Joel Kullberg

Eva Breznik, Hoel Kervadec, Filip Malmberg et al.

This paper investigates the combination of intensity-based distance maps with boundary loss for point-supervised semantic segmentation. By design the boundary loss imposes a stronger penalty on the false positives the farther away from the object they occur. Hence it is intuitively inappropriate for weak supervision, where the ground truth label may be much smaller than the actual object and a certain amount of false positives (w.r.t. the weak ground truth) is actually desirable. Using intensity-aware distances instead may alleviate this drawback, allowing for a certain amount of false positives without a significant increase to the training loss. The motivation for applying the boundary loss directly under weak supervision lies in its great success for fully supervised segmentation tasks, but also in not requiring extra priors or outside information that is usually required -- in some form -- with existing weakly supervised methods in the literature. This formulation also remains potentially more attractive than existing CRF-based regularizers, due to its simplicity and computational efficiency. We perform experiments on two multi-class datasets; ACDC (heart segmentation) and POEM (whole-body abdominal organ segmentation). Preliminary results are encouraging and show that this supervision strategy has great potential. On ACDC it outperforms the CRF-loss based approach, and on POEM data it performs on par with it. The code for all our experiments is openly available.

Sambit Tarai, Ashish Chauhan, Elin Lundström et al.

Purpose: Automated medical image-based prediction of clinical outcomes, such as overall survival (OS), has great potential in improving patient prognostics and personalized treatment planning. We developed a deep regression framework using tissue-wise FDG-PET/CT projections as input, along with a temporal input representing a scalar time horizon (in days) to predict OS in patients with Non-Small Cell Lung Cancer (NSCLC). Methods: The proposed framework employed a ResNet-50 backbone to process input images and generate corresponding image embeddings. The embeddings were then combined with temporal data to produce OS probabilities as a function of time, effectively parameterizing the predictions based on time. The overall framework was developed using the U-CAN cohort (n = 556) and evaluated by comparing with a baseline method on the test set (n = 292). The baseline utilized the ResNet-50 architecture, processing only the images as input and providing OS predictions at pre-specified intervals, such as 2- or 5-year. Results: The incorporation of temporal data with image embeddings demonstrated an advantage in predicting OS, outperforming the baseline method with an improvement in AUC of 4.3%. The proposed model using clinical + IDP features achieved strong performance, and an ensemble of imaging and clinical + IDP models achieved the best overall performance (0.788), highlighting the complementary value of multimodal inputs. The proposed method also enabled risk stratification of patients into distinct categories (high vs low risk). Heat maps from the saliency analysis highlighted tumor regions as key structures for the prediction. Conclusion: Our method provided an automated framework for predicting OS as a function of time and demonstrates the potential of combining imaging and tabular data for improved survival prediction.

Yasemin Utkueri, Elin Lundström, Håkan Ahlström et al.

The UK Biobank is a large-scale study collecting whole-body MR imaging and non-imaging health data. Robust and accurate inter-subject image registration of these whole-body MR images would enable their body-wide spatial standardization, and region-/voxel-wise correlation analysis of non-imaging data with image-derived parameters (e.g., tissue volume or fat content). We propose a sex-stratified inter-subject whole-body MR image registration approach that uses subcutaneous adipose tissue- and muscle-masks from the state-of-the-art VIBESegmentator method to augment intensity-based graph-cut registration. The proposed method was evaluated on a subset of 4000 subjects by comparing it to an intensity-only method as well as two previously published registration methods, uniGradICON and MIRTK. The evaluation comprised overlap measures applied to the 71 VIBESegmentator masks: 1) Dice scores, and 2) voxel-wise label error frequency. Additionally, voxel-wise correlation between age and each of fat content and tissue volume was studied to exemplify the usefulness for medical research. The proposed method exhibited a mean dice score of 0.77 / 0.75 across the cohort and the 71 masks for males/females, respectively. When compared to the intensity-only registration, the mean values were 6 percentage points (pp) higher for both sexes, and the label error frequency was decreased in most tissue regions. These differences were 9pp / 8pp against uniGradICON and 12pp / 13pp against MIRTK. Using the proposed method, the age-correlation maps were less noisy and showed higher anatomical alignment. In conclusion, the image registration method using two tissue masks improves whole-body registration of UK Biobank images.

Johan Öfverstedt, Elin Lundström, Håkan Ahlström et al.

