Eli Gibson

Han Liu, Bogdan Georgescu, Yanbo Zhang et al.

3D medical image classification is essential for modern clinical workflows. Medical foundation models (FMs) have emerged as a promising approach for scaling to new tasks, yet current research suffers from three critical pitfalls: data-regime bias, suboptimal adaptation, and insufficient task coverage. In this paper, we address these pitfalls and introduce AnyMC3D, a scalable 3D classifier adapted from 2D FMs. Our method scales efficiently to new tasks by adding only lightweight plugins (about 1M parameters per task) on top of a single frozen backbone. This versatile framework also supports multi-view inputs, auxiliary pixel-level supervision, and interpretable heatmap generation. We establish a comprehensive benchmark of 12 tasks covering diverse pathologies, anatomies, and modalities, and systematically analyze state-of-the-art 3D classification techniques. Our analysis reveals key insights: (1) effective adaptation is essential to unlock FM potential, (2) general-purpose FMs can match medical-specific FMs if properly adapted, and (3) 2D-based methods surpass 3D architectures for 3D classification. For the first time, we demonstrate the feasibility of achieving state-of-the-art performance across diverse applications using a single scalable framework (including 1st place in the VLM3D challenge), eliminating the need for separate task-specific models.

Eli Gibson, Wenqi Li, Carole Sudre et al.

Medical image analysis and computer-assisted intervention problems are increasingly being addressed with deep-learning-based solutions. Established deep-learning platforms are flexible but do not provide specific functionality for medical image analysis and adapting them for this application requires substantial implementation effort. Thus, there has been substantial duplication of effort and incompatible infrastructure developed across many research groups. This work presents the open-source NiftyNet platform for deep learning in medical imaging. The ambition of NiftyNet is to accelerate and simplify the development of these solutions, and to provide a common mechanism for disseminating research outputs for the community to use, adapt and build upon. NiftyNet provides a modular deep-learning pipeline for a range of medical imaging applications including segmentation, regression, image generation and representation learning applications. Components of the NiftyNet pipeline including data loading, data augmentation, network architectures, loss functions and evaluation metrics are tailored to, and take advantage of, the idiosyncracies of medical image analysis and computer-assisted intervention. NiftyNet is built on TensorFlow and supports TensorBoard visualization of 2D and 3D images and computational graphs by default. We present 3 illustrative medical image analysis applications built using NiftyNet: (1) segmentation of multiple abdominal organs from computed tomography; (2) image regression to predict computed tomography attenuation maps from brain magnetic resonance images; and (3) generation of simulated ultrasound images for specified anatomical poses. NiftyNet enables researchers to rapidly develop and distribute deep learning solutions for segmentation, regression, image generation and representation learning applications, or extend the platform to new applications.

Youngjin Yoo, Bogdan Georgescu, Yanbo Zhang et al.

Recent advancements in AI and medical imaging offer transformative potential in emergency head CT interpretation for reducing assessment times and improving accuracy in the face of an increasing request of such scans and a global shortage in radiologists. This study introduces a 3D foundation model for detecting diverse neuro-trauma findings with high accuracy and efficiency. Using large language models (LLMs) for automatic labeling, we generated comprehensive multi-label annotations for critical conditions. Our approach involved pretraining neural networks for hemorrhage subtype segmentation and brain anatomy parcellation, which were integrated into a pretrained comprehensive neuro-trauma detection network through multimodal fine-tuning. Performance evaluation against expert annotations and comparison with CT-CLIP demonstrated strong triage accuracy across major neuro-trauma findings, such as hemorrhage and midline shift, as well as less frequent critical conditions such as cerebral edema and arterial hyperdensity. The integration of neuro-specific features significantly enhanced diagnostic capabilities, achieving an average AUC of 0.861 for 16 neuro-trauma conditions. This work advances foundation models in medical imaging, serving as a benchmark for future AI-assisted neuro-trauma diagnostics in emergency radiology.

