•Subcortical segmentation is performed in 3D fetal brain US with a 3D CNN.•High performance can be achieved using only nine manually annotated US volumes.•Pre-alignment increases segmentation ...performance but is not essential.•Subcortical growth curves during the second trimester of gestation are presented.•The cerebellar volume trajectories are in line with previous publications
Display omitted
The quantification of subcortical volume development from 3D fetal ultrasound can provide important diagnostic information during pregnancy monitoring. However, manual segmentation of subcortical structures in ultrasound volumes is time-consuming and challenging due to low soft tissue contrast, speckle and shadowing artifacts. For this reason, we developed a convolutional neural network (CNN) for the automated segmentation of the choroid plexus (CP), lateral posterior ventricle horns (LPVH), cavum septum pellucidum et vergae (CSPV), and cerebellum (CB) from 3D ultrasound. As ground-truth labels are scarce and expensive to obtain, we applied few-shot learning, in which only a small number of manual annotations (n = 9) are used to train a CNN. We compared training a CNN with only a few individually annotated volumes versus many weakly labelled volumes obtained from atlas-based segmentations. This showed that segmentation performance close to intra-observer variability can be obtained with only a handful of manual annotations. Finally, the trained models were applied to a large number (n = 278) of ultrasound image volumes of a diverse, healthy population, obtaining novel US-specific growth curves of the respective structures during the second trimester of gestation.
Objective: The segmentation of the breast from the chest wall is an important first step in the analysis of breast magnetic resonance images. 3D U-Nets have been shown to obtain high segmentation ...accuracy and appear to generalize well when trained on one scanner type and tested on another scanner, provided that a very similar MR protocol is used. There has, however, been little work addressing the problem of domain adaptation when image intensities or patient orientation differ markedly between the training set and an unseen test set. In this work we aim to address this domain shift problem. Method: We propose to apply extensive intensity augmentation in addition to geometric augmentation during training. We explored both style transfer and a novel intensity remapping approach as intensity augmentation strategies. For our experiments, we trained a 3D U-Net on T1-weighted scans. We tested our network on T2-weighted scans from the same dataset as well as on an additional independent test set acquired with a T1-weighted TWIST sequence and a different coil configuration. Results: By applying intensity augmentation we increased segmentation performance for the T2-weighted scans from a Dice of 0.71 to 0.88. This performance is very close to the baseline performance of training with T2-weighted scans (0.92). On the T1-weighted dataset we obtained a performance increase from 0.77 to 0.85. Conclusion: Our results show that the proposed intensity augmentation increases segmentation performance across different datasets. Significance: The proposed method can improve whole breast segmentation of clinical MR scans acquired with different protocols.
Three-dimensional (3D) ultrasound imaging has contributed to our understanding of fetal developmental processes by providing rich contextual information of the inherently 3D anatomies. However, its ...use is limited in clinical settings, due to the high purchasing costs and limited diagnostic practicality. Freehand 2D ultrasound imaging, in contrast, is routinely used in standard obstetric exams, but inherently lacks a 3D representation of the anatomies, which limits its potential for more advanced assessment. Such full representations are challenging to recover even with external tracking devices due to internal fetal movement which is independent from the operator-led trajectory of the probe. Capitalizing on the flexibility offered by freehand 2D ultrasound acquisition, we propose ImplicitVol to reconstruct 3D volumes from non-sensor-tracked 2D ultrasound sweeps. Conventionally, reconstructions are performed on a discrete voxel grid. We, however, employ a deep neural network to represent, for the first time, the reconstructed volume as an implicit function. Specifically, ImplicitVol takes a set of 2D images as input, predicts their locations in 3D space, jointly refines the inferred locations, and learns a full volumetric reconstruction. When testing natively-acquired and volume-sampled 2D ultrasound video sequences collected from different manufacturers, the 3D volumes reconstructed by ImplicitVol show significantly better visual and semantic quality than the existing interpolation-based reconstruction approaches. The inherent continuity of implicit representation also enables ImplicitVol to reconstruct the volume to arbitrarily high resolutions. As formulated, ImplicitVol has the potential to integrate seamlessly into the clinical workflow, while providing richer information for diagnosis and evaluation of the developing brain.
Display omitted
•We propose a framework for sensorless 3D reconstruction from freehand 2D ultrasound.•External positional sensor is impractical due to uncorrelated fetal and probe motion.•Our framework uses implicit neural representation for the 2D-to-3D reconstruction.•It outperforms the baselines on natively-acquired and volume-sampled 2D scans.•Have better visual and semantic quality than existing methods (interpolation-based).
Maturation of the human fetal brain should follow precisely scheduled structural growth and folding of the cerebral cortex for optimal postnatal function
. We present a normative digital atlas of ...fetal brain maturation based on a prospective international cohort of healthy pregnant women
, selected using World Health Organization recommendations for growth standards
. Their fetuses were accurately dated in the first trimester, with satisfactory growth and neurodevelopment from early pregnancy to 2 years of age
. The atlas was produced using 1,059 optimal quality, three-dimensional ultrasound brain volumes from 899 of the fetuses and an automated analysis pipeline
. The atlas corresponds structurally to published magnetic resonance images
, but with finer anatomical details in deep grey matter. The between-study site variability represented less than 8.0% of the total variance of all brain measures, supporting pooling data from the eight study sites to produce patterns of normative maturation. We have thereby generated an average representation of each cerebral hemisphere between 14 and 31 weeks' gestation with quantification of intracranial volume variability and growth patterns. Emergent asymmetries were detectable from as early as 14 weeks, with peak asymmetries in regions associated with language development and functional lateralization between 20 and 26 weeks' gestation. These patterns were validated in 1,487 three-dimensional brain volumes from 1,295 different fetuses in the same cohort. We provide a unique spatiotemporal benchmark of fetal brain maturation from a large cohort with normative postnatal growth and neurodevelopment.