Age prediction from medical images or other health-related non-imaging data is an important approach to data-driven aging research, providing knowledge of how much information a specific tissue or organ carries about the chronological age of the individual. In this work, we studied the prediction of age from computed tomography angiography (CTA) images, which provide detailed representations of the heart morphology, with the goals of (i) studying the relationship between morphology and aging, and (ii) developing a novel \emph{morphological heart age} biomarker. We applied an image registration-based method that standardizes the images from the whole cohort into a single space. We then extracted supervoxels (using unsupervised segmentation), and corresponding robust features of density and local volume, which provide a detailed representation of the heart morphology while being robust to registration errors. Machine learning models are then trained to fit regression models from these features to the chronological age. We applied the method to a subset of the images from the Swedish CArdioPulomonary bioImage Study (SCAPIS) dataset, consisting of 721 females and 666 males. We observe a mean absolute error of $2.74$ years for females and $2.77$ years for males. The predictions from different sub-regions of interest were observed to be more highly correlated with the predictions from the whole heart, compared to the chronological age, revealing a high consistency in the predictions from morphology. Saliency analysis was also performed on the prediction models to study what regions are associated positively and negatively with the predicted age. This resulted in detailed association maps where the density and volume of known, as well as some novel sub-regions of interest, are determined to be important. The saliency analysis aids in the interpretability of the models and their predictions.

Johan Öfverstedt, Elin Lundström, Göran Bergström et al.

The study of associations between an individual's age and imaging and non-imaging data is an active research area that attempts to aid understanding of the effects and patterns of aging. In this work we have conducted a supervoxel-wise association study between both volumetric and tissue density features in coronary computed tomography angiograms and the chronological age of a subject, to understand the localized changes in morphology and tissue density with age. To enable a supervoxel-wise study of volume and tissue density, we developed a novel method based on image segmentation, inter-subject image registration, and robust supervoxel-based correlation analysis, to achieve a statistical association study between the images and age. We evaluate the registration methodology in terms of the Dice coefficient for the heart chambers and myocardium, and the inverse consistency of the transformations, showing that the method works well in most cases with high overlap and inverse consistency. In a sex-stratified study conducted on a subset of $n=1388$ images from the SCAPIS study, the supervoxel-wise analysis was able to find localized associations with age outside of the commonly segmented and analyzed sub-regions, and several substantial differences between the sexes in the association of age and volume.

Taro Langner, Andrés Martínez Mora, Robin Strand et al.

UK Biobank (UKB) conducts large-scale examinations of more than half a million volunteers, collecting health-related information on genetics, lifestyle, blood biochemistry, and more. Medical imaging of 100,000 subjects, with 70,000 follow-up sessions, enables measurements of organs, muscle, and body composition. With up to 170,000 mounting MR images, various methodologies are accordingly engaged in large-scale image analysis. This work presents an experimental inference engine that can automatically predict a comprehensive profile of subject metadata from UKB neck-to-knee body MRI. It was evaluated in cross-validation for baseline characteristics such as age, height, weight, and sex, but also measurements of body composition, organ volumes, and abstract properties like grip strength, pulse rate, and type 2 diabetic status. It predicted subsequently released test data covering twelve body composition metrics with a 3% median error. The proposed system can automatically analyze one thousand subjects within ten minutes, providing individual confidence intervals. The underlying methodology utilizes convolutional neural networks for image-based mean-variance regression on two-dimensional representations of the MRI data. This work aims to make the proposed system available for free to researchers, who can use it to obtain fast and fully-automated estimates of 72 different measurements immediately upon release of new UKB image data.

Taro Langner, Robin Strand, Håkan Ahlström et al.

Large-scale medical studies such as the UK Biobank examine thousands of volunteer participants with medical imaging techniques. Combined with the vast amount of collected metadata, anatomical information from these images has the potential for medical analyses at unprecedented scale. However, their evaluation often requires manual input and long processing times, limiting the amount of reference values for biomarkers and other measurements available for research. Recent approaches with convolutional neural networks for regression can perform these evaluations automatically. On magnetic resonance imaging (MRI) data of more than 40,000 UK Biobank subjects, these systems can estimate human age, body composition and more. This style of analysis is almost entirely data-driven and no manual intervention or guidance with manually segmented ground truth images is required. The networks often closely emulate the reference method that provided their training data and can reach levels of agreement comparable to the expected variability between established medical gold standard techniques. The risk of silent failure can be individually quantified by predictive uncertainty obtained from a mean-variance criterion and ensembling. Saliency analysis furthermore enables an interpretation of the underlying relevant image features and showed that the networks learned to correctly target specific organs, limbs, and regions of interest.

Taro Langner, Fredrik K. Gustafsson, Benny Avelin et al.

