Search

Introduction: An increasing shortage of donor blood is expected, considering the demographic change in Germany. Due to the short shelf life and varying daily fluctuations in consumption, the storage of platelet concentrates (PCs) becomes challenging. This emphasizes the need for reliable prediction of needed PCs for the blood bank inventories. Therefore, the objective of this study was to evaluate multimodal data from multiple source systems within a hospital to predict the number of platelet transfusions in 3 days on a per-patient level. Methods: Data were collected from 25,190 (42% female and 58% male) patients between 2017 and 2021. For each patient, the number of received PCs, platelet count blood tests, drugs causing thrombocytopenia, acute platelet diseases, procedures, age, gender, and the period of a patient’s hospital stay were collected. Two models were trained on samples using a sliding window of 7 days as input and a day 3 target. The model predicts whether a patient will be transfused 3 days in the future. The model was trained with an excessive hyperparameter search using patient-level repeated 5-fold cross-validation to optimize the average macro F2-score. Results: The trained models were tested on 5,022 unique patients. The best-performing model has a specificity of 0.99, a sensitivity of 0.37, an area under the precision-recall curve score of 0.45, an MCC score of 0.43, and an F1-score of 0.43. However, the model does not generalize well for cases when the need for a platelet transfusion is recognized. Conclusion: A patient AI-based platelet forecast could improve logistics management and reduce blood product waste. In this study, we build the first model to predict patient individual platelet demand. To the best of our knowledge, we are the first to introduce this approach. Our model predicts the need for platelet units for 3 days in the future. While sensitivity underperforms, specificity performs reliably. The model may be of clinical use as a pretest for potential patients needing a platelet transfusion within the next 3 days. As sensitivity needs to be improved, further studies should introduce deep learning and wider patient characterization to the methodological multimodal, multisource data approach. Furthermore, a hospital-wide consumption of PCs could be derived from individual predictions.

Objective: Body tissue composition plays a crucial role in the multisystemic processes of advanced liver disease and has been shown to be influenced by transjugular intrahepatic portosystemic shunt (TIPS). A differentiated analysis of the various tissue compartments has not been performed until now. The purpose of this study was to evaluate the value of imaging biomarkers derived from automated body composition analysis (BCA) to predict clinical and functional outcome. Methods: A retrospective analysis of 56 patients undergoing TIPS procedure between 2013 and 2021 was performed. BCA on the base of pre-interventional CT examination was used to determine quantitative data as well as ratios of bone, muscle and fat masses. Furthermore, a BCA-derived sarcopenia marker was investigated. Regarding potential correlations between BCA imaging biomarkers and the occurrence of hepatic encephalopathy (HE) as well as 1-year survival, an exploratory analysis was conducted. Results: No BCA imaging biomarker was associated with the occurrence of HE after TIPS placement. However, there were significant differences in alive and deceased patients regarding the BCA-derived sarcopenia marker (alive: 1.60, deceased: 1.83, p = 0.046), ratios of intra- and intermuscular fat/skeletal volume (alive: 0.53, deceased: 0.31, p = 0.015) and intra- and intermuscular fat/muscle volume (alive: 0.21, deceased: 0.14, p = 0.031). Conclusion: A lower amount of intra- and intermuscular adipose tissue might have protective effects regarding liver derived complications and survival. Advances in knowledge: Precise characterization of body tissue components with automated BCA might provide prognostic information in patients with advanced liver disease undergoing TIPS procedure.

Introduction: Liver volumetry based on a CT-Scan is widely used for estimating liver volume before any liver resection, especially for pediatric living donor liver transplantation (LDLT). The "One-to-one" conversion rule for liver volume to weight has been widely adopted worldwide for many years. However, most recent analyses discuss this approach. Methods: The study retrospectively included consecutive donors undergoing left lateral hepatectomy for pediatric LDLT between December 2008 and September 2020. All donors were healthy adults who met the evaluation criteria for pediatric LDLT and underwent a preoperative contrast-enhanced CT scan. Manual segmentation of the left lateral liver lobe for volume (GV) estimation and intraoperative measurement of an actual graft weight (AGW) was performed. A relationship between estimated GV and AGW was analyzed. Results: 94 living liver donors were included in the study. The mean AGW was approximately 283.4 ± 68.5 grams, and the mean GV was 244.9 ± 63.86 ml. There was a strong correlation between the GV and AGW (r = 0.804, p < 0.001). The Bland-Altman analysis revealed an interreader agreement of 38.0 ± 97.25 and an intraclass correlation coefficient showed an almost excellent agreement (0.840, p < 0.001). The conversion formula for calculating graft weight (GW) based on computed tomography volumetry was determined based on regression analysis: 0.88 x GV + 41.63. Conclusion: Estimating left liver GW using "one-to-one" rule is subject to measurable variability in calculated graft weights and tends to underestimate it. Instead, a conversion formula should be used to determine donor GW more accurately.