Badhan Kumar Das, Ajay Singh, Gengyan Zhao et al.

Self-supervised pretrain techniques have been widely used to improve the downstream tasks' performance. However, real-world magnetic resonance (MR) studies usually consist of different sets of contrasts due to different acquisition protocols, which poses challenges for the current deep learning methods on large-scale pretrain and different downstream tasks with different input requirements, since these methods typically require a fixed set of input modalities or, contrasts. To address this challenge, we propose variable-input ViT (VIViT), a transformer-based framework designed for self-supervised pretraining and segmentation finetuning for variable contrasts in each study. With this ability, our approach can maximize the data availability in pretrain, and can transfer the learned knowledge from pretrain to downstream tasks despite variations in input requirements. We validate our method on brain infarct and brain tumor segmentation, where our method outperforms current CNN and ViT-based models with a mean Dice score of 0.624 and 0.883 respectively. These results highlight the efficacy of our design for better adaptability and performance on tasks with real-world heterogeneous MR data.

Badhan Kumar Das, Gengyan Zhao, Han Liu et al.

The Masked Autoencoder (MAE) has recently demonstrated effectiveness in pre-training Vision Transformers (ViT) for analyzing natural images. By reconstructing complete images from partially masked inputs, the ViT encoder gathers contextual information to predict the missing regions. This capability to aggregate context is especially important in medical imaging, where anatomical structures are functionally and mechanically linked to surrounding regions. However, current methods do not consider variations in the number of input images, which is typically the case in real-world Magnetic Resonance (MR) studies. To address this limitation, we propose a 3D Adaptive Masked Autoencoders (AMAE) architecture that accommodates a variable number of 3D input contrasts per subject. A magnetic resonance imaging (MRI) dataset of 45,364 subjects was used for pretraining and a subset of 1648 training, 193 validation and 215 test subjects were used for finetuning. The performance demonstrates that self pre-training of this adaptive masked autoencoders can enhance the infarct segmentation performance by 2.8%-3.7% for ViT-based segmentation models.

Parhom Esmaeili, Chayanin Tangwiriyasakul, Eli Gibson et al.

Interactive segmentation enables clinicians to guide annotation, but existing zero-shot models like nnInteractive fail to consistently reach expert-level performance across diverse medical imaging tasks. Because annotation campaigns produce a growing stream of task-specific labelled data, online adaptation of the segmentation model is a natural complement to zero-shot inference. We propose CLoPA, a continual adaptation strategy that tunes a small fraction of nnInteractive's parameters on the annotation cache, triggered by lightweight episode scheduling. CLoPA requires no new parameters or changes to the inference pipeline, and operates entirely within the existing annotation workflow. Across eight Medical Segmentation Decathlon tasks spanning diverse anatomical targets and imaging characteristics, CLoPA rapidly elevates performance to expert-level, even for tasks where nnInteractive previously failed, with the majority of gains realised after a single training episode. We show that the benefits of tuning different parameter groups depends on task characteristics and data regimes. Also, that for targets with complex geometries (e.g., hepatic vessels), instance normalisation and low-level feature tuning saturates, suggesting a need for deeper feature-representation alignment in the most challenging scenarios.

Parhom Esmaeili, Virginia Fernandez, Pedro Borges et al.

Interactive segmentation is a promising strategy for building robust, generalisable algorithms for volumetric medical image segmentation. However, inconsistent and clinically unrealistic evaluation hinders fair comparison and misrepresents real-world performance. We propose a clinically grounded methodology for defining evaluation tasks and metrics, and built a software framework for constructing standardised evaluation pipelines. We evaluate state-of-the-art algorithms across heterogeneous and complex tasks and observe that (i) minimising information loss when processing user interactions is critical for model robustness, (ii) adaptive-zooming mechanisms boost robustness and speed convergence, (iii) performance drops if validation prompting behaviour/budgets differ from training, (iv) 2D methods perform well with slab-like images and coarse targets, but 3D context helps with large or irregularly shaped targets, (v) performance of non-medical-domain models (e.g. SAM2) degrades with poor contrast and complex shapes.

