Refine
Year of publication
Document Type
- conference proceeding (article) (52)
- Article (50)
- conference proceeding (presentation, abstract) (33)
- Preprint (3)
- Book (1)
- Part of a Book (1)
Has Fulltext
- no (140) (remove)
Is part of the Bibliography
- no (140)
Keywords
- Bildgebendes Verfahren (17)
- Deep Learning (14)
- Diagnose (13)
- Gehirn (11)
- Künstliche Intelligenz (11)
- Maschinelles Lernen (11)
- Artificial Intelligence (9)
- Kernspintomografie (9)
- Dreidimensionale Bildverarbeitung (8)
- Registrierung <Bildverarbeitung> (8)
- Speiseröhrenkrankheit (8)
- Lernprogramm (7)
- Schnittdarstellung (7)
- Bildsegmentierung (6)
- Handchirurgie (6)
- Speiseröhrenkrebs (6)
- Adenocarcinoma (5)
- Bildverarbeitung (5)
- Deep learning (5)
- Machine learning (5)
- Osteosynthese (5)
- Polarisiertes Licht (5)
- Barrett's esophagus (4)
- Farbkonstanz (4)
- Gehirnkarte (4)
- Histologie (4)
- ICP-Massenspektrometrie (4)
- Image analysis (4)
- Image processing (4)
- Laryngoskopie (4)
- Medizin (4)
- Metalle (4)
- Positronen-Emissions-Tomografie (4)
- Schwächung (4)
- 3D-Druck (3)
- Automatische Klassifikation (3)
- Barrett-Ösophagus (3)
- Endoscopy (3)
- Endoskopie (3)
- Hirntumor (3)
- Magnetic Resonance Imaging (3)
- Metallproteide (3)
- Neuronales Netz (3)
- Simulation (3)
- Adenocarcinom (2)
- Anomalie (2)
- Barrett’s esophagus (2)
- Bilderkennung (2)
- Bilderzeugung (2)
- Brain (2)
- Computed tomography (2)
- Computertomographie (2)
- Computerunterstützte Medizin (2)
- Computerunterstütztes Verfahren (2)
- Convolutional neural networks (2)
- Dichromatisches Reflexionsmodell (2)
- Dreidimensionale Rekonstruktion (2)
- Dual-material 3D printing (2)
- Elektrophorese (2)
- Generative adversarial networks (2)
- HaptiVisT (2)
- Image classification (2)
- Image fusion (2)
- Image registration (2)
- Image segmentation (2)
- Konturverfolgung (2)
- Laser ablation inductively coupled plasma mass spectrometry (2)
- Literaturbericht (2)
- MRI (2)
- Machine Learning (2)
- Maschinelles Sehen (2)
- Medical Image Computing (2)
- Medical imaging (2)
- Mustererkennung (2)
- Objekterkennung (2)
- Operationstechnik (2)
- PET/CT (2)
- Segmentation (2)
- Snakes (2)
- Third-Space Endoscopy (2)
- Ultrastruktur (2)
- Video (2)
- Visualization (2)
- Zilie (2)
- artificial intelligence (2)
- dichromatisches Reflexionsmodell (2)
- 3D analysis (1)
- 3D breast scan registration (1)
- 3D image processing (1)
- 3D imaging (1)
- 3D reconstruction (1)
- 3D segmentation (1)
- 3D surface imaging (1)
- 4FC (1)
- A1 adenosine receptor (1)
- Adenokarzinom (1)
- Adenosinrezeptor (1)
- Aktives Konturmodell (1)
- Algorithmus (1)
- Alloplastik (1)
- Amino acid transport (1)
- Aminosäuren (1)
- Analytical models (1)
- Anatomy (1)
- Antiangiogenese (1)
- Antiangiogenic treatment (1)
- Arthrodese (1)
- Artificial intelligence (1)
- Attenuation correction (1)
- Aufnahme (1)
- Ausrichtung (1)
- Autogene Transplantation (1)
- Balloon Model (1)
- Barrett (1)
- Barrett's Carcinoma (1)
- Barrett's Esophagus (1)
- Barrett's esphagus (1)
- Barrett’s Esophagus (1)
- Barrett’s cancer (1)
- Barrett’s esophagus detection (1)
- Beryllium (1)
- Bewegungsanalyse (1)
- Bewegungsschatzung (1)
- Bildbasierte Verfahren (1)
- Bioimaging (1)
- Bioimaging of metals (1)