Detection and validation of the chest X-ray view position with use of convolutional neural networks to improve meta-information for data cleaning within a hospital data infrastructure. Within this paper we developed a convolutional neural network which automatically detects the anteroposterior and posteroanterior view position of a chest radiograph. We trained two different network architectures (VGG variant and ResNet-34) with data published by the RSNA (26 684 radiographs, class distribution 46 % AP, 54 % PA) and validated these on a self-compiled dataset with data from the University Hospital Essen (4507, radiographs, class distribution 55 % PA, 45 % AP) labeled by a human reader. For visualization and better understanding of the network predictions, a Grad-CAM was generated for each network decision. The network results were evaluated based on the accuracy, the area under the curve (AUC), and the F1-score against the human reader labels. Also a final performance comparison between model predictions and DICOM labels was performed. The ensemble models reached accuracy and F1-scores greater than 95 %. The AUC reaches more than 0.99 for the ensemble models. The Grad-CAMs provide insight as to which anatomical structures contributed to a decision by the networks which are comparable with the ones a radiologist would use. Furthermore, the trained models were able to generalize over mislabeled examples, which was found by comparing the human reader labels to the predicted labels as well as the DICOM labels. The results show that certain incorrectly entered meta-information of radiological images can be effectively corrected by deep learning in order to increase data quality in clinical application as well as in research. · The predictions for both view positions are accurate with respect to external validation data.. · The networks based their decisions on anatomical structures and key points that were in-line with prior knowledge and human understanding.. · Final models were able to detect labeling errors within the test dataset..· Hosch R, Kroll L, Nensa F et al. Differentiation Between Anteroposterior and Posteroanterior Chest X-Ray View Position With Convolutional Neural Networks. Fortschr Röntgenstr 2021; 193: 168 - 176.ZIEL: Detektion der Röntgen-Thorax-Aufnahmeposition anhand von Convolutional Neural Networks zur Verbesserung und Bereinigung von Metainformationen innerhalb der Dateninfrastruktur eines Krankenhauses. Innerhalb dieser Studie wurde ein Convolutional Neural Network entwickelt, das automatisch die verwendete Anterior-posterior- bzw. Posterior-anterior-Aufnahmeprojektion einer Röntgen-Thoraxaufnahme erkennt. Es wurden 2 unterschiedliche Netzwerkarchitekturen (VGG Variante und ResNet-34) auf Basis von Daten der RSNA (26 684 Röntgenaufnahmen, Klassenverteilung: 46 % AP, 54 % PA) trainiert und anschließend auf einem zusammengestellten hauseigenen Datensatz (Verwendung von manuellen Labeln) aus dem Datenbestand des Universitätsklinikums Essen (4507 Röntgenaufnahmen, Klassenverteilung: 55 % PA, 45 % AP) getestet. Für eine bessere Nachvollziehbarkeit der getätigten Vorhersagen der Modelle wurde zudem für jede Vorhersage eine Grad-CAM generiert. Die Resultate der Modelle wurden anhand der Accuracy, der Area under the Curve (AUC) und dem F1-Score berechnet auf Basis des Abgleichs der manuellen Label. Abschließend wurde zudem die Genauigkeit der Modellvorhersagen und der DICOM-Label anhand des Vergleichs mit den manuellen Labeln berechnet. Die zusammengefassten Modelle erreichten Accuracy- und F1-Score-Werte von mehr als 95 %. Alle Modelle erreichten eine AUC von über 0,99. Die generierten Grad-CAMSs zeigen, dass die Modelle relevante anatomische Referenzpunkte für ihre Vorhersage nutzen, die auch ein Radiologe für eine Unterscheidung heranziehen würde. Zudem zeigen die antrainierten Modelle die Fähigkeit zur Generalisierung, da diese auch falsch gekennzeichnete Röntgenbilder richtig einordnen können, was durch den Vergleich der manuellen Label mit den jeweiligen Modellvorhersagen und den DICOM-Labeln ersichtlich wurde. Die Resultate zeigen, dass falsch eingetragene Metainformationen innerhalb der radiologischen Bildgebung effektiv durch den Einsatz von Deep Learning korrigiert und somit die Datenqualität sowohl für die klinische Anwendung als auch für die Forschung erhöht werden können. · Die trainierten Modelle erzielen akkurate Vorhersagen auf externen Validierungsdaten.. · Die Netzwerke treffen ihre Vorhersagen basierend auf anatomischen Strukturen und Referenzpunkten, die mit dem menschlichen Fachwissen übereinstimmen.. · Die finalen Modelle konnten Label-Fehler in dem Testdatensatz finden..

