be easily adapted to any model fitting task on image data, independent of image modality or model. Several ready-to-use libraries for model fitting and use-cases, including fit evaluation and visualization, were implemented. Their embedding into MITK allows for easy data loading, pre- and post-processing and thus a natural inclusion of model fitting into an overarching workflow. As an example, we present a comprehensive set of plug-ins for the analysis of DCE MRI data, which we validated on existing and novel digital phantoms, yielding competitive deviations between fit and ground truth. CONCLUSIONS: Providing a very flexible environment, our software mainly addresses developers of medical imaging software that includes model fitting algorithms and tools. Additionally, the framework is of high interest to users in the domain of perfusion MRI, as it offers feature-rich, freely available, validated tools to perform pharmacokinetic analysis on DCE MRI data, with both interactive and automatized batch processing workflows.

Quantifying blood-brain barrier leakage in small vessel disease: Review and consensus recommendations.

Cerebral small vessel disease (cSVD) comprises pathological processes of the small vessels in the brain that may manifest clinically as stroke, cognitive impairment, dementia, or gait disturbance. It is generally accepted that endothelial dysfunction, including blood-brain barrier (BBB) failure, is pivotal in the pathophysiology. Recent years have seen increasing use of imaging, primarily dynamic contrast-enhanced magnetic resonance imaging, to assess BBB leakage, but there is considerable variability in the approaches and findings reported in the literature. Although dynamic contrast-enhanced magnetic resonance imaging is well established, challenges emerge in cSVD because of the subtle nature of BBB impairment. The purpose of this work, authored by members of the HARNESS Initiative, is to provide an in-depth review and position statement on magnetic resonance imaging measurement of subtle BBB leakage in clinical research studies, with aspects requiring further research identified. We further aim to provide information and consensus recommendations for new investigators wishing to study BBB failure in cSVD and dementia.

Detection of pulmonary embolism with free-breathing dynamic contrast-enhanced MRI.

PURPOSE: To evaluate the use of dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI) during free breathing for the detection of acute pulmonary embolism (PE). MATERIALS AND METHODS: Eighteen subjects underwent free-breathing DCE MRI at 1.5T, eight of whom were patients with acute PE, as confirmed by routine computed tomography pulmonary angiography (CTPA). The remaining 10 subjects were healthy volunteers with no history or signs of pulmonary disease. From all DCE MRI data, maps of relative signal enhancement were calculated and assessed for the presence or absence of perfusion defects in each lung by two readers. Interreader variability, sensitivity, and specificity of free-breathing DCE MRI for the detection of PE were calculated using CTPA as the gold standard. RESULTS: Of the 16 patient's lungs, 15 were affected by acute PE according to CTPA. In patients and volunteers, DCE MRI sensitivity was 93% and 87% for readers 1 and 2, with specificities of 95% and 90%, respectively. Interreader agreement was substantial, with κ = 0.77 (95% confidence interval: 0.44-1.0). CONCLUSION: Free-breathing DCE MRI may have potential use for the assessment of PE, and does not require patient cooperation in breath-holding.

Tracer kinetic modeling in myocardial perfusion quantification using MRI.

PURPOSE: To investigate and compare several quantification methods of myocardial perfusion measurements, paying special attention to the relation between the techniques and the required measurement duration. METHODS: Seven patients underwent contrast-enhanced rest and stress cardiac perfusion measurements at 3T. Three slices were acquired in each patient and were divided into 16 segments, leading to 112 rest and stress data curves, which were analyzed using various tracer kinetic models as well as a model-free deconvolution. Plasma flow, plasma volume, and myocardial perfusion reserve were analyzed for the complete acquisition as well as for the first pass data only. RESULTS: Deconvolution analysis yielded stable results for both rest and stress analysis, while Fermi and one compartment models agree well for first pass data (rest measurements only) and prolonged data acquisition (stress measurements only). More complex models do not yield satisfactory results for the short measurement times investigated in this study. CONCLUSIONS: When performing MRI-based quantification of myocardial perfusion, care must be taken that the method used is appropriate for the time frame under investigation. When a numerical deconvolution is used instead of tracer kinetic models, more stable results are obtained.

Reducing tumor growth and angiogenesis using a triple therapy measured with Contrast-enhanced ultrasound (CEUS).