- Biomarker (1)
- Biomaterial (1)
- Biomedical imaging (1)
- Biopsy (1)
- Blood–brain barrier (1)
- Blut-Hirn-Schranke (1)
- Blutdruck (1)
- Blutgefäß (1)
- Boltzmann-Maschine (1)
- Brain Segmentation (1)
- Brain modeling (1)
- Brain tissue (1)
- Brain tumour (1)
- Breast imaging (1)
- Breast reconstruction (1)
- Breast symmetry (1)
- Brustkorb (1)
- Celiac Disease (1)
- Chest X-Ray (1)
- Co-occurrence Matrices (1)
- Co-occurrence matrix (1)
- Cochlea-Implantat (1)
- Color Texture (1)
- Color texture (1)
- Colorconstancy (1)
- Computer aided diagnosis and therapy (1)
- Computer-aided diagnosis (1)
- Computerassistierte Chirurgie (1)
- Computertomografie (1)
- Connectome (1)
- Contour Detection (1)
- Convex optimization (1)
- Cooccurrence Matrix (1)
- Data acquisition (1)
- Data models (1)
- Data privacy (1)
- Datenfusion (1)
- Datenschutz (1)
- Diagnosis (1)
- Diagnostik (1)
- Diffusion-weighted imaging (1)
- Digital anthropometry (1)
- Digital endoscopy (1)
- Dokument (1)
- Electron microscopy (1)
- Encoder-Decoder Network (1)
- Endobrachyösophagus (1)
- Eosinophilic Esophagitis (1)
- Farbanalyse (1)
- Farbbild (1)
- Farbbildverarbeitung (1)
- Farbenraum (1)
- Fehlerbehandlung (1)
- Felsenbein (1)
- Fluorescence imaging (1)
- Force-feedback haptic (1)
- Fräsen (1)
- GBM (1)
- Gabor Filter (1)
- Gamification in der Medizin (1)
- Gastroenterologie (1)
- General Purpose Graphic Processing Unit (1)
- Glanzlichtelimination (1)
- Gliom (1)
- Graphical user interface (1)
- Graphische Benutzeroberfläche (1)
- Graustufe (1)
- Hand surgery training (1)
- Haptische Feedback-Technologie (1)
- Haptisches Feedback (1)
- Hierarchische Wasserscheiden-Transformation (1)
- High-grade glioma (1)
- High-resolution imaging (1)
- Hochschuldidaktik (1)
- Human brain (1)
- Humans (1)
- Hämodynamik (1)
- Image processing (1)
- Image Processing (1)
- Image generation (1)
- Image resolution (1)
- Implantatwerkstoff (1)
- In vitro testing (1)
- In vivo imaging (1)
- Indexierung <Inhaltserschließung> (1)
- Inductively Couple Plasma Mass Spectrometry (1)
- Infinity Restricted Boltzmann Machines (1)
- Information Retrieval (1)
- Integrative features (1)
- Inverse registration consistency (1)
- K-wire drilling (1)
- Klassifikation (1)
- Kmeans algorithm (1)
- KolmogKorov distance (1)
- Komponentenanalyse (1)
- Konvexe Optimierung (1)
- Krankheitsverlauf (1)
- LA-ICP-MS (1)
- Laryngoscopy (1)
- Laser Ablation Inductively Couple Plasma Mass Spectrometry (1)
- Laser microdissection inductively coupled plasma mass spectrometry (1)
- Literaturdatenbank (1)
- MPTP (1)
- MPTP Treatment (1)
- Magnetic resonance imaging (1)
- Mammoplastik (1)
- Massenspektrometrie (1)
- Materialprüfung (1)
- Medical diagnostic imaging (1)
- Medical training system (1)
- Meta-heuristics (1)
- Metaheuristik (1)
- Metallomics (1)
- Metals (1)
- Metamaterial (1)
- Method (1)
- Minimally invasive hand surgery (1)
- Model-based imaging (1)
- Multi-modal imaging (1)
- Multimodal Imaging (1)
- Multimodal response assessment (1)
- Multimodales Verfahren (1)
- Multiskalenbilder (1)
- Multispektralbilder (1)
- Multistep training (1)
- NMDA receptors (1)
- Nano-LA-ICP-MS (1)
- Neoplasms (1)
- Nervenfaser (1)
- Neuroimaging (1)
- Neuronale Netze (1)