Prototype Learning for Explainable Brain Age Prediction Hesse, Linde S.; Dinsdale, Nicola K.; Namburete, Ana I.L.
2024 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV),
2024-Jan.-3
Conference Proceeding
The lack of explainability of deep learning models limits the adoption of such models in clinical practice. Prototype-based models can provide inherent explainable predictions, but these have ...predominantly been designed for classification tasks, despite many important tasks in medical imaging being continuous regression problems. Therefore, in this work, we present ExPeRT: an explainable prototype-based model specifically designed for regression tasks. Our proposed model makes a sample prediction from the distances to a set of learned prototypes in latent space, using a weighted mean of prototype labels. The distances in latent space are regularized to be relative to label differences, and each of the prototypes can be visualized as a sample from the training set. The image-level distances are further constructed from patch-level distances, in which the patches of both images are structurally matched using optimal transport. This thus provides an example-based explanation with patch-level detail at inference time. We demonstrate our proposed model for brain age prediction on two imaging datasets: adult MR and fetal ultrasound. Our approach achieved state-of-the-art prediction performance while providing insight into the model's reasoning process.
The lack of explainability of deep learning models limits the adoption of such models in clinical practice. Prototype-based models can provide inherent explainable predictions, but these have ...predominantly been designed for classification tasks, despite many important tasks in medical imaging being continuous regression problems. Therefore, in this work, we present ExPeRT: an explainable prototype-based model specifically designed for regression tasks. Our proposed model makes a sample prediction from the distances to a set of learned prototypes in latent space, using a weighted mean of prototype labels. The distances in latent space are regularized to be relative to label differences, and each of the prototypes can be visualized as a sample from the training set. The image-level distances are further constructed from patch-level distances, in which the patches of both images are structurally matched using optimal transport. This thus provides an example-based explanation with patch-level detail at inference time. We demonstrate our proposed model for brain age prediction on two imaging datasets: adult MR and fetal ultrasound. Our approach achieved state-of-the-art prediction performance while providing insight into the model's reasoning process.
Convolutional neural networks (CNNs) have shown exceptional performance for a range of medical imaging tasks. However, conventional CNNs are not able to explain their reasoning process, therefore ...limiting their adoption in clinical practice. In this work, we propose an inherently interpretable CNN for regression using similarity-based comparisons (INSightR-Net) and demonstrate our methods on the task of diabetic retinopathy grading. A prototype layer incorporated into the architecture enables visualization of the areas in the image that are most similar to learned prototypes. The final prediction is then intuitively modeled as a mean of prototype labels, weighted by the similarities. We achieved competitive prediction performance with our INSightR-Net compared to a ResNet baseline, showing that it is not necessary to compromise performance for interpretability. Furthermore, we quantified the quality of our explanations using sparsity and diversity, two concepts considered important for a good explanation, and demonstrated the effect of several parameters on the latent space embeddings.
The segmentation of the breast from the chest wall is an important first step in the analysis of breast magnetic resonance images. 3D U-nets have been shown to obtain high segmentation accuracy and ...appear to generalize well when trained on one scanner type and tested on another scanner, provided that a very similar T1-weighted MR protocol is used. There has, however, been little work addressing the problem of domain adaptation when image intensities or patient orientation differ markedly between the training set and an unseen test set. To overcome the domain shift we propose to apply extensive intensity augmentation in addition to geometric augmentation during training. We explored both style transfer and a novel intensity remapping approach as intensity augmentation strategies. For our experiments, we trained a 3D U-net on T1-weighted scans and tested on T2-weighted scans. By applying intensity augmentation we increased segmentation performance from a DSC of 0.71 to 0.90. This performance is very close to the baseline performance of training and testing on T2-weighted scans (0.92). Furthermore, we applied our network to an independent test set made up of publicly available scans acquired using a T1-weighted TWIST sequence and a different coil configuration. On this dataset we obtained a performance of 0.89, close to the inter-observer variability of the ground truth segmentations (0.92). Our results show that using intensity augmentation in addition to geometric augmentation is a suitable method to overcome the intensity domain shift and we expect it to be useful for a wide range of segmentation tasks.
Early detection of lung cancer has been proven to decrease mortality significantly. A recent development in computed tomography (CT), spectral CT, can potentially improve diagnostic accuracy, as it ...yields more information per scan than regular CT. However, the shear workload involved with analyzing a large number of scans drives the need for automated diagnosis methods. Therefore, we propose a detection and classification system for lung nodules in CT scans. Furthermore, we want to observe whether spectral images can increase classifier performance. For the detection of nodules we trained a VGG-like 3D convolutional neural net (CNN). To obtain a primary tumor classifier for our dataset we pre-trained a 3D CNN with similar architecture on nodule malignancies of a large publicly available dataset, the LIDC-IDRI dataset. Subsequently we used this pre-trained network as feature extractor for the nodules in our dataset. The resulting feature vectors were classified into two (benign/malignant) and three (benign/primary lung cancer/metastases) classes using support vector machine (SVM). This classification was performed both on nodule- and scan-level. We obtained state-of-the art performance for detection and malignancy regression on the LIDC-IDRI database. Classification performance on our own dataset was higher for scan- than for nodule-level predictions. For the three-class scan-level classification we obtained an accuracy of 78\%. Spectral features did increase classifier performance, but not significantly. Our work suggests that a pre-trained feature extractor can be used as primary tumor origin classifier for lung nodules, eliminating the need for elaborate fine-tuning of a new network and large datasets. Code is available at \url{https://github.com/tueimage/lung-nodule-msc-2018}.