Purpose Both digital positron emission tomography (PET) detector technologies and artificial intelligence based image post-reconstruction methods allow to reduce the PET acquisition time while maintaining diagnostic quality. The aim of this study was to acquire ultra-low-count fluorodeoxyglucose (FDG) ExtremePET images on a digital PET/computed tomography (CT) scanner at an acquisition time comparable to a CT scan and to generate synthetic full-dose PET images using an artificial neural network. Methods This is a prospective, single-arm, single-center phase I/II imaging study. A total of 587 patients were included. For each patient, a standard and an ultra-low-count FDG PET/CT scan (whole-body acquisition time about 30 s) were acquired. A modified pix2pixHD deep-learning network was trained employing 387 data sets as training and 200 as test cohort. Three models (PET-only and PET/CT with or without group convolution) were compared. Detectability and quantification were evaluated. Results The PET/CT input model with group convolution performed best regarding lesion signal recovery and was selected for detailed evaluation. Synthetic PET images were of high visual image quality; mean absolute lesion SUVmax (maximum standardized uptake value) difference was 1.5. Patient-based sensitivity and specificity for lesion detection were 79% and 100%, respectively. Not-detected lesions were of lower tracer uptake and lesion volume. In a matched-pair comparison, patient-based (lesion-based) detection rate was 89% (78%) for PERCIST (PET response criteria in solid tumors)-measurable and 36% (22%) for non PERCIST-measurable lesions. Conclusion Lesion detectability and lesion quantification were promising in the context of extremely fast acquisition times. Possible application scenarios might include re-staging of late-stage cancer patients, in whom assessment of total tumor burden can be of higher relevance than detailed evaluation of small and low-uptake lesions.

ZusammenfassungRadiomics ist eine Methode der medizinischen Bildanalyse, bei der quantitative Merkmale aus Bilddaten extrahiert und mittels Machine Learning zu prädiktiven Modellen weiterverarbeitet werden. Ziel dieser Arbeit ist es, die technischen Grundlagen von Radiomics und mögliche klinische Anwendungen unter besonderer Berücksichtigung nuklearmedizinischer Daten zu erläutern. Dabei wird zunächst die klassische Radiomics-Methode besprochen, welche auf einer exakten Segmentierung der zu analysierenden Pathologie beruht und bei der die Features manuell definiert werden müssen. Anschließend wird auf das noch wenig verbreitete, allerdings vielversprechende Deep Learning basierte Radiomics eingegangen, dessen Vorteile darin liegen, dass ausschließlich datengetrieben gearbeitet wird und daher weder exakte Segmentierungen noch manuelle Definitionen der Features benötigt werden. Abschließend werden einige Anwendungen von Radiomics besprochen, die zukünftig im klinischen Alltag eine Rolle spielen könnten.

Objectives To reduce the dose of intravenous iodine-based contrast media (ICM) in CT through virtual contrast-enhanced images using generative adversarial networks. Methods Dual-energy CTs in the arterial phase of 85 patients were randomly split into an 80/20 train/test collective. Four different generative adversarial networks (GANs) based on image pairs, which comprised one image with virtually reduced ICM and the original full ICM CT slice, were trained, testing two input formats (2D and 2.5D) and two reduced ICM dose levels (−50% and −80%). The amount of intravenous ICM was reduced by creating virtual non-contrast series using dual-energy and adding the corresponding percentage of the iodine map. The evaluation was based on different scores (L1 loss, SSIM, PSNR, FID), which evaluate the image quality and similarity. Additionally, a visual Turing test (VTT) with three radiologists was used to assess the similarity and pathological consistency. Results The −80% models reach an SSIM of > 98%, PSNR of > 48, L1 of between 7.5 and 8, and an FID of between 1.6 and 1.7. In comparison, the −50% models reach a SSIM of > 99%, PSNR of > 51, L1 of between 6.0 and 6.1, and an FID between 0.8 and 0.95. For the crucial question of pathological consistency, only the 50% ICM reduction networks achieved 100% consistency, which is required for clinical use. Conclusions The required amount of ICM for CT can be reduced by 50% while maintaining image quality and diagnostic accuracy using GANs. Further phantom studies and animal experiments are required to confirm these initial results. Key Points • The amount of contrast media required for CT can be reduced by 50% using generative adversarial networks. • Not only the image quality but especially the pathological consistency must be evaluated to assess safety. • A too pronounced contrast media reduction could influence the pathological consistency in our collective at 80%.