- Nichtlineare Optimierung (1)
- Non-rigid surface registration (1)
- Object detector (1)
- Optical Flow (1)
- Optimierung (1)
- PET (1)
- PET/MRI (1)
- Parallel Execution (1)
- Parallelverarbeitung (1)
- Parametric serial MR image registration (1)
- Pathologische Anatomie (1)
- Pattern recognition (1)
- Pflanzen (1)
- Point Distribution Model (1)
- Polarized light imaging (1)
- Positron emission tomography (1)
- Pradikatenlogik (1)
- Quantitative Image analysis (1)
- Receptor autoradiography (1)
- Referenzdaten (1)
- Reflexion (1)
- SLAC wrist (1)
- SNAC wrist (1)
- Schnittpräparat (1)
- Segmentierung (1)
- Segmentierung der Lippen (1)
- Sehrinde (1)
- Semi-Supervised Learning (1)
- Semi-automated segmentation (1)
- Senile Makuladegeneration (1)
- Signaltrennung (1)
- Signalverarbeitung (1)
- Smart ESD (1)
- Statistical shape mode (1)
- Stroboscopic Images (1)
- Substantia Nigra (1)
- Suchmaschine (1)
- Support material (1)
- Surgical instrument segmentation (1)
- Surgical outcome simulation (1)
- Systems biology (1)
- Table lookup (1)
- Texture Analysis (1)
- Texturerkennung (1)
- Tissue-imitating hand phantom (1)
- Tractography (1)
- Tumor disease progression (1)
- Tumour (1)
- Vascular Malformations (1)
- Vascular malformation (1)
- Vektorbilder (1)
- Vesicle membrane analysis (1)
- Vibration Profile (1)
- Virtual fixtures (1)
- Virtual reality (1)
- Virtualisierung (1)
- Virtuelle Realität (1)
- Virtuelles Training (1)
- Visualisierung (1)
- Vizualization (1)
- Voxel Spacing (1)
- Zielverfolgung (1)
- attenuation (1)
- camera calibration (1)
- celiac disease (1)
- combined imaging (1)
- correction (1)
- d/l-serine (1)
- deep learning (1)
- diagnostic laryngoscopy (1)
- digital anthropometry (1)
- electronic imaging (1)
- endoscopy (1)
- endoscopy detection (1)
- hand surgery training (1)
- healthcare (1)
- herbarium specimens (1)
- image co-registration (1)
- image distortion (1)
- magnetic resonance imaging (1)
- medizinische Bildverarbeitung (1)
- metal distribution (1)
- metallomics (1)
- metamaterial (1)
- multiple regression analysis (1)
- neoplastic larynx disease (1)
- neurodegenerative diseases (1)
- object detection (1)
- quantitative Farbmessung (1)
- real-time (1)
- reconstructive surgery (1)
- robot-assisted surgery (1)
- smoke simulation (1)
- submucosal invasion (1)
- tissue imitating phantom hand (1)
- unpaired image-to-image translation (1)
- villous atrophy (1)
- visual recognition (1)
- Ähnlichkeitssuche (1)
Institute
- Fakultät Informatik und Mathematik (140) (remove)
Begutachtungsstatus
- peer-reviewed (69)
- begutachtet (1)
Bestimmung der Lichtquellenfarbe bei der Endoskopie mikrotexturierter Oberflächen des Kehlkopfes
(1999)
Zur Unterstützung der Diagnose von Stimmlippenerkrankungen werden innerhalb des Forschungsprojektes Quantitative Digitale Laryngoskopie objektive Parameter zur Beschreibung der Bewegung, der Farbe sowie der Form der Stimmlippen entwickelt und klinisch evaluiert. Während die Bewegungsanalyse Aufschluß über funktionelle Stimmstörungen gibt, beschreiben Parameter der Farb- und Formanalyse morphologische Veränderungen des Stimmlippengewebes. In diesem Beitrag werden die Methoden und bisherigen Ergebnisse zur Bewegungs- und Farbanalyse vorgestellt.
