Tackling the class imbalance problem of deep learning-based head and neck organ segmentation.

PURPOSE: The segmentation of organs at risk (OAR) is a required precondition for the cancer treatment with image- guided radiation therapy. The automation of the segmentation task is therefore of high clinical relevance. Deep learning (DL)-based medical image segmentation is currently the most successful approach, but suffers from the over-presence of the background class and the anatomically given organ size difference, which is most severe in the head and neck (HAN) area. METHODS: To tackle the HAN area-specific class imbalance problem, we first optimize the patch size of the currently best performing general-purpose segmentation framework, the nnU-Net, based on the introduced class imbalance measurement, and second introduce the class adaptive Dice loss to further compensate for the highly imbalanced setting. RESULTS: Both the patch size and the loss function are parameters with direct influence on the class imbalance, and their optimization leads to a 3% increase in the Dice score and 22% reduction in the 95% Hausdorff distance compared to the baseline, finally reaching [Formula: see text] and [Formula: see text] mm for the segmentation of seven HAN organs using a single and simple neural network. CONCLUSION: The patch size optimization and the class adaptive Dice loss are both simply integrable in current DL-based segmentation approaches and allow to increase the performance for class imbalance segmentation tasks.

Heart rate estimation from ballistocardiographic signals using deep learning.

Objective.Ballistocardiography (BCG) is an unobtrusive approach for cost-effective and patient-friendly health monitoring. In this work, deep learning methods are used for heart rate estimation from BCG signals and are compared against five digital signal processing methods found in literature.Approach.The models are evaluated on a dataset featuring BCG recordings from 42 patients, acquired with a pneumatic system. Several different deep learning architectures, including convolutional, recurrent and a combination of both are investigated. Besides model performance, we are also concerned about model size and specifically investigate less complex and smaller networks.Main results.Deep learning models outperform traditional methods by a large margin. Across 14 patients in a held-out testing set, an architecture with stacked convolutional and recurrent layers achieves an average mean absolute error (MAE) of 2.07 beat min-1, whereas the best-performing traditional method reaches 4.24 beat min-1. Besides smaller errors, deep learning models show more consistent performance across different patients, indicating the ability to better deal with inter-patient variability, a prevalent issue in BCG analysis. In addition, we develop a smaller version of the best-performing architecture, that only features 8283 parameters, yet still achieves an average MAE of 2.32 beat min-1on the testing set.Significance.This is the first study that applies and compares different deep learning architectures to heart rate estimation from bed-based BCG signals. Compared to signal processing algorithms, deep learning models show dramatically smaller errors and more consistent results across different individuals. The results show that using smaller models instead of excessively large ones can lead to sufficient performance for specific biosignal processing applications. Additionally, we investigate the use of fully convolutional networks for 1D signal processing, which is rarely applied in literature.

Training of head and neck segmentation networks with shape prior on small datasets.

PURPOSE: Cancer in the head and neck area is commonly treated with radiotherapy. A key step for low-risk treatment is the accurate delineation of organs at risk in the planning imagery. The success of deep learning in image segmentation led to automated algorithms achieving human expert performance on certain datasets. However, such algorithms require large datasets for training and fail to segment previously unseen pathologies, where human experts still succeed. As pathologies are rare and large datasets costly to generate, we investigate the effect of: reduced training data, batch sizes and incorporation of prior knowledge. METHODS: The small data problem is studied by training a full-volume segmentation network with the reduced amount of data from the MICCAI 2015 head and neck segmentation challenge. To improve the segmentation, we evaluate the batch size as a hyper-parameter and first study and then incorporate a stacked autoencoder as shape prior into the training process. RESULTS: We found that using half of the training data (12 images of 25) results in an accuracy drop of only 3% for the segmentation of organs at risk. Also, the batch size turns out to be relevant for the quality of the segmentation when trained with less than half of the data. By applying PCA on the autoencoder's latent space we achieve a compact and accurate shape model, which is used as a regularizer and significantly improves the segmentation results. CONCLUSION: Small training data of up to 12 training images is enough to train accurate head and neck segmentation models. By using a shape prior for regularization, the performance of the segmentation can be improved significantly on the full dataset. When training on fewer than 12 images, the batch size is relevant and models have to be trained much longer until convergence.