Badhan Kumar Das, Gengyan Zhao, Boris Mailhe et al.

Identifying brain hemorrhages from magnetic resonance imaging (MRI) is a critical task for healthcare professionals. The diverse nature of MRI acquisitions with varying contrasts and orientation introduce complexity in identifying hemorrhage using neural networks. For acquisitions with varying orientations, traditional methods often involve resampling images to a fixed plane, which can lead to information loss. To address this, we propose a 3D multi-plane vision transformer (MP-ViT) for hemorrhage classification with varying orientation data. It employs two separate transformer encoders for axial and sagittal contrasts, using cross-attention to integrate information across orientations. MP-ViT also includes a modality indication vector to provide missing contrast information to the model. The effectiveness of the proposed model is demonstrated with extensive experiments on real world clinical dataset consists of 10,084 training, 1,289 validation and 1,496 test subjects. MP-ViT achieved substantial improvement in area under the curve (AUC), outperforming the vision transformer (ViT) by 5.5% and CNN-based architectures by 1.8%. These results highlight the potential of MP-ViT in improving performance for hemorrhage detection when different orientation contrasts are needed.

Badhan Kumar Das, Gengyan Zhao, Han Liu et al.

Pretrain techniques, whether supervised or self-supervised, are widely used in deep learning to enhance model performance. In real-world clinical scenarios, different sets of magnetic resonance (MR) contrasts are often acquired for different subjects/cases, creating challenges for deep learning models assuming consistent input modalities among all the cases and between pretrain and finetune. Existing methods struggle to maintain performance when there is an input modality/contrast set mismatch with the pretrained model, often resulting in degraded accuracy. We propose an adaptive Vision Transformer (AdaViT) framework capable of handling variable set of input modalities for each case. We utilize a dynamic tokenizer to encode different input image modalities to tokens and take advantage of the characteristics of the transformer to build attention mechanism across variable length of tokens. Through extensive experiments, we demonstrate that this architecture effectively transfers supervised pretrained models to new datasets with different input modality/contrast sets, resulting in superior performance on zero-shot testing, few-shot finetuning, and backward transferring in brain infarct and brain tumor segmentation tasks. Additionally, for self-supervised pretrain, the proposed method is able to maximize the pretrain data and facilitate transferring to diverse downstream tasks with variable sets of input modalities.

Florin C. Ghesu, Bogdan Georgescu, Awais Mansoor et al.

The interpretation of medical images is a challenging task, often complicated by the presence of artifacts, occlusions, limited contrast and more. Most notable is the case of chest radiography, where there is a high inter-rater variability in the detection and classification of abnormalities. This is largely due to inconclusive evidence in the data or subjective definitions of disease appearance. An additional example is the classification of anatomical views based on 2D Ultrasound images. Often, the anatomical context captured in a frame is not sufficient to recognize the underlying anatomy. Current machine learning solutions for these problems are typically limited to providing probabilistic predictions, relying on the capacity of underlying models to adapt to limited information and the high degree of label noise. In practice, however, this leads to overconfident systems with poor generalization on unseen data. To account for this, we propose a system that learns not only the probabilistic estimate for classification, but also an explicit uncertainty measure which captures the confidence of the system in the predicted output. We argue that this approach is essential to account for the inherent ambiguity characteristic of medical images from different radiologic exams including computed radiography, ultrasonography and magnetic resonance imaging. In our experiments we demonstrate that sample rejection based on the predicted uncertainty can significantly improve the ROC-AUC for various tasks, e.g., by 8% to 0.91 with an expected rejection rate of under 25% for the classification of different abnormalities in chest radiographs. In addition, we show that using uncertainty-driven bootstrapping to filter the training data, one can achieve a significant increase in robustness and accuracy.