Die Bewegungsanalyse wurde mit einem erweiterten Konturmodell (Snakes) durchgeführt. Aufgrund des modifizierten Konturmodells konnten die Konturen der Stimmlippen automatisch über die gesmate Bildsequenz zuverlässig detektiert werden. Die Vermssung der Konturen liefert neue quantitative Parameter zur Befundung von laryngoskopischen Stimmlippenaufnahmen.
Um die Farbeigenschaften der Stimmlippen zu bestimmen, wurde ausgehend vom RGB-Bild die Objektfarbe unabhängig von der Farbe der Lichtquelle durch Verwendung von Clusterverfahren und der Viertelkreisanalyse berechnet. Mit dieser Farbanalyse konnte die Farbe der Lichtquelle ermittelt und das beleuchtungsunabhängige Farbbild berechnet werden. Die Quanitifizierung der Rötung der Stimmlippen ist z.B. ein entscheidendes Kriterium zur Diagnostik der akuten Laryngitis.
GraphMIC is a cross-platform image processing application utilizing the libraries ITK and OpenCV. The abstract structure of image processing pipelines is visually represented by user interface components based on modern QtQuick technology and allows users to focus on arrangement and parameterization of operations rather than implementing the equivalent functionality natively in C++. The application's central goal is to improve and simplify the typical workflow by providing various high level features and functions like multi threading, image sequence processing and advanced error handling. A built-in python interpreter allows the creation of custom nodes, where user defined algorithms can be integrated to extend basic functionality. An embedded 2d/3d visual-izer gives feedback of the resulting image of an operation or the whole pipeline. User inputs like seed points, contours or regions are forwarded to the processing pipeline as parameters to offer semi-automatic image computing. We report the main concept of the application and introduce several features and their implementation. Finally, the current state of development as well as future perspectives of GraphMIC are discussed
Zur Lösung komplexer Segmentierungsprobleme wird eine hierarchische und farbbasierte Wasserscheidentransformation vorgestellt. Geringe Modifikationen bezüglich Startpunktwahl und Flutungsprozess resultieren in signifikanten Verbesserungen der Segmentierung. Das Verfahren wurde zur Lippendetektion in Farbbildsequenzen eingesetzt, die zur quantitativen Beschreibung von Sprechbewegungsabläufen automatisch ausgewertet werden. Die Experimente mit 245 Bildern aus 6 Sequenzen zeigten eine Fehlerrate von 13%.
Color Texture Analysis of Moving Vocal Cords Using Approaches from Statistics and Signal Theory
(2000)
Textural features are applied for detection of morphological pathologies of vocal cords. Cooccurrence matrices as statistical features are presented as well as filter bank analysis by Gabor filters. Both methods are extended to handle color images. Their robustness against camera movement and vibration of vocal cords is evaluated. Classification results due to three in vivo sequences are in between 94.4 % and 98.9%. The classification errors decrease if color features are used instead of grayscale features for both statistical and Fourier features
Purpose
Age-related macular degeneration (AMD) is a common threat to vision. While classification of disease stages is critical to understanding disease risk and progression, several systems based on color fundus photographs are known. Most of these require in-depth and time-consuming analysis of fundus images. Herein, we present an automated computer-based classification algorithm.
Design Algorithm development for AMD classification based on a large collection of color fundus images. Validation is performed on a cross-sectional, population-based study.
Participants.
We included 120 656 manually graded color fundus images from 3654 Age-Related Eye Disease Study (AREDS) participants. AREDS participants were >55 years of age, and non-AMD sight-threatening diseases were excluded at recruitment. In addition, performance of our algorithm was evaluated in 5555 fundus images from the population-based Kooperative Gesundheitsforschung in der Region Augsburg (KORA; Cooperative Health Research in the Region of Augsburg) study.
Methods.
We defined 13 classes (9 AREDS steps, 3 late AMD stages, and 1 for ungradable images) and trained several convolution deep learning architectures. An ensemble of network architectures improved prediction accuracy. An independent dataset was used to evaluate the performance of our algorithm in a population-based study.
Main Outcome Measures.
