34 results
Now showing1 - 10 of 34
- Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , A review of deep learning-based unsupervised anomaly detection in brain MRI(Elsevier, 2026-07-01); ; ; ; The manual assessment of brain Magnetic Resonance Imaging (MRI) scans can be labor-intensive and time-consuming for radiologists. Deep Learning methods have demonstrated the potential to aid this process. However, their effectiveness relies on the availability of large, annotated data sets. Unsupervised Anomaly Detection (UAD) presents a promising alternative, offering the potential to identify and localize anomalies without per-pixel annotations. Instead, a normative distribution is learned using healthy data, enabling the identification of abnormalities as deviations. This allows UAD methods to detect abnormalities that were unseen during training. This appealing feature has led to numerous studies proposing innovations and novel approaches. In this work, we provide a review of the literature and systematically collect and compare the proposed approaches. We observe that UAD has made significant advancements in brain MRI analysis. However, individual approaches are often evaluated in different contexts, i.e., changes in acquisition parameters, pre- and post-processing, and anomaly scoring. This variability makes it challenging to assess which models perform best, underscoring the need for comprehensive comparative studies concerning the specific context of MRI scans. Our collection, featuring public data sets, research studies, and open implementations, is available at our GitHub repository https://github.com/FinnBehrendt/Unsupervised-Anomaly-Detection-in-Brain-MRI .Publicationtype: Review Article11 25 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Deep generative models for unsupervised anomaly detection in magnetic resonance imaging of the brainThis thesis explores unsupervised anomaly detection in magnetic resonance imaging of the brain, leveraging generative models to model the distribution of healthy anatomy and identify deviations as anomalies. We examine the integration of additional contextual information into generative models and anomaly scoring methods. Our approaches enhance reconstruction quality and anomaly detection accuracy.Publicationtype: Doctoral Thesis113 66 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: without files , Deep denoising of volumetric OCT images for in vivo motion detection(IEEE, 2025-07); ; ; ; Optical Coherence Tomography (OCT) is an imaging modality with high temporal and spatial resolution. There are different applications, e.g. in ophthalmology, dermatology, cardiology and recently OCT image based motion compensation has been considered. However, OCT suffers from speckle noise which affects image quality and subsequent interpretation. We propose denoising based on unsupervised 3D convolutional Autoencoder (AE) and systematically evaluate the model on in vivo and post mortem datasets. Moreover, we study motion detection as a subsequent task and to investigate the effects of AE based denoising. Our results demonstrate that speckle noise in the OCT images can lead to substantial outliers. Denoising based on AEs is effective in reducing the outliers and results in improved motion detection. Furthermore, the proposed AE processes OCT volumes in 0.5 ms, making it suitable for real-time applications. Finally, the results illustrate that the AE can effectively improve motion detection performance on in vivo data, despite being trained on different data. In conclusion, our AE model presents a simple and unsupervised deep learning approach to obtain fast denoising adapted to OCT imaging. For motion detection, denoising can be crucial to avoid artifacts due to outliers.Clinical relevance: This work provides an efficient denoising Autoencoder for fast data processing that can be applied in various clinical scenarios to improve noisy 3D Optical Coherence Tomography images.Publicationtype: Conference Paper7 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: without files , A-scan sequence transformers for palpation with optical coherence elastography(2025-05-01); ; ; ; Recognizing the properties of elastic tissue can facilitate surgical navigation, e.g., when localizing lesions by palpation. However, palpation is very subjective and often unavailable in minimally invasive surgery. High-speed optical coherence elastography (OCE) adapted for intraoperative use could enable elasticity estimation by measuring the propagation of mechanically stimulated waves. However, robust estimation of wave velocity can be challenging, and reconstruction of the elastic modulus is highly dependent on the correct modeling of wave propagation. We therefore consider deep learning for the end-to-end estimation of elasticity from OCE phase data. Since optical coherence tomography inherently produces a temporal sequence of one-dimensional axial scans (A-scans), we consider transformer-based deep learning models to directly process A-scan sequences. For homogeneous tissue phantoms with known elastic properties, we obtain a mean error of 1.64 kPa, which significantly improves elasticity reconstruction compared to conventional processing and the best CNN-based approach with 7.80 kPa and 5.55 kPa, respectively. Furthermore, we demonstrate generalization to heterogeneous phantoms with inclusions and assess the elasticity of soft tissue samples, including heart, kidney, and liver. The results show that transformer architectures are well suited for reconstructing