J. Alison Noble

Matthias Eisenmann, Annika Reinke, Vivienn Weru et al. · utoronto

The number of international benchmarking competitions is steadily increasing in various fields of machine learning (ML) research and practice. So far, however, little is known about the common practice as well as bottlenecks faced by the community in tackling the research questions posed. To shed light on the status quo of algorithm development in the specific field of biomedical imaging analysis, we designed an international survey that was issued to all participants of challenges conducted in conjunction with the IEEE ISBI 2021 and MICCAI 2021 conferences (80 competitions in total). The survey covered participants' expertise and working environments, their chosen strategies, as well as algorithm characteristics. A median of 72% challenge participants took part in the survey. According to our results, knowledge exchange was the primary incentive (70%) for participation, while the reception of prize money played only a minor role (16%). While a median of 80 working hours was spent on method development, a large portion of participants stated that they did not have enough time for method development (32%). 25% perceived the infrastructure to be a bottleneck. Overall, 94% of all solutions were deep learning-based. Of these, 84% were based on standard architectures. 43% of the respondents reported that the data samples (e.g., images) were too large to be processed at once. This was most commonly addressed by patch-based training (69%), downsampling (37%), and solving 3D analysis tasks as a series of 2D tasks. K-fold cross-validation on the training set was performed by only 37% of the participants and only 50% of the participants performed ensembling based on multiple identical models (61%) or heterogeneous models (39%). 48% of the respondents applied postprocessing steps.

Matthias Eisenmann, Annika Reinke, Vivienn Weru et al.

International benchmarking competitions have become fundamental for the comparative performance assessment of image analysis methods. However, little attention has been given to investigating what can be learnt from these competitions. Do they really generate scientific progress? What are common and successful participation strategies? What makes a solution superior to a competing method? To address this gap in the literature, we performed a multi-center study with all 80 competitions that were conducted in the scope of IEEE ISBI 2021 and MICCAI 2021. Statistical analyses performed based on comprehensive descriptions of the submitted algorithms linked to their rank as well as the underlying participation strategies revealed common characteristics of winning solutions. These typically include the use of multi-task learning (63%) and/or multi-stage pipelines (61%), and a focus on augmentation (100%), image preprocessing (97%), data curation (79%), and postprocessing (66%). The "typical" lead of a winning team is a computer scientist with a doctoral degree, five years of experience in biomedical image analysis, and four years of experience in deep learning. Two core general development strategies stood out for highly-ranked teams: the reflection of the metrics in the method design and the focus on analyzing and handling failure cases. According to the organizers, 43% of the winning algorithms exceeded the state of the art but only 11% completely solved the respective domain problem. The insights of our study could help researchers (1) improve algorithm development strategies when approaching new problems, and (2) focus on open research questions revealed by this work.

Zeyu Fu, Jianbo Jiao, Robail Yasrab et al.

Self-supervised contrastive representation learning offers the advantage of learning meaningful visual representations