Donghao Zhang, Siqi Liu, Shikha Chaganti et al.

With the injection of contrast material into blood vessels, multi-phase contrasted CT images can enhance the visibility of vessel networks in the human body. Reconstructing the 3D geometric morphology of liver vessels from the contrasted CT images can enable multiple liver preoperative surgical planning applications. Automatic reconstruction of liver vessel morphology remains a challenging problem due to the morphological complexity of liver vessels and the inconsistent vessel intensities among different multi-phase contrasted CT images. On the other side, high integrity is required for the 3D reconstruction to avoid decision making biases. In this paper, we propose a framework for liver vessel morphology reconstruction using both a fully convolutional neural network and a graph attention network. A fully convolutional neural network is first trained to produce the liver vessel centerline heatmap. An over-reconstructed liver vessel graph model is then traced based on the heatmap using an image processing based algorithm. We use a graph attention network to prune the false-positive branches by predicting the presence probability of each segmented branch in the initial reconstruction using the aggregated CNN features. We evaluated the proposed framework on an in-house dataset consisting of 418 multi-phase abdomen CT images with contrast. The proposed graph network pruning improves the overall reconstruction F1 score by 6.4% over the baseline. It also outperformed the other state-of-the-art curvilinear structure reconstruction algorithms.

Siqi Liu, Arnaud Arindra Adiyoso Setio, Florin C. Ghesu et al.

Detecting malignant pulmonary nodules at an early stage can allow medical interventions which may increase the survival rate of lung cancer patients. Using computer vision techniques to detect nodules can improve the sensitivity and the speed of interpreting chest CT for lung cancer screening. Many studies have used CNNs to detect nodule candidates. Though such approaches have been shown to outperform the conventional image processing based methods regarding the detection accuracy, CNNs are also known to be limited to generalize on under-represented samples in the training set and prone to imperceptible noise perturbations. Such limitations can not be easily addressed by scaling up the dataset or the models. In this work, we propose to add adversarial synthetic nodules and adversarial attack samples to the training data to improve the generalization and the robustness of the lung nodule detection systems. To generate hard examples of nodules from a differentiable nodule synthesizer, we use projected gradient descent (PGD) to search the latent code within a bounded neighbourhood that would generate nodules to decrease the detector response. To make the network more robust to unanticipated noise perturbations, we use PGD to search for noise patterns that can trigger the network to give over-confident mistakes. By evaluating on two different benchmark datasets containing consensus annotations from three radiologists, we show that the proposed techniques can improve the detection performance on real CT data. To understand the limitations of both the conventional networks and the proposed augmented networks, we also perform stress-tests on the false positive reduction networks by feeding different types of artificially produced patches. We show that the augmented networks are more robust to both under-represented nodules as well as resistant to noise perturbations.

Yipeng Hu, Eli Gibson, Dean C. Barratt et al.

Classical pairwise image registration methods search for a spatial transformation that optimises a numerical measure that indicates how well a pair of moving and fixed images are aligned. Current learning-based registration methods have adopted the same paradigm and typically predict, for any new input image pair, dense correspondences in the form of a dense displacement field or parameters of a spatial transformation model. However, in many applications of registration, the spatial transformation itself is only required to propagate points or regions of interest (ROIs). In such cases, detailed pixel- or voxel-level correspondence within or outside of these ROIs often have little clinical value. In this paper, we propose an alternative paradigm in which the location of corresponding image-specific ROIs, defined in one image, within another image is learnt. This results in replacing image registration by a conditional segmentation algorithm, which can build on typical image segmentation networks and their widely-adopted training strategies. Using the registration of 3D MRI and ultrasound images of the prostate as an example to demonstrate this new approach, we report a median target registration error (TRE) of 2.1 mm between the ground-truth ROIs defined on intraoperative ultrasound images and those propagated from the preoperative MR images. Significantly lower (>34%) TREs were obtained using the proposed conditional segmentation compared with those obtained from a previously-proposed spatial-transformation-predicting registration network trained with the same multiple ROI labels for individual image pairs. We conclude this work by using a quantitative bias-variance analysis to provide one explanation of the observed improvement in registration accuracy.

Florin C. Ghesu, Bogdan Georgescu, Eli Gibson et al.

The interpretation of chest radiographs is an essential task for the detection of thoracic diseases and abnormalities. However, it is a challenging problem with high inter-rater variability and inherent ambiguity due to inconclusive evidence in the data, limited data quality or subjective definitions of disease appearance. Current deep learning solutions for chest radiograph abnormality classification are typically limited to providing probabilistic predictions, relying on the capacity of learning models to adapt to the high degree of label noise and become robust to the enumerated causal factors. In practice, however, this leads to overconfident systems with poor generalization on unseen data. To account for this, we propose an automatic system that learns not only the probabilistic estimate on the presence of an abnormality, but also an explicit uncertainty measure which captures the confidence of the system in the predicted output. We argue that explicitly learning the classification uncertainty as an orthogonal measure to the predicted output, is essential to account for the inherent variability characteristic of this data. Experiments were conducted on two datasets of chest radiographs of over 85,000 patients. Sample rejection based on the predicted uncertainty can significantly improve the ROC-AUC, e.g., by 8% to 0.91 with an expected rejection rate of under 25%. Eliminating training samples using uncertainty-driven bootstrapping, enables a significant increase in robustness and accuracy. In addition, we present a multi-reader study showing that the predictive uncertainty is indicative of reader errors.

Sebastian Guendel, Florin C. Ghesu, Sasa Grbic et al.

Chest X-ray (CXR) is the most common X-ray examination performed in daily clinical practice for the diagnosis of various heart and lung abnormalities. The large amount of data to be read and reported, with 100+ studies per day for a single radiologist, poses a challenge in maintaining consistently high interpretation accuracy. In this work, we propose a method for the classification of different abnormalities based on CXR scans of the human body. The system is based on a novel multi-task deep learning architecture that in addition to the abnormality classification, supports the segmentation of the lungs and heart and classification of regions where the abnormality is located. We demonstrate that by training these tasks concurrently, one can increase the classification performance of the model. Experiments were performed on an extensive collection of 297,541 chest X-ray images from 86,876 patients, leading to a state-of-the-art performance level of 0.883 AUC on average for 12 different abnormalities. We also conducted a detailed performance analysis and compared the accuracy of our system with 3 board-certified radiologists. In this context, we highlight the high level of label noise inherent to this problem. On a reduced subset containing only cases with high confidence reference labels based on the consensus of the 3 radiologists, our system reached an average AUC of 0.945.

Jie Yang, Siqi Liu, Sasa Grbic et al.

Though large-scale datasets are essential for training deep learning systems, it is expensive to scale up the collection of medical imaging datasets. Synthesizing the objects of interests, such as lung nodules, in medical images based on the distribution of annotated datasets can be helpful for improving the supervised learning tasks, especially when the datasets are limited by size and class balance. In this paper, we propose the class-aware adversarial synthesis framework to synthesize lung nodules in CT images. The framework is built with a coarse-to-fine patch in-painter (generator) and two class-aware discriminators. By conditioning on the random latent variables and the target nodule labels, the trained networks are able to generate diverse nodules given the same context. By evaluating on the public LIDC-IDRI dataset, we demonstrate an example application of the proposed framework for improving the accuracy of the lung nodule malignancy estimation as a binary classification problem, which is important in the lung screening scenario. We show that combining the real image patches and the synthetic lung nodules in the training set can improve the mean AUC classification score across different network architectures by 2%.

Siqi Liu, Eli Gibson, Sasa Grbic et al.

The performance of medical image analysis systems is constrained by the quantity of high-quality image annotations. Such systems require data to be annotated by experts with years of training, especially when diagnostic decisions are involved. Such datasets are thus hard to scale up. In this context, it is hard for supervised learning systems to generalize to the cases that are rare in the training set but would be present in real-world clinical practices. We believe that the synthetic image samples generated by a system trained on the real data can be useful for improving the supervised learning tasks in the medical image analysis applications. Allowing the image synthesis to be manipulable could help synthetic images provide complementary information to the training data rather than simply duplicating the real-data manifold. In this paper, we propose a framework for synthesizing 3D objects, such as pulmonary nodules, in 3D medical images with manipulable properties. The manipulation is enabled by decomposing of the object of interests into its segmentation mask and a 1D vector containing the residual information. The synthetic object is refined and blended into the image context with two adversarial discriminators. We evaluate the proposed framework on lung nodules in 3D chest CT images and show that the proposed framework could generate realistic nodules with manipulable shapes, textures and locations, etc. By sampling from both the synthetic nodules and the real nodules from 2800 3D CT volumes during the classifier training, we show the synthetic patches could improve the overall nodule detection performance by average 8.44% competition performance metric (CPM) score.

Yipeng Hu, Marc Modat, Eli Gibson et al.

One of the fundamental challenges in supervised learning for multimodal image registration is the lack of ground-truth for voxel-level spatial correspondence. This work describes a method to infer voxel-level transformation from higher-level correspondence information contained in anatomical labels. We argue that such labels are more reliable and practical to obtain for reference sets of image pairs than voxel-level correspondence. Typical anatomical labels of interest may include solid organs, vessels, ducts, structure boundaries and other subject-specific ad hoc landmarks. The proposed end-to-end convolutional neural network approach aims to predict displacement fields to align multiple labelled corresponding structures for individual image pairs during the training, while only unlabelled image pairs are used as the network input for inference. We highlight the versatility of the proposed strategy, for training, utilising diverse types of anatomical labels, which need not to be identifiable over all training image pairs. At inference, the resulting 3D deformable image registration algorithm runs in real-time and is fully-automated without requiring any anatomical labels or initialisation. Several network architecture variants are compared for registering T2-weighted magnetic resonance images and 3D transrectal ultrasound images from prostate cancer patients. A median target registration error of 3.6 mm on landmark centroids and a median Dice of 0.87 on prostate glands are achieved from cross-validation experiments, in which 108 pairs of multimodal images from 76 patients were tested with high-quality anatomical labels.

Yipeng Hu, Eli Gibson, Nooshin Ghavami et al.

We describe an adversarial learning approach to constrain convolutional neural network training for image registration, replacing heuristic smoothness measures of displacement fields often used in these tasks. Using minimally-invasive prostate cancer intervention as an example application, we demonstrate the feasibility of utilizing biomechanical simulations to regularize a weakly-supervised anatomical-label-driven registration network for aligning pre-procedural magnetic resonance (MR) and 3D intra-procedural transrectal ultrasound (TRUS) images. A discriminator network is optimized to distinguish the registration-predicted displacement fields from the motion data simulated by finite element analysis. During training, the registration network simultaneously aims to maximize similarity between anatomical labels that drives image alignment and to minimize an adversarial generator loss that measures divergence between the predicted- and simulated deformation. The end-to-end trained network enables efficient and fully-automated registration that only requires an MR and TRUS image pair as input, without anatomical labels or simulated data during inference. 108 pairs of labelled MR and TRUS images from 76 prostate cancer patients and 71,500 nonlinear finite-element simulations from 143 different patients were used for this study. We show that, with only gland segmentation as training labels, the proposed method can help predict physically plausible deformation without any other smoothness penalty. Based on cross-validation experiments using 834 pairs of independent validation landmarks, the proposed adversarial-regularized registration achieved a target registration error of 6.3 mm that is significantly lower than those from several other regularization methods.

Yipeng Hu, Marc Modat, Eli Gibson et al.

Spatially aligning medical images from different modalities remains a challenging task, especially for intraoperative applications that require fast and robust algorithms. We propose a weakly-supervised, label-driven formulation for learning 3D voxel correspondence from higher-level label correspondence, thereby bypassing classical intensity-based image similarity measures. During training, a convolutional neural network is optimised by outputting a dense displacement field (DDF) that warps a set of available anatomical labels from the moving image to match their corresponding counterparts in the fixed image. These label pairs, including solid organs, ducts, vessels, point landmarks and other ad hoc structures, are only required at training time and can be spatially aligned by minimising a cross-entropy function of the warped moving label and the fixed label. During inference, the trained network takes a new image pair to predict an optimal DDF, resulting in a fully-automatic, label-free, real-time and deformable registration. For interventional applications where large global transformation prevails, we also propose a neural network architecture to jointly optimise the global- and local displacements. Experiment results are presented based on cross-validating registrations of 111 pairs of T2-weighted magnetic resonance images and 3D transrectal ultrasound images from prostate cancer patients with a total of over 4000 anatomical labels, yielding a median target registration error of 4.2 mm on landmark centroids and a median Dice of 0.88 on prostate glands.

Yipeng Hu, Eli Gibson, Tom Vercauteren et al.

In this paper, we describe how a patient-specific, ultrasound-probe-induced prostate motion model can be directly generated from a single preoperative MR image. Our motion model allows for sampling from the conditional distribution of dense displacement fields, is encoded by a generative neural network conditioned on a medical image, and accepts random noise as additional input. The generative network is trained by a minimax optimisation with a second discriminative neural network, tasked to distinguish generated samples from training motion data. In this work, we propose that 1) jointly optimising a third conditioning neural network that pre-processes the input image, can effectively extract patient-specific features for conditioning; and 2) combining multiple generative models trained separately with heuristically pre-disjointed training data sets can adequately mitigate the problem of mode collapse. Trained with diagnostic T2-weighted MR images from 143 real patients and 73,216 3D dense displacement fields from finite element simulations of intraoperative prostate motion due to transrectal ultrasound probe pressure, the proposed models produced physically-plausible patient-specific motion of prostate glands. The ability to capture biomechanically simulated motion was evaluated using two errors representing generalisability and specificity of the model. The median values, calculated from a 10-fold cross-validation, were 2.8+/-0.3 mm and 1.7+/-0.1 mm, respectively. We conclude that the introduced approach demonstrates the feasibility of applying state-of-the-art machine learning algorithms to generate organ motion models from patient images, and shows significant promise for future research.

Yipeng Hu, Eli Gibson, Li-Lin Lee et al.

Sonography synthesis has a wide range of applications, including medical procedure simulation, clinical training and multimodality image registration. In this paper, we propose a machine learning approach to simulate ultrasound images at given 3D spatial locations (relative to the patient anatomy), based on conditional generative adversarial networks (GANs). In particular, we introduce a novel neural network architecture that can sample anatomically accurate images conditionally on spatial position of the (real or mock) freehand ultrasound probe. To ensure an effective and efficient spatial information assimilation, the proposed spatially-conditioned GANs take calibrated pixel coordinates in global physical space as conditioning input, and utilise residual network units and shortcuts of conditioning data in the GANs' discriminator and generator, respectively. Using optically tracked B-mode ultrasound images, acquired by an experienced sonographer on a fetus phantom, we demonstrate the feasibility of the proposed method by two sets of quantitative results: distances were calculated between corresponding anatomical landmarks identified in the held-out ultrasound images and the simulated data at the same locations unseen to the networks; a usability study was carried out to distinguish the simulated data from the real images. In summary, we present what we believe are state-of-the-art visually realistic ultrasound images, simulated by the proposed GAN architecture that is stable to train and capable of generating plausibly diverse image samples.