[Emergency care in nursing homes in the City of Braunschweig: A cross-sectoral view with indicators from deaths, resuscitation, emergency department visits and EMS deployment]. / Notfallversorgung in Pflegeeinrichtungen in der Stadt Braunschweig: Ein sektorenübergreifender Blick anhand von Kennzahlen aus Sterbefällen, Reanimationen, Krankenhauszuweisungen und Rettungsdiensteinsätzen.

INTRODUCTION: General survey of emergency care in nursing homes in the City of Braunschweig. METHODS: Retrospective analysis of data from death registry, resuscitation registry and further routine data from the local health authorities and the emergency medical services (EMS). RESULTS: 30 nursing homes with 3,100 beds (mean: 103; range: 35-250) were included; operators of nursing homes were 18 non-profit organizations; 7 private (local); 5 private (nationwide). Among the inhabitants of these 30 nursing homes 880 deaths occurred, 406 (46 %) in hospital; 4,895 EMS deployments for emergency care; 4,493 (92 %) resulting in emergency department visits; 19 CPRs. EMS deployments without a physician order per bed 1.0 (0.4-1.6); emergency department visits per bed 1.4 (0.7-3.1); rate of EMS deployments without physicians order / emergency department visits 70 % (41-96 %); deaths per bed 0,29 (0.12-0.48); rate of deaths in hospital 46 % (0-62 %); CPRs per 1,000 beds 6.1 (0-28); CPR failure rate 22 (0-83) per 1,000 deaths per year. EMS deployment without physician order was significantly more frequent in privately (nationwide) operated nursing homes 1.2 (1.0-1.4). In the entire urban region, the incidence of EMS deployment without a physician order was 0.2 per inhabitant per year and the rate of hospital deaths was 64 %. CONCLUSION: Compared to the entire population of the City of Braunschweig, EMS deployment was more frequent in nursing homes. The chance of hospital death or failed CPR was slightly lower. There are large variations between the different nursing homes. Indicators from routine data can provide guidance for more specific surveys but do not allow benchmarking.

Multi-organ segmentation of the head and neck area: an efficient hierarchical neural networks approach.

PURPOSE: In radiation therapy, a key step for a successful cancer treatment is image-based treatment planning. One objective of the planning phase is the fast and accurate segmentation of organs at risk and target structures from medical images. However, manual delineation of organs, which is still the gold standard in many clinical environments, is time-consuming and prone to inter-observer variations. Consequently, many automated segmentation methods have been developed. METHODS: In this work, we train two hierarchical 3D neural networks to segment multiple organs at risk in the head and neck area. First, we train a coarse network on size-reduced medical images to locate the organs of interest. Second, a subsequent fine network on full-resolution images is trained for a final accurate segmentation. The proposed method is purely deep learning based; accordingly, no pre-registration or post-processing is required. RESULTS: The approach has been applied on a publicly available computed tomography dataset, created for the MICCAI 2015 Auto-Segmentation challenge. In an extensive evaluation process, the best configurations for the trained networks have been determined. Compared to the existing methods, the presented approach shows state-of-the-art performance for the segmentation of seven different structures in the head and neck area. CONCLUSION: We conclude that 3D neural networks outperform the most existing model- and atlas-based methods for the segmentation of organs at risk in the head and neck area. The ease of use, high accuracy and the test time efficiency of the method make it promising for image-based treatment planning in clinical practice.

Ejection Wave Segmentation for Contact-Free Heart Rate Estimation from Ballistocardiographic Signals.

We present a new algorithm for peak detection in ballistocardiographic (BCG) signals and use it for heart rate estimation. Systolic complexes of the BCG signal are enhanced and coarse heart beat locations estimated. Ejection waves I, J and K are detected simultaneously around coarse locations, only using detection of local maxima and weighted summation of peak heights. Due to a lack of reference BCG annotations, the algorithm's performance is assessed by using the detected peaks for heart rate estimation. On a dataset acquired with a pneumatic BCG system, we evaluate the heart rate estimation performance and compare the introduced algorithm against other methods found in literature. The dataset is gathered from 42 patients in a clinical environment and provides low-quality signals taken from a realistic scenario. With a mean absolute percentage error of 2.58 % at 65 % coverage, the presented method is on par with the best-performing state-of-the-art algorithms investigated. Limits of agreement (5th/95th percentiles) in a comparison with ECG-based heart rate measurements lie within P5 = -3.63 and P95 = 5.78 beat/min.