elasticity from A-scan sequences in OCE.Publicationtype: Journal Article26 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Guided reconstruction with conditioned diffusion models for unsupervised anomaly detection in brain MRIs(2025-01-22); ; ; ; The application of supervised models to clinical screening tasks is challenging due to the need for annotated data for each considered pathology. Unsupervised Anomaly Detection (UAD) is an alternative approach that aims to identify any anomaly as an outlier from a healthy training distribution. A prevalent strategy for UAD in brain MRI involves using generative models to learn the reconstruction of healthy brain anatomy for a given input image. As these models should fail to reconstruct unhealthy structures, the reconstruction errors indicate anomalies. However, a significant challenge is to balance the accurate reconstruction of healthy anatomy and the undesired replication of abnormal structures. While diffusion models have shown promising results with detailed and accurate reconstructions, they face challenges in preserving intensity characteristics, resulting in false positives. We propose conditioning the denoising process of diffusion models with additional information derived from a latent representation of the input image. We demonstrate that this conditioning allows for accurate and local adaptation to the general input intensity distribution while avoiding the replication of unhealthy structures. We compare the novel approach to different state-of-the-art methods and for different data sets. Our results show substantial improvements in the segmentation performance, with the Dice score improved by 11.9%, 20.0%, and 44.6%, for the BraTS, ATLAS and MSLUB data sets, respectively, while maintaining competitive performance on the WMH data set. Furthermore, our results indicate effective domain adaptation across different MRI acquisitions and simulated contrasts, an important attribute for general anomaly detection methods. The code for our work is available at https://github.com/FinnBehrendt/Conditioned-Diffusion-Models-UAD.Publicationtype: Journal Article39 40 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Computer-aided diagnosis of maxillary sinus anomalies : validation and clinical correlation(Wiley, 2024-09); ; ; ; Objective: Computer aided diagnostics (CAD) systems can automate the differentiation of maxillary sinus (MS) with and without opacification, simplifying the typically laborious process and aiding in clinical insight discovery within large cohorts. Methods: This study uses Hamburg City Health Study (HCHS) a large, prospective, long-term, population-based cohort study of participants between 45 and 74 years of age. We develop a CAD system using an ensemble of 3D Convolutional Neural Network (CNN) to analyze cranial MRIs, distinguishing MS with opacifications (polyps, cysts, mucosal thickening) from MS without opacifications. The system is used to find correlations of participants with and without MS opacifications with clinical data (smoking, alcohol, BMI, asthma, bronchitis, sex, age, leukocyte count, C-reactive protein, allergies). Results: The evaluation metrics of CAD system (Area Under Receiver Operator Characteristic: 0.95, sensitivity: 0.85, specificity: 0.90) demonstrated the effectiveness of our approach. MS with opacification group exhibited higher alcohol consumption, higher BMI, higher incidence of intrinsic asthma and extrinsic asthma. Male sex had higher prevalence of MS opacifications. Participants with MS opacifications had higher incidence of hay fever and house dust allergy but lower incidence of bee/wasp venom allergy. Conclusion: The study demonstrates a 3D CNN's ability to distinguish MS with and without opacifications, improving automated diagnosis and aiding in correlating clinical data in population studies. Level of Evidence: 3 Laryngoscope, 134:3927–3934, 2024.Publicationtype: Journal Article20 63 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , PolypNextLSTM: a lightweight and fast polyp video segmentation network using ConvNext and ConvLSTM(Springer, 2024-08-08); ; ; ; Purpose: Commonly employed in polyp segmentation, single-image UNet architectures lack the temporal insight clinicians gain from video data in diagnosing polyps. To mirror clinical practices more faithfully, our proposed solution, PolypNextLSTM, leverages video-based deep learning, harnessing temporal information for superior segmentation performance with least parameter overhead, making it possibly suitable for edge devices. Methods: PolypNextLSTM employs a UNet-like structure with ConvNext-Tiny as its backbone, strategically omitting the last two layers to reduce parameter overhead. Our temporal fusion module, a Convolutional Long Short Term Memory (ConvLSTM), effectively exploits temporal features. Our primary novelty lies in PolypNextLSTM, which stands out as the leanest in parameters and the fastest model, surpassing the performance of five state-of-the-art image and video-based deep learning models. The evaluation of the SUN-SEG dataset spans easy-to-detect and hard-to-detect polyp scenarios, along with videos containing challenging artefacts like fast motion and occlusion. Results: Comparison against 5 image-based and 5 video-based models demonstrates PolypNextLSTM’s superiority, achieving a Dice score of 0.7898 on the hard-to-detect polyp test set, surpassing image-based PraNet (0.7519) and video-based PNS+ (0.7486). Notably, our model excels in videos featuring complex artefacts such as