κ Statistics and accuracy to evaluate the concordance between predicted and expert human grader classification.
Results.
A network ensemble of 6 different neural net architectures predicted the 13 classes in the AREDS test set with a quadratic weighted κ of 92% (95% confidence interval, 89%–92%) and an overall accuracy of 63.3%. In the independent KORA dataset, images wrongly classified as AMD were mainly the result of a macular reflex observed in young individuals. By restricting the KORA analysis to individuals >55 years of age and prior exclusion of other retinopathies, the weighted and unweighted κ increased to 50% and 63%, respectively. Importantly, the algorithm detected 84.2% of all fundus images with definite signs of early or late AMD. Overall, 94.3% of healthy fundus images were classified correctly.
Conclusions
Our deep learning algoritm revealed a weighted κ outperforming human graders in the AREDS study and is suitable to classify AMD fundus images in other datasets using individuals >55 years of age.
Integrative Co-occurrence matrices are introduced as novel features for color texture classification. The extended Co-occurrence notation allows the comparison between integrative and parallel color texture concepts. The information profit of the new matrices is shown quantitatively using the Kolmogorov distance and by extensive classification experiments on two datasets. Applying them to the RGB and the LUV color space the combined color and intensity textures are studied and the existence of intensity independent pure color patterns is demonstrated. The results are compared with two baselines: gray-scale texture analysis and color histogram analysis. The novel features improve the classification results up to 20% and 32% for the first and second baseline, respectively.
Purpose
MR-based attenuation correction (AC) will become an integral part of combined PET/MR systems. Here, we propose a toolbox to validate MR-AC of clinical PET/MRI data sets.
Methods
Torso scans of ten patients were acquired on a combined PET/CT and on a 1.5-T MRI system. MR-based attenuation data were derived from the CT following MR–CT image co-registration and subsequent histogram matching. PET images were reconstructed after CT- (PET/CT) and MR-based AC (PET/MRI). Lesion-to-background (L/B) ratios were estimated on PET/CT and PET/MRI.
Results
MR–CT histogram matching leads to a mean voxel intensity difference in the CT- and MR-based attenuation images of 12% (max). Mean differences between PET/MRI and PET/CT were 19% (max). L/B ratios were similar except for the lung where local misregistration and intensity transformation leads to a biased PET/MRI.
Conclusion
Our toolbox can be used to study pitfalls in MR-AC. We found that co-registration accuracy and pixel value transformation determine the accuracy of PET/MRI.
Mass spectrometric imaging (MSI) is a young innovative analytical technique and combines different fields of advanced mass spectrometry and biomedical research with the aim to provide maps of elements and molecules, complexes or fragments. Especially essential metals such as zinc, copper, iron and manganese play a functional role in signaling, metabolism and homeostasis of the cell. Due to the high degree of spatial organization of metals in biological systems their distribution analysis is of key interest in life sciences. We have developed analytical techniques termed BrainMet using laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) imaging to measure the distribution of trace metals in biological tissues for biomedical research and feasibility studies—including bioaccumulation and bioavailability studies, ecological risk assessment and toxicity studies in humans and other organisms. The analytical BrainMet techniques provide quantitative images of metal distributions in brain tissue slices which can be combined with other imaging modalities such as photomicrography of native or processed tissue (histochemistry, immunostaining) and autoradiography or with in vivo techniques such as positron emission tomography or magnetic resonance tomography.
Prospective and instrumental developments will be discussed concerning the development of the metalloprotein microscopy using a laser microdissection (LMD) apparatus for specific sample introduction into an inductively coupled plasma mass spectrometer (LMD-ICP-MS) or an application of the near field effect in LA-ICP-MS (NF-LA-ICP-MS). These nano-scale mass spectrometric techniques provide improved spatial resolution down to the single cell level.
Local gray level dependencies of natural images can be modelled by means of co-occurrence matrices containing joint probabilities of gray-level pairs. Texture, however, is a resolution-dependent phenomenon and hence, classification depends on the chosen scale. Since there is no optimal scale for all textures we employ a multiscale approach that acquires textural features at several scales. Thus linear and nonlinear scale-spaces are analyzed by multiscale co-occurrence matrices that describe the statistical behavior of a texture in scale-space. Classification is then performed on the basis of texture features taken from the individual scale with the highest discriminatory power. By considering cross-scale occurrences of gray level pairs, the impact of filters on the feature is described and used for classification of natural textures. This novel method was found to improve classification rates of the common co-occurrence matrix approach on standard textures significantly.