Objectives Body tissue composition is a long-known biomarker with high diagnostic and prognostic value not only in cardiovascular, oncological, and orthopedic diseases but also in rehabilitation medicine or drug dosage. In this study, the aim was to develop a fully automated, reproducible, and quantitative 3D volumetry of body tissue composition from standard CT examinations of the abdomen in order to be able to offer such valuable biomarkers as part of routine clinical imaging. Methods Therefore, an in-house dataset of 40 CTs for training and 10 CTs for testing were fully annotated on every fifth axial slice with five different semantic body regions: abdominal cavity, bones, muscle, subcutaneous tissue, and thoracic cavity. Multi-resolution U-Net 3D neural networks were employed for segmenting these body regions, followed by subclassifying adipose tissue and muscle using known Hounsfield unit limits. Results The Sørensen Dice scores averaged over all semantic regions was 0.9553 and the intra-class correlation coefficients for subclassified tissues were above 0.99. Conclusions Our results show that fully automated body composition analysis on routine CT imaging can provide stable biomarkers across the whole abdomen and not just on L3 slices, which is historically the reference location for analyzing body composition in the clinical routine. Key Points • Our study enables fully automated body composition analysis on routine abdomen CT scans. • The best segmentation models for semantic body region segmentation achieved an averaged Sørensen Dice score of 0.9553. • Subclassified tissue volumes achieved intra-class correlation coefficients over 0.99.

Pediatric endocrinologists regularly order radiographs of the left hand to estimate the degree of bone maturation in order to assess their patients for advanced or delayed growth, physical development, and to monitor consecutive therapeutic measures. The reading of such images is a labor-intensive task that requires a lot of experience and is normally performed by highly trained experts like pediatric radiologists. In this paper we build an automated system for pediatric bone age estimation that mimics and accelerates the workflow of the radiologist without breaking it. The complete system is based on two neural network based models: on the one hand a detector network, which identifies the ossification areas, on the other hand gender and region specific regression networks, which estimate the bone age from the detected areas. With a small annotated dataset an ossification area detection network can be trained, which is stable enough to work as part of a multi-stage approach. Furthermore, our system achieves competitive results on the RSNA Pediatric Bone Age Challenge test set with an average error of 4.56 months. In contrast to other approaches, especially purely encoder-based architectures, our two-stage approach provides self-explanatory results. By detecting and evaluating the individual ossification areas, thus simulating the workflow of the Tanner-Whitehouse procedure, the results are interpretable for a radiologist.

The original version of this article, published on 18 September 2020, unfortunately contained a mistake.