BACKGROUND: To evaluate the in vivo response by detecting the anti-angiogenic and invasion-inhibiting effects of a triple-combination-therapy in an experimental-small-animal-squamous-cell-carcinoma-model using the "flash-replenishment" (FR) method to assess tissue hemodynamics via contrast-enhanced-ultrasound (CEUS). METHODS: Human hypopharynx-carcinoma-cells were subcutaneously injected into the left flank of 22-female-athymic-nude-rats. After seven days of subcutaneous tumor growth, FR-measurements were performed on each rat. Treatment-group and control-group were treated every day for a period of one week, with the treatment-group receiving solvents containing a triple therapy of Upamostat®, Celecoxib® and Ilomastat® and the control-group solvents only. On day seven, follow-up measurements were performed using the same measurement protocol to assess the effects of the triple therapy. VueBox® was used to quantify the kinetic parameters and additional immunohistochemistry analyses were performed for comparison with and validation of the CEUS results against established methods (Proliferation/Ki-67, vascularization/CD31, apoptosis/caspase3). RESULTS: Compared to the control-group, the treatment-group that received the triple-therapy resulted in a reduction of tumor growth by 48.6% in size. Likewise, the immunohistochemistry results showed significant decreases in tumor proliferation and vascularization in the treatment-group in comparison to the control-group of 26%(p ≤ 0.05) and 32.2%(p ≤ 0.05) respectively. Correspondingly, between the baseline and follow-up measurements, the therapy-group was associated with a significant(p ≤ 0.01) decrease in the relative-Blood-Volume(rBV) in both the whole tumor(wt) and hypervascular tumor(ht) areas (p ≤ 0.01), while the control-group was associated with a significant (p ≤ 0.01) increase of the rBV in the wt area and a non-significant increase (p ≤ 0.16) in the ht area. The mean-transit-time (mTT) of the wt and the ht areas showed a significant increase (p ≤ 0.01) in the follow-up measurements in the therapy group. CONCLUSION: The triple-therapy is feasible and effective in reducing both tumor growth and vascularization. In particular, compared with the placebo-group, the triple-therapy-group resulted in a reduction in tumor growth of 48.6% in size when assessed by CEUS and a significant reduction in the number of vessels in the tumor of 32% as assessed by immunohistochemistry. As the immunohistochemistry supports the CEUS findings, CEUS using the "flash replenishment"(FR) method appears to provide a useful assessment of the anti-angiogenic and invasion-inhibiting effects of a triple combination therapy.

Resting-state networks in healthy adult subjects: a comparison between a 32-element and an 8-element phased array head coil at 3.0 Tesla.

BACKGROUND: Little research exists on the influence of a magnetic resonance imaging (MRI) head coil's channel count on measured resting-state functional connectivity. PURPOSE: To compare a 32-element (32ch) and an 8-element (8ch) phased array head coil with respect to their potential to detect functional connectivity within resting-state networks. MATERIAL AND METHODS: Twenty-six healthy adults (mean age, 21.7 years; SD, 2.1 years) underwent resting-state functional MRI at 3.0 Tesla with both coils using equal standard imaging parameters and a counterbalanced design. Independent component analysis (ICA) at different model orders and a dual regression approach were performed. Voxel-wise non-parametric statistical between-group contrasts were determined using permutation-based non-parametric inference. RESULTS: Phantom measurements demonstrated a generally higher image signal-to-noise ratio using the 32ch head coil. However, the results showed no significant differences between corresponding resting-state networks derived from both coils (p < 0.05, FWE-corrected). CONCLUSION: Using the identical standard acquisition parameters, the 32ch head coil does not offer any significant advantages in detecting ICA-based functional connectivity within RSNs.

DCE-MRI biomarkers for monitoring an anti-angiogenic triple combination therapy in experimental hypopharynx carcinoma xenografts with immunohistochemical validation.

BACKGROUND: Novel anti-angiogenic treatments are increasingly complementing established cancer therapy strategies in head and neck tumors. Contrast-enhanced magnetic resonance imaging (MRI) can be applied for early and non-invasive therapy monitoring by non-invasive quantitative assessment of tumor microcirculation as in vivo imaging biomarkers of therapy response. PURPOSE: To monitor the anti-angiogenic effects of a novel combination therapy on experimental head and neck squamous cell carcinomas (HNSCC) with dynamic contrast-enhanced (DCE)-MRI. MATERIAL AND METHODS: Athymic rats (n = 18) with subcutaneous HNSCC xenografts were investigated by DCE-MRI before and after 7 days of a daily triple therapy regimen combining the COX-II-inhibitor celecoxib, the matrix-metalloproteinase-inhibitor GM6001, and the uPA-inhibitor upamostat. Quantitative measurements of tumor blood flow (tBF), tumor blood volume (tBV), and permeability-surface area product (PS) were calculated and validated by immunohistochemistry. RESULTS: Mean tBF and tBV in triple-therapy animals decreased significantly from day 0 to day 7 (tBF, 41.0 ± 14.2 to 20.4 ± 5.7 mL/100 mL/min; P < 0.01; tBV, 17.7 ± 3.9 to 7.5 ± 3.3%; P < 0.01). No significant effects on PS were observed in either group (P > 0.05). Immunohistochemical analysis showed a significantly lower tumor vascularity in the therapy group than in the control group (CD31), significantly fewer Ki-67+ proliferating tumor cells and significantly more Capase-3+ apoptotic tumor cells (P < 0.05). Significant (P < 0.05) correlations were observed between tBF/tBV and CD31 (tBF, r = 0.84; tBV, r = 0.70), tBV and Ki-67 (r = 0.62), as well as tBF and caspase-3 (r = -0.64). CONCLUSION: DCE-MRI may be a suitable tool for the non-invasive monitoring of the anti-vascular effects of this innovative triple therapy regimen with potential for clinical translation.