Background and aims
Celiac disease with its endoscopic manifestation of villous atrophy is underdiagnosed worldwide. The application of artificial intelligence (AI) for the macroscopic detection of villous atrophy at routine esophagogastroduodenoscopy may improve diagnostic performance.
Methods
A dataset of 858 endoscopic images of 182 patients with villous atrophy and 846 images from 323 patients with normal duodenal mucosa was collected and used to train a ResNet 18 deep learning model to detect villous atrophy. An external data set was used to test the algorithm, in addition to six fellows and four board certified gastroenterologists. Fellows could consult the AI algorithm’s result during the test. From their consultation distribution, a stratification of test images into “easy” and “difficult” was performed and used for classified performance measurement.
Results
External validation of the AI algorithm yielded values of 90 %, 76 %, and 84 % for sensitivity, specificity, and accuracy, respectively. Fellows scored values of 63 %, 72 % and 67 %, while the corresponding values in experts were 72 %, 69 % and 71 %, respectively. AI consultation significantly improved all trainee performance statistics. While fellows and experts showed significantly lower performance for “difficult” images, the performance of the AI algorithm was stable.
Conclusion
In this study, an AI algorithm outperformed endoscopy fellows and experts in the detection of villous atrophy on endoscopic still images. AI decision support significantly improved the performance of non-expert endoscopists. The stable performance on “difficult” images suggests a further positive add-on effect in challenging cases.
Background:
Reliable, time- and cost-effective, and clinician-friendly diagnostic tools are cornerstones in facial palsy (FP) patient management. Different automated FP grading systems have been developed but revealed persisting downsides such as insufficient accuracy and cost-intensive hardware. We aimed to overcome these barriers and programmed an automated grading system for FP patients utilizing the House and Brackmann scale (HBS).
Methods:
Image datasets of 86 patients seen at the Department of Plastic, Hand, and Reconstructive Surgery at the University Hospital Regensburg, Germany, between June 2017 and May 2021, were used to train the neural network and evaluate its accuracy. Nine facial poses per patient were analyzed by the algorithm.
Results:
The algorithm showed an accuracy of 100%. Oversampling did not result in altered outcomes, while the direct form displayed superior accuracy levels when compared to the modular classification form (n = 86; 100% vs. 99%). The Early Fusion technique was linked to improved accuracy outcomes in comparison to the Late Fusion and sequential method (n = 86; 100% vs. 96% vs. 97%).
Conclusions:
Our automated FP grading system combines high-level accuracy with cost- and time-effectiveness. Our algorithm may accelerate the grading process in FP patients and facilitate the FP surgeon’s workflow.
In the field of computer- and robot-assisted minimally invasive surgery, enormous progress has been made in recent years based on the recognition of surgical instruments in endoscopic images. Especially the determination of the position and type of the instruments is of great interest here. Current work involves both spatial and temporal information with the idea, that the prediction of movement of surgical tools over time may improve the quality of final segmentations. The provision of publicly available datasets has recently encouraged the development of new methods, mainly based on deep learning. In this review, we identify datasets used for method development and evaluation, as well as quantify their frequency of use in the literature. We further present an overview of the current state of research regarding the segmentation and tracking of minimally invasive surgical instruments in endoscopic images. The paper focuses on methods that work purely visually without attached markers of any kind on the instruments, taking into account both single-frame segmentation approaches as well as those involving temporal information. A discussion of the reviewed literature is provided, highlighting existing shortcomings and emphasizing available potential for future developments. The publications considered were identified through the platforms Google Scholar, Web of Science, and PubMed. The search terms used were "instrument segmentation", "instrument tracking", "surgical tool segmentation", and "surgical tool tracking" and result in 408 articles published between 2015 and 2022 from which 109 were included using systematic selection criteria.