Importance: Mammography screening currently relies on subjective human interpretation. Artificial intelligence (AI) advances could be used to increase mammography screening accuracy by reducing missed cancers and false positives. Objective: To evaluate whether AI can overcome human mammography interpretation limitations with a rigorous, unbiased evaluation of machine learning algorithms. Design, Setting, and Participants: In this diagnostic accuracy study conducted between September 2016 and November 2017, an international, crowdsourced challenge was hosted to foster AI algorithm development focused on interpreting screening mammography. More than 1100 participants comprising 126 teams from 44 countries participated. Analysis began November 18, 2016. Main Outcomes and Measurements: Algorithms used images alone (challenge 1) or combined images, previous examinations (if available), and clinical and demographic risk factor data (challenge 2) and output a score that translated to cancer yes/no within 12 months. Algorithm accuracy for breast cancer detection was evaluated using area under the curve and algorithm specificity compared with radiologists' specificity with radiologists' sensitivity set at 85.9% (United States) and 83.9% (Sweden). An ensemble method aggregating top-performing AI algorithms and radiologists' recall assessment was developed and evaluated. Results: Overall, 144 231 screening mammograms from 85 580 US women (952 cancer positive ≤12 months from screening) were used for algorithm training and validation. A second independent validation cohort included 166 578 examinations from 68 008 Swedish women (780 cancer positive). The top-performing algorithm achieved an area under the curve of 0.858 (United States) and 0.903 (Sweden) and 66.2% (United States) and 81.2% (Sweden) specificity at the radiologists' sensitivity, lower than community-practice radiologists' specificity of 90.5% (United States) and 98.5% (Sweden). Combining top-performing algorithms and US radiologist assessments resulted in a higher area under the curve of 0.942 and achieved a significantly improved specificity (92.0%) at the same sensitivity. Conclusions and Relevance: While no single AI algorithm outperformed radiologists, an ensemble of AI algorithms combined with radiologist assessment in a single-reader screening environment improved overall accuracy. This study underscores the potential of using machine learning methods for enhancing mammography screening interpretation. CA extern

Depression is ranked as the largest contributor to global disability and is also a major reason for suicide. Still, many individuals suffering from forms of depression are not treated for various reasons. Previous studies have shown that depression also has an effect on language usage and that many depressed individuals use social media platforms or the internet in general to get information or discuss their problems. This paper addresses the early detection of depression using machine learning models based on messages on a social platform. In particular, a convolutional neural network based on different word embeddings is evaluated and compared to a classification based on user-level linguistic metadata. An ensemble of both approaches is shown to achieve state-of-the-art results in a current early detection task. Furthermore, the currently popular ERDE score as metric for early detection systems is examined in detail and its drawbacks in the context of shared tasks are illustrated. A slightly modified metric is proposed and compared to the original score. Finally, a new word embedding was trained on a large corpus of the same domain as the described task and is evaluated as well., This work has been submitted to the IEEE and has been accepted for future publication in IEEE Transactions on Knowledge and Data Engineering. Copyright may be transferred without notice, after which this version may no longer be accessible. 14 pages, 3 figures, 7 tables

Early detection of depression based on written texts has become an important research area due to the rise of social media platforms and because many affected individuals are still not treated. During the eRisk task for early detection of depression at CLEF 2017, FHDO Biomedical Computer Science Group (BCSG) submitted results based on five text classification models. This paper builds upon this work to examine the task and especially the \(ERDE_o\) metric in further detail and to analyze how an additional type of metadata features can help in this task. Finally, different prediction thresholds and ensembles of the developed models are utilized to investigate the possible improvements, and a newly proposed alternative early detection metric is evaluated.

This work introduces a new multimodal image dataset, with the aim of detecting the interplay between visual elements and semantic relations present in radiology images. The objective is accomplished by retrieving all image-caption pairs from the open-access biomedical literature database PubMedCentral, as these captions describe the visual content in their semantic context. All compound, multi-pane, and non-radiology images were eliminated using an automatic binary classifier fine-tuned with a deep convolutional neural network system. Radiology Objects in COntext (ROCO) dataset contains over 81k radiology images with several medical imaging modalities including Computer Tomography, Ultrasound, X-Ray, Fluoroscopy, Positron Emission Tomography, Mammography, Magnetic Resonance Imaging, Angiography. All images in ROCO have corresponding caption, keywords, Unified Medical Language Systems Concept Unique Identifiers and Semantic Type. An out-of-class set with 6k images ranging from synthetic radiology figures to digital arts is provided, to improve prediction and classification performance. Adopting ROCO, systems for caption and keywords generation can be modeled, which allows multimodal representation for datasets lacking text representation. Systems with the goal of image structuring and semantic information tagging can be created using ROCO, which is beneficial and of assistance for image and information retrieval purposes.

Automatic classification systems are required to support medical literature databases like PubMedCentral, which allow an easy access to millions of articles. FHDO Biomedical Computer Science Group (BCSG) participated at the ImageCLEF 2016 Subfigure Classification Task to improve existing approaches for classifying figures from medical literature. In this work, a data analysis is conducted in order to improve image preprocessing for deep learning approaches. Evaluations on the dataset show better ensemble classification accuracies using only visual information with an optimized training, in comparison to the mixed feature approaches of BCSG at ImageCLEF 2016. Additionally, a self-training approach is investigated to generate more labeled data in the medical domain.