Out-of-distribution detection with in-distribution voting using the medical example of chest x-ray classification.

BACKGROUND: Deep learning models are being applied to more and more use cases with astonishing success stories, but how do they perform in the real world? Models are typically tested on specific cleaned data sets, but when deployed in the real world, the model will encounter unexpected, out-of-distribution (OOD) data. PURPOSE: To investigate the impact of OOD radiographs on existing chest x-ray classification models and to increase their robustness against OOD data. METHODS: The study employed the commonly used chest x-ray classification model, CheXnet, trained on the chest x-ray 14 data set, and tested its robustness against OOD data using three public radiography data sets: IRMA, Bone Age, and MURA, and the ImageNet data set. To detect OOD data for multi-label classification, we proposed in-distribution voting (IDV). The OOD detection performance is measured across data sets using the area under the receiver operating characteristic curve (AUC) analysis and compared with Mahalanobis-based OOD detection, MaxLogit, MaxEnergy, self-supervised OOD detection (SS OOD), and CutMix. RESULTS: Without additional OOD detection, the chest x-ray classifier failed to discard any OOD images, with an AUC of 0.5. The proposed IDV approach trained on ID (chest x-ray 14) and OOD data (IRMA and ImageNet) achieved, on average, 0.999 OOD AUC across the three data sets, surpassing all other OOD detection methods. Mahalanobis-based OOD detection achieved an average OOD detection AUC of 0.982. IDV trained solely with a few thousand ImageNet images had an AUC 0.913, which was considerably higher than MaxLogit (0.726), MaxEnergy (0.724), SS OOD (0.476), and CutMix (0.376). CONCLUSIONS: The performance of all tested OOD detection methods did not translate well to radiography data sets, except Mahalanobis-based OOD detection and the proposed IDV method. Consequently, training solely on ID data led to incorrect classification of OOD images as ID, resulting in increased false positive rates. IDV substantially improved the model's ID classification performance, even when trained with data that will not occur in the intended use case or test set (ImageNet), without additional inference overhead or performance decrease in the target classification. The corresponding code is available at https://gitlab.lrz.de/IP/a-knee-cannot-have-lung-disease.

Language model-based labeling of German thoracic radiology reports.

The aim of this study was to explore the potential of weak supervision in a deep learning-based label prediction model. The goal was to use this model to extract labels from German free-text thoracic radiology reports on chest X-ray images and for training chest X-ray classification models.The proposed label extraction model for German thoracic radiology reports uses a German BERT encoder as a backbone and classifies a report based on the CheXpert labels. For investigating the efficient use of manually annotated data, the model was trained using manual annotations, weak rule-based labels, and both. Rule-based labels were extracted from 66071 retrospectively collected radiology reports from 2017-2021 (DS 0), and 1091 reports from 2020-2021 (DS 1) were manually labeled according to the CheXpert classes. Label extraction performance was evaluated with respect to mention extraction, negation detection, and uncertainty detection by measuring F1 scores. The influence of the label extraction method on chest X-ray classification was evaluated on a pneumothorax data set (DS 2) containing 6434 chest radiographs with associated reports and expert diagnoses of pneumothorax. For this, DenseNet-121 models trained on manual annotations, rule-based and deep learning-based label predictions, and publicly available data were compared.The proposed deep learning-based labeler (DL) performed on average considerably stronger than the rule-based labeler (RB) for all three tasks on DS 1 with F1 scores of 0.938 vs. 0.844 for mention extraction, 0.891 vs. 0.821 for negation detection, and 0.624 vs. 0.518 for uncertainty detection. Pre-training on DS 0 and fine-tuning on DS 1 performed better than only training on either DS 0 or DS 1. Chest X-ray pneumothorax classification results (DS 2) were highest when trained with DL labels with an area under the receiver operating curve (AUC) of 0.939 compared to RB labels with an AUC of 0.858. Training with manual labels performed slightly worse than training with DL labels with an AUC of 0.934. In contrast, training with a public data set resulted in an AUC of 0.720.Our results show that leveraging a rule-based report labeler for weak supervision leads to improved labeling performance. The pneumothorax classification results demonstrate that our proposed deep learning-based labeler can serve as a substitute for manual labeling requiring only 1000 manually annotated reports for training. · The proposed deep learning-based label extraction model for German thoracic radiology reports performs better than the rule-based model.. · Training with limited supervision outperformed training with a small manually labeled data set.. · Using predicted labels for pneumothorax classification from chest radiographs performed equally to using manual annotations.. Wollek A, Haitzer P, Sedlmeyr T et al. Language modelbased labeling of German thoracic radiology reports. Fortschr Röntgenstr 2024; DOI 10.1055/a-2287-5054.