We investigate contrastive learning in a multi-task learning setting classifying and segmenting early Barrett’s cancer. How can contrastive learning be applied in a domain with few classes and low inter-class and inter-sample variance, potentially enabling image retrieval or image attribution? We introduce a data sampling strategy that mines per-lesion data for positive samples and keeps a queue of the recent projections as negative samples. We propose a masking strategy for the NT-Xent loss that keeps the negative set pure and removes samples from the same lesion. We show cohesion and uniqueness improvements of the proposed method in feature space. The introduction of the auxiliary objective does not affect the performance but adds the ability to indicate similarity between lesions. Therefore, the approach could enable downstream auto-documentation tasks on homogeneous medical image data.
Einleitung Die Endoskopische Retrograde Cholangiopankreatikographie (ERCP) ist der Goldstandard in der Diagnostik und Therapie von Erkrankungen des pankreatobiliären Trakts. Jedoch ist sie technisch sehr anspruchsvoll und weist eine vergleichsweise hohe Komplikationsrate auf.
Ziele
In der vorliegenden Machbarkeitsstudie soll geprüft werden, ob mithilfe eines Deep-learning-Algorithmus die Papille und das Ostium zuverlässig detektiert werden können und somit für Endoskopiker mit geringer Erfahrung ein geeignetes Hilfsmittel, insbesondere für die Ausbildungssituation, darstellen könnten.
Methodik
Wir betrachteten insgesamt 606 Bilddatensätze von 65 Patienten. In diesen wurde sowohl die Papilla duodeni major als auch das Ostium segmentiert. Anschließend wurde eine neuronales Netz mittels eines Deep-learning-Algorithmus trainiert. Außerdem erfolgte eine 5-fache Kreuzvaldierung.
Ergebnisse
Bei einer 5-fachen Kreuzvaldierung auf den 606 gelabelten Daten konnte für die Klasse Papille eine F1-Wert von 0,7908, eine Sensitivität von 0,7943 und eine Spezifität von 0,9785 erreicht werden, für die Klasse Ostium eine F1-Wert von 0,5538, eine Sensitivität von 0,5094 und eine Spezifität von 0,9970 (vgl. [Tab. 1]). Unabhängig von der Klasse zeigte sich gemittelt (Klasse Papille und Klasse Ostium) ein F1-Wert von 0,6673, eine Sensitivität von 0,6519 und eine Spezifität von 0,9877 (vgl. [Tab. 2]).
Schlussfolgerung
In vorliegende Machbarkeitsstudie konnte das neuronale Netz die Papilla duodeni major mit einer hohen Sensitivität und sehr hohen Spezifität identifizieren. Bei der Detektion des Ostiums war die Sensitivität deutlich geringer. Zukünftig soll das das neuronale Netz mit mehr Daten trainiert werden. Außerdem ist geplant, den Algorithmus auch auf Videos anzuwenden. Somit könnte langfristig ein geeignetes Hilfsmittel für die ERCP etabliert werden.
BACKGROUND
Differentiation of high-flow from low-flow vascular malformations (VMs) is crucial for therapeutic management of this orphan disease.
OBJECTIVE
A convolutional neural network (CNN) was evaluated for differentiation of peripheral vascular malformations (VMs) on T2-weighted short tau inversion recovery (STIR) MRI.
METHODS
527 MRIs (386 low-flow and 141 high-flow VMs) were randomly divided into training, validation and test set for this single-center study. 1) Results of the CNN's diagnostic performance were compared with that of two expert and four junior radiologists. 2) The influence of CNN's prediction on the radiologists' performance and diagnostic certainty was evaluated. 3) Junior radiologists' performance after self-training was compared with that of the CNN.
RESULTS
Compared with the expert radiologists the CNN achieved similar accuracy (92% vs. 97%, p = 0.11), sensitivity (80% vs. 93%, p = 0.16) and specificity (97% vs. 100%, p = 0.50). In comparison to the junior radiologists, the CNN had a higher specificity and accuracy (97% vs. 80%, p < 0.001; 92% vs. 77%, p < 0.001). CNN assistance had no significant influence on their diagnostic performance and certainty. After self-training, the junior radiologists' specificity and accuracy improved and were comparable to that of the CNN.
CONCLUSIONS
Diagnostic performance of the CNN for differentiating high-flow from low-flow VM was comparable to that of expert radiologists. CNN did not significantly improve the simulated daily practice of junior radiologists, self-training was more effective.