Along with rich health-related metadata, medical images have been acquired for over 40,000 male and female UK Biobank participants, aged 44-82, since 2014. Phenotypes derived from these images, such as measurements of body composition from MRI, can reveal new links between genetics, cardiovascular disease, and metabolic conditions. In this work, six measurements of body composition and adipose tissues were automatically estimated by image-based, deep regression with ResNet50 neural networks from neck-to-knee body MRI. Despite the potential for high speed and accuracy, these networks produce no output segmentations that could indicate the reliability of individual measurements. The presented experiments therefore examine uncertainty quantification with mean-variance regression and ensembling to estimate individual measurement errors and thereby identify potential outliers, anomalies, and other failure cases automatically. In 10-fold cross-validation on data of about 8,500 subjects, mean-variance regression and ensembling showed complementary benefits, reducing the mean absolute error across all predictions by 12%. Both improved the calibration of uncertainties and their ability to identify high prediction errors. With intra-class correlation coefficients (ICC) above 0.97, all targets except the liver fat content yielded relative measurement errors below 5%. Testing on another 1,000 subjects showed consistent performance, and the method was finally deployed for inference to 30,000 subjects with missing reference values. The results indicate that deep regression ensembles could ultimately provide automated, uncertainty-aware measurements of body composition for more than 120,000 UK Biobank neck-to-knee body MRI that are to be acquired within the coming years.

Taro Langner, Robin Strand, Håkan Ahlström et al.

The UK Biobank Imaging Study has acquired medical scans of more than 40,000 volunteer participants. The resulting wealth of anatomical information has been made available for research, together with extensive metadata including measurements of liver fat. These values play an important role in metabolic disease, but are only available for a minority of imaged subjects as their collection requires the careful work of image analysts on dedicated liver MRI. Another UK Biobank protocol is neck-to-knee body MRI for analysis of body composition. The resulting volumes can also quantify fat fractions, even though they were reconstructed with a two- instead of a three-point Dixon technique. In this work, a novel framework for automated inference of liver fat from UK Biobank neck-to-knee body MRI is proposed. A ResNet50 was trained for regression on two-dimensional slices from these scans and the reference values as target, without any need for ground truth segmentations. Once trained, it performs fast, objective, and fully automated predictions that require no manual intervention. On the given data, it closely emulates the reference method, reaching a level of agreement comparable to different gold standard techniques. The network learned to rectify non-linearities in the fat fraction values and identified several outliers in the reference. It outperformed a multi-atlas segmentation baseline and inferred new estimates for all imaged subjects lacking reference values, expanding the total number of liver fat measurements by factor six.

Taro Langner, Andreas Östling, Lukas Maldonis et al.

The UK Biobank is collecting extensive data on health-related characteristics of over half a million volunteers. The biological samples of blood and urine can provide valuable insight on kidney function, with important links to cardiovascular and metabolic health. Further information on kidney anatomy could be obtained by medical imaging. In contrast to the brain, heart, liver, and pancreas, no dedicated Magnetic Resonance Imaging (MRI) is planned for the kidneys. An image-based assessment is nonetheless feasible in the neck-to-knee body MRI intended for abdominal body composition analysis, which also covers the kidneys. In this work, a pipeline for automated segmentation of parenchymal kidney volume in UK Biobank neck-to-knee body MRI is proposed. The underlying neural network reaches a relative error of 3.8%, with Dice score 0.956 in validation on 64 subjects, close to the 2.6% and Dice score 0.962 for repeated segmentation by one human operator. The released MRI of about 40,000 subjects can be processed within two days, yielding volume measurements of left and right kidney. Algorithmic quality ratings enabled the exclusion of outliers and potential failure cases. The resulting measurements can be studied and shared for large-scale investigation of associations and longitudinal changes in parenchymal kidney volume.

Taro Langner, Robin Strand, Håkan Ahlström et al.

In a large-scale medical examination, the UK Biobank study has successfully imaged more than 32,000 volunteer participants with magnetic resonance imaging (MRI). Each scan is linked to extensive metadata, providing a comprehensive medical survey of imaged anatomy and related health states. Despite its potential for research, this vast amount of data presents a challenge to established methods of evaluation, which often rely on manual input. To date, the range of reference values for cardiovascular and metabolic risk factors is therefore incomplete. In this work, neural networks were trained for image-based regression to infer various biological metrics from the neck-to-knee body MRI automatically. The approach requires no manual intervention or direct access to reference segmentations for training. The examined fields span 64 variables derived from anthropometric measurements, dual-energy X-ray absorptiometry (DXA), atlas-based segmentations, and dedicated liver scans. With the ResNet50, the standardized framework achieves a close fit to the target values (median R^2 > 0.97) in cross-validation. Interpretation of aggregated saliency maps suggests that the network correctly targets specific body regions and limbs, and learned to emulate different modalities. On several body composition metrics, the quality of the predictions is within the range of variability observed between established gold standard techniques.