ghosting and occlusion. Conclusion: PolypNextLSTM, integrating pruned ConvNext-Tiny with ConvLSTM for temporal fusion, not only exhibits superior segmentation performance but also maintains the highest frames per speed among evaluated models. Code can be found here: https://github.com/mtec-tuhh/PolypNextLSTM.Publicationtype: Journal Article19 21 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Needle tracking in low-resolution ultrasound volumes using deep learning(Springer, 2024-07-13); ; ; ; Purpose: Clinical needle insertion into tissue, commonly assisted by 2D ultrasound imaging for real-time navigation, faces the challenge of precise needle and probe alignment to reduce out-of-plane movement. Recent studies investigate 3D ultrasound imaging together with deep learning to overcome this problem, focusing on acquiring high-resolution images to create optimal conditions for needle tip detection. However, high-resolution also requires a lot of time for image acquisition and processing, which limits the real-time capability. Therefore, we aim to maximize the US volume rate with the trade-off of low image resolution. We propose a deep learning approach to directly extract the 3D needle tip position from sparsely sampled US volumes. Methods: We design an experimental setup with a robot inserting a needle into water and chicken liver tissue. In contrast to manual annotation, we assess the needle tip position from the known robot pose. During insertion, we acquire a large data set of low-resolution volumes using a 16 × 16 element matrix transducer with a volume rate of 4 Hz. We compare the performance of our deep learning approach with conventional needle segmentation. Results: Our experiments in water and liver show that deep learning outperforms the conventional approach while achieving sub-millimeter accuracy. We achieve mean position errors of 0.54 mm in water and 1.54 mm in liver for deep learning. Conclusion: Our study underlines the strengths of deep learning to predict the 3D needle positions from low-resolution ultrasound volumes. This is an important milestone for real-time needle navigation, simplifying the alignment of needle and ultrasound probe and enabling a 3D motion analysis.Publicationtype: Journal Article18 81 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Combining Reconstruction-based Unsupervised Anomaly Detection with Supervised Segmentation for Brain MRIsIn contrast to supervised deep learning approaches, unsupervised anomaly detection (UAD) methods can be trained with healthy data only and do not require pixel-level annotations, enabling the identification of unseen pathologies. While this is promising for clinical screening tasks, reconstruction-based UAD methods fall short in segmentation accuracy compared to supervised models. Therefore, self-supervised UAD approaches have been proposed to improve segmentation accuracy. Typically, synthetic anomalies are used to train a segmentation network in a supervised fashion. However, this approach does not effectively generalize to real pathologies. We propose a framework combining reconstruction-based and self-supervised UAD methods to improve both segmentation performance for known anomalies and generalization to unknown pathologies. The framework includes an unsupervised diffusion model trained on healthy data to produce pseudo-healthy reconstructions and a supervised Unet trained to delineate anomalies from deviations between input- reconstruction pairs. Besides the effective use of synthetic training data, this framework allows for weakly-supervised training with small annotated data sets, generalizing to unseen pathologies. Our results show that with our approach, utilizing annotated data sets during training can substantially improve the segmentation performance for in-domain data while maintaining the generalizability of reconstruction-based approaches to pathologies unseen during trainingPublicationtype: Conference Paper30 26 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: without files , Efficient anatomy segmentation in laparoscopic surgery using multi-teacher knowledge distillation(Microtome Publishing, 2024-07); ; ; ; Automatic segmentation of anatomical structures in laparoscopic images or videos is an important prerequisite for visual assistance tools which are designed to increase efficiency and safety during an intervention. In order to be used in a realistic clinical scenario, both high accuracy and real-time capability are required. Current deep learning networks for anatomy segmentation show high accuracy, but are not suitable for real-time clinical application due to their large size. As smaller, real-time capable deep learning networks show lower segmentation performance, we propose a multi-teacher knowledge distillation approach applicable to partially labeled datasets. We leverage the knowledge of multiple anatomy-specific, high-accuracy teacher networks to improve the segmentation performance of a single and efficient student network capable of segmenting multiple anatomies simultaneously. To do so, we minimize the Kullback-Leibler divergence between the normalized anatomy-specific teacher logits and the respective normalized logits of the student. We conduct experiments on the Dresden Surgical Anatomy Dataset, which provides multiple subsets of binary segmented anatomical structures. Results show that our approach can increase the overall Dice score for different real-time capable network architectures for anatomy segmentation.Publicationtype: Conference Paper18