Sociodemographic determinants of spatial disparities in early childhood caries: An ecological analysis in Braunschweig, Germany.

OBJECTIVES: To identify spatial disparities in dental caries experience (measured by dmft (decayed missing filled teeth) index) of children in the city of Braunschweig and to evaluate whether these disparities can be explained by sociodemographic characteristics. METHODS: We examined the dental health of children aged 3-6 years visiting a daycare centre (DCC) in the metropolitan area of Braunschweig between 2009 and 2014 by combining data on dental health from the annual visits of the local health service with aggregated data on sociodemographic factors for Braunschweig's city districts. We assessed longitudinal patterns of change in average dmft index at district level from 2009 to 2014 using a finite mixture model. We analysed spatial autocorrelation of the district's average dmft indices by Moran's I to identify spatial clusters. With a spatial lag model, we evaluated whether sociodemographic risk factors (data from 2012) were associated with high dmft scores (data from 2014) and whether spatial disparities remained after adjusting for these sociodemographic characteristics. RESULTS: The average dmft index decreased slightly (ß=-0.048; P<.03; CI 95% [-0.079; -0.017]) from 2009 to 2014. The finite mixture model resulted in four different groups of trajectories over time. While three groups showed a decrease in dmft score, one group showed an increase from 2009 to 2014. Moran's I test statistic showed strong evidence for spatial clustering (Moran's I 0.30, P=.002). A cluster of districts with high dmft values was identified in the centre of the city. The spatial lag model showed that both the proportion of unemployed persons (aged 16-65) and the proportion of persons with migration background were associated with the dmft values at district level. After adjusting for these, no further spatial heterogeneity was observed. CONCLUSION: We identified regional clusters for poor dental health in a German city and showed that these clusters can be explained by sociodemographic characteristics. The findings support the need of targeted interventions and prevention measures in regions with less favourable sociodemographic characteristics.

Evaluation of segmentation methods on head and neck CT: Auto-segmentation challenge 2015.

PURPOSE: Automated delineation of structures and organs is a key step in medical imaging. However, due to the large number and diversity of structures and the large variety of segmentation algorithms, a consensus is lacking as to which automated segmentation method works best for certain applications. Segmentation challenges are a good approach for unbiased evaluation and comparison of segmentation algorithms. METHODS: In this work, we describe and present the results of the Head and Neck Auto-Segmentation Challenge 2015, a satellite event at the Medical Image Computing and Computer Assisted Interventions (MICCAI) 2015 conference. Six teams participated in a challenge to segment nine structures in the head and neck region of CT images: brainstem, mandible, chiasm, bilateral optic nerves, bilateral parotid glands, and bilateral submandibular glands. RESULTS: This paper presents the quantitative results of this challenge using multiple established error metrics and a well-defined ranking system. The strengths and weaknesses of the different auto-segmentation approaches are analyzed and discussed. CONCLUSIONS: The Head and Neck Auto-Segmentation Challenge 2015 was a good opportunity to assess the current state-of-the-art in segmentation of organs at risk for radiotherapy treatment. Participating teams had the possibility to compare their approaches to other methods under unbiased and standardized circumstances. The results demonstrate a clear tendency toward more general purpose and fewer structure-specific segmentation algorithms.

Model-based Vestibular Afferent Stimulation: Modular Workflow for Analyzing Stimulation Scenarios in Patient Specific and Statistical Vestibular Anatomy.