Jonathan Andersson, Håkan Ahlström, Joel Kullberg

Purpose: To perform and evaluate water-fat signal separation of whole-body gradient echo scans using convolutional neural networks. Methods: Whole-body gradient echo scans of 240 subjects, each consisting of 5 bipolar echoes, were used. Reference fat fraction maps were created using a conventional method. Convolutional neural networks, more specifically 2D U-nets, were trained using 5-fold cross-validation with 1 or several echoes as input, using the squared difference between the output and the reference fat fraction maps as the loss function. The outputs of the networks were assessed by the loss function, measured liver fat fractions, and visually. Training was performed using a graphics processing unit (GPU). Inference was performed using the GPU as well as a central processing unit (CPU). Results: The loss curves indicated convergence, and the final loss of the validation data decreased when using more echoes as input. The liver fat fractions could be estimated using only 1 echo, but results were improved by use of more echoes. Visual assessment found the quality of the outputs of the networks to be similar to the reference even when using only 1 echo, with slight improvements when using more echoes. Training a network took at most 28.6 h. Inference time of a whole-body scan took at most 3.7 s using the GPU and 5.8 min using the CPU. Conclusion: It is possible to perform water-fat signal separation of whole-body gradient echo scans using convolutional neural networks. Separation was possible using only 1 echo, although using more echoes improved the results.

Simon Ekström, Filip Malmberg, Håkan Ahlström et al.

Objective: Deformable image registration is a fundamental problem in medical image analysis, with applications such as longitudinal studies, population modeling, and atlas based image segmentation. Registration is often phrased as an optimization problem, i.e., finding a deformation field that is optimal according to a given objective function. Discrete, combinatorial, optimization techniques have successfully been employed to solve the resulting optimization problem. Specifically, optimization based on $α$-expansion with minimal graph cuts has been proposed as a powerful tool for image registration. The high computational cost of the graph-cut based optimization approach, however, limits the utility of this approach for registration of large volume images. Methods: Here, we propose to accelerate graph-cut based deformable registration by dividing the image into overlapping sub-regions and restricting the $α$-expansion moves to a single sub-region at a time. Results: We demonstrate empirically that this approach can achieve a large reduction in computation time -- from days to minutes -- with only a small penalty in terms of solution quality. Conclusion: The reduction in computation time provided by the proposed method makes graph cut based deformable registration viable for large volume images. Significance: Graph cut based image registration has previously been shown to produce excellent results, but the high computational cost has hindered the adoption of the method for registration of large medical volume images. Our proposed method lifts this restriction, requiring only a small fraction of the computational cost to produce results of comparable quality.

Taro Langner, Anders Hedström, Katharina Mörwald et al.

Purpose: An approach for the automated segmentation of visceral adipose tissue (VAT) and subcutaneous adipose tissue (SAT) in multicenter water-fat MRI scans of the abdomen was investigated, using two different neural network architectures. Methods: The two fully convolutional network architectures U-Net and V-Net were trained, evaluated and compared on the water-fat MRI data. Data of the study Tellus with 90 scans from a single center was used for a 10-fold cross-validation in which the most successful configuration for both networks was determined. These configurations were then tested on 20 scans of the multicenter study beta-cell function in JUvenile Diabetes and Obesity (BetaJudo), which involved a different study population and scanning device. Results: The U-Net outperformed the used implementation of the V-Net in both cross-validation and testing. In cross-validation, the U-Net reached average dice scores of 0.988 (VAT) and 0.992 (SAT). The average of the absolute quantification errors amount to 0.67% (VAT) and 0.39% (SAT). On the multi-center test data, the U-Net performs only slightly worse, with average dice scores of 0.970 (VAT) and 0.987 (SAT) and quantification errors of 2.80% (VAT) and 1.65% (SAT). Conclusion: The segmentations generated by the U-Net allow for reliable quantification and could therefore be viable for high-quality automated measurements of VAT and SAT in large-scale studies with minimal need for human intervention. The high performance on the multicenter test data furthermore shows the robustness of this approach for data of different patient demographics and imaging centers, as long as a consistent imaging protocol is used.