Our sense of balance and spatial orientation strongly depends on the correct functionality of our vestibular system. Vestibular dysfunction can lead to blurred vision and impaired balance and spatial orientation, causing a significant decrease in quality of life. Recent studies have shown that vestibular implants offer a possible treatment for patients with vestibular dysfunction. The close proximity of the vestibular nerve bundles, the facial nerve and the cochlear nerve poses a major challenge to targeted stimulation of the vestibular system. Modeling the electrical stimulation of the vestibular system allows for an efficient analysis of stimulation scenarios previous to time and cost intensive in vivo experiments. Current models are based on animal data or CAD models of human anatomy. In this work, a (semi-)automatic modular workflow is presented for the stepwise transformation of segmented vestibular anatomy data of human vestibular specimens to an electrical model and subsequently analyzed. The steps of this workflow include (i) the transformation of labeled datasets to a tetrahedra mesh, (ii) nerve fiber anisotropy and fiber computation as a basis for neuron models, (iii) inclusion of arbitrary electrode designs, (iv) simulation of quasistationary potential distributions, and (v) analysis of stimulus waveforms on the stimulation outcome. Results obtained by the workflow based on human datasets and the average shape of a statistical model revealed a high qualitative agreement and a quantitatively comparable range compared to data from literature, respectively. Based on our workflow, a detailed analysis of intra- and extra-labyrinthine electrode configurations with various stimulation waveforms and electrode designs can be performed on patient specific anatomy, making this framework a valuable tool for current optimization questions concerning vestibular implants in humans.

Automatic segmentation of head and neck CT images for radiotherapy treatment planning using multiple atlases, statistical appearance models, and geodesic active contours.

PURPOSE: Accurate delineation of organs at risk (OARs) is a precondition for intensity modulated radiation therapy. However, manual delineation of OARs is time consuming and prone to high interobserver variability. Because of image artifacts and low image contrast between different structures, however, the number of available approaches for autosegmentation of structures in the head-neck area is still rather low. In this project, a new approach for automated segmentation of head-neck CT images that combine the robustness of multiatlas-based segmentation with the flexibility of geodesic active contours and the prior knowledge provided by statistical appearance models is presented. METHODS: The presented approach is using an atlas-based segmentation approach in combination with label fusion in order to initialize a segmentation pipeline that is based on using statistical appearance models and geodesic active contours. An anatomically correct approximation of the segmentation result provided by atlas-based segmentation acts as a starting point for an iterative refinement of this approximation. The final segmentation result is based on using model to image registration and geodesic active contours, which are mutually influencing each other. RESULTS: 18 CT images in combination with manually segmented labels of parotid glands and brainstem were used in a leave-one-out cross validation scheme in order to evaluate the presented approach. For this purpose, 50 different statistical appearance models have been created and used for segmentation. Dice coefficient (DC), mean absolute distance and max. Hausdorff distance between the autosegmentation results and expert segmentations were calculated. An average Dice coefficient of DC = 0.81 (right parotid gland), DC = 0.84 (left parotid gland), and DC = 0.86 (brainstem) could be achieved. CONCLUSIONS: The presented framework provides accurate segmentation results for three important structures in the head neck area. Compared to a segmentation approach based on using multiple atlases in combination with label fusion, the proposed hybrid approach provided more accurate results within a clinically acceptable amount of time.

Hip fracture discrimination from dual-energy X-ray absorptiometry by statistical model registration.

Although the areal Bone Mineral Density (BMD) measurements from dual-energy X-ray absorptiometry (DXA) are able to discriminate between hip fracture cases and controls, the femoral strength is largely determined by the 3D bone structure. In a previous work a statistical model was presented which parameterizes the 3D shape and BMD distribution of the proximal femur. In this study the parameter values resulting from the registration of the model onto DXA images are evaluated for their hip fracture discrimination ability with respect to regular DXA derived areal BMD measurements. The statistical model was constructed from a large database of QCT scans of females with an average age of 67.8 ± 17.0 years. This model was subsequently registered onto the DXA images of a fracture and control group. The fracture group consisted of 175 female patients with an average age of 66.4 ± 9.9 years who suffered a fracture on the contra lateral femur. The control group consisted of 175 female subjects with an average age of 65.3 ± 10.0 years and no fracture history. The discrimination ability of the resulting model parameter values, as well as the areal BMD measurements extracted from the DXA images were evaluated using a logistic regression analysis. The area under the receiver operating curve (AUC) of the combined model parameters and areal BMD values was 0.840 (95% CI 0.799-0.881), whilst using only the areal BMD values resulted in an AUC of 0.802 (95% CI 0.757-0.848). These results indicate that the discrimination ability of the areal BMD values is improved by supplementing them with the model parameter values, which give a more complete representation of the subject specific shape and internal bone distribution. Thus, the presented method potentially allows for an improved hip fracture risk estimation whilst maintaining DXA as the current standard modality.

A statistical model of shape and bone mineral density distribution of the proximal femur for fracture risk assessment.

This work presents a statistical model of both the shape and Bone Mineral Density (BMD) distribution of the proximal femur for fracture risk assessment. The shape and density model was built from a dataset of Quantitative Computed Tomography scans of fracture patients and a control group. Principal Component Analysis and Horn's parallel analysis were used to reduce the dimensionality of the shape and density model to the main modes of variation. The input data was then used to analyze the model parameters for the optimal separation between the fracture and control group. Feature selection using the Fisher criterion determined the parameters with the best class separation, which were used in Fisher Linear Discriminant Analysis to find the direction in the parameter space that best separates the fracture and control group. This resulted in a Fisher criterion value of 6.70, while analyzing the Dual-energy X-ray Absorptiometry derived femur neck areal BMD of the same subjects resulted in a Fisher criterion value of 0.98. This indicates that a fracture risk estimation approach based on the presented model might improve upon the current standard clinical practice.

3-D graph cut segmentation with Riemannian metrics to avoid the shrinking problem.

Though graph cut based segmentation is a widely-used technique, it is known that segmentation of a thin, elongated structure is challenging due to the "shrinking problem". On the other hand, many segmentation targets in medical image analysis have such thin structures. Therefore, the conventional graph cut method is not suitable to be applied to them. In this study, we developed a graph cut segmentation method with novel Riemannian metrics. The Riemannian metrics are determined from the given "initial contour," so that any level-set surface of the distance transformation of the contour has the same surface area in the Riemannian space. This will ensure that any shape similar to the initial contour will not be affected by the shrinking problem. The method was evaluated with clinical CT datasets and showed a fair result in segmenting vertebral bones.

Texture analysis, bone mineral density, and cortical thickness of the proximal femur: fracture risk prediction.

OBJECTIVE: The objectives of this study were to perform a clinical study analyzing bone quality in multidetector computed tomographic images of the femur using bone mineral density (BMD), cortical thickness, and texture algorithms in differentiating osteoporotic fracture and control subjects; to differentiate fracture types. METHODS: Femoral head, trochanteric, intertrochanteric, and upper and lower neck were segmented (fracture, n = 30; control, n = 10). Cortical thickness, BMD, and texture analysis were obtained using co-occurrence matrices, Minkowski dimension, and functional and scaling index method. RESULTS: Bone mineral density and cortical thickness performed best in the neck region, and texture measures performed best in the trochanter. Only cortical thickness and texture measures differentiated femoral neck and intertrochanteric fractures. CONCLUSIONS: This study demonstrates that differentiation of osteoporotic fracture subjects and controls is achieved with texture measures, cortical thickness, and BMD; however, performance is region specific.

eHealth vision towards cooperative patient care--domain fields and architectural challenges of regional health care networks.

Numerous eHealth projects and efforts to establish inter-organizational communication and to build up regional health care networks could be observed in the last ten years. Nevertheless the success of such efforts is profoundly different. The aim of this paper is to introduce the lately started regional initiative eHealth.Braunschweig compounding of the major health care players (hospitals, physician offices, nursing services and nursing homes) in the region of Braunschweig, participants from research institutions and industry. We propose in this paper the main goals of the regional initiative eHealth.Braunschweig, its constitution and major approaches. Based on respective literature and our former projects as well as experiences in this field we discuss our vision of a patient-oriented cooperative health care by depicting regional distinctions, identifying the major domain fields in this context and discussing the architectural challenges for the regional health care network eHealth.Braunschweig. In our view this work can be considered as a systematical approach to the establishment of regional health care networks with lasting and sustainable effects on patient-centered health care in a region.

Assessment of the individual fracture risk of the proximal femur by using statistical appearance models.

PURPOSE: Standard diagnostic techniques to quantify bone mineral density (BMD) include dual-energy x-ray absorptiometry (DXA) and quantitative computed tomography. However, BMD alone is not sufficient to predict the fracture risk for an individual patient. Therefore, the development of tools, which can assess the bone quality in order to predict individual biomechanics of a bone, would mean a significant improvement for the prevention of fragility fractures. In this study, a new approach to predict the fracture risk of proximal femora using a statistical appearance model will be presented. METHODS: 100 CT data sets of human femur cadaver specimens are used to create statistical appearance models for the prediction of the individual fracture load (FL). Calculating these models offers the possibility to use information about the inner structure of the proximal femur, as well as geometric properties of the femoral bone for FL prediction. By applying principal component analysis, statistical models have been calculated in different regions of interest. For each of these models, the individual model parameters for each single data set were calculated and used as predictor variables in a multilinear regression model. By this means, the best working region of interest for the prediction of FL was identified. The accuracy of the FL prediction was evaluated by using a leave-one-out cross validation scheme. Performance of DXA in predicting FL was used as a standard of comparison. RESULTS: The results of the evaluative tests demonstrate that significantly better results for FL prediction can be achieved by using the proposed model-based approach (R = 0.91) than using DXA-BMD (R = 0.81) for the prediction of fracture load. CONCLUSIONS: The results of the evaluation show that the presented model-based approach is very promising and also comparable to studies that partly used higher image resolutions for bone quality assessment and fracture risk prediction.

Analysis of the micro-migration of sliding hip screws by using point-based registration.

PURPOSE: The favored treatment for many hip fractures is a sliding hip screw, and its usage is expected to increase in the future. Failures can be reduced, and complications detected earlier by semi-automated CT image analysis. The most frequent failure is due to the screw cut-out from the femoral head. METHODS: An image-based method was developed for early detection of complications and assessment of anchorage quality relative to implant model, bone quality or tip-apex distance (TAD). This method evaluates micro-migration using CT images acquired at different time points (immediately post-op and 3-month later). Serial CT image registration and transformation methods were applied, including point-based registration, to achieve semi-automated evaluations. RESULTS: Qualitative and quantitative validation of the image registration was performed with measurement mean error determination by different observers. The micro-migration evaluation by clinicians compared favorably with semi-automated image-based results. CONCLUSION: Semi-automatic evaluation of hip screw micro-migration using CT images is feasible and can aid observation of convalescence. The method may be amenable to full automation, a future goal for this work.

Statistical model based analysis of bone mineral density of lumbar spine.

OBJECTIVE: For planning surgical interventions at the spine affected by osteoporosis, accurate information about the local bone quality in terms of anchorage strength for implants is very important. Based on previous work on automated bone quality assessment on the proximal femur with a completely automated model-based approach, this paper describes first applications and results on the lumbar vertebrae. MATERIALS AND METHODS: As basis for the analysis, CT datasets of 17 spinal specimens, with a resolution of 0.7 mm x 0.7 mm x 0.7 mm have been used. A combined statistical model of 3D shape and intensity value distribution was created for these datasets and used to predict the measured bone mineral density (BMD). Different regions of interest were tested, model parameters with high correlation with BMD were identified. Leave-one-out tests were performed to evaluate the capability for the BMD-prediction using regression models. RESULTS: High correlation values (R = 0.94) between measured and predicted BMD were achieved and the high predictive quality of the model could be shown. CONCLUSION: Although the results are only valid for an insufficient small sample size of specimen data, they show a clear potential for clinical application. Therefore, work in the future will focus on clinical validation with larger sample size and the inclusion of biomechanical properties in addition to BMD.

Trabecular bone analysis in CT and X-ray images of the proximal femur for the assessment of local bone quality.

Currently, conventional X-ray and CT images as well as invasive methods performed during the surgical intervention are used to judge the local quality of a fractured proximal femur. However, these approaches are either dependent on the surgeon's experience or cannot assist diagnostic and planning tasks preoperatively. Therefore, in this work a method for the individual analysis of local bone quality in the proximal femur based on model-based analysis of CT- and X-ray images of femur specimen will be proposed. A combined representation of shape and spatial intensity distribution of an object and different statistical approaches for dimensionality reduction are used to create a statistical appearance model in order to assess the local bone quality in CT and X-ray images. The developed algorithms are tested and evaluated on 28 femur specimen. It will be shown that the tools and algorithms presented herein are highly adequate to automatically and objectively predict bone mineral density values as well as a biomechanical parameter of the bone that can be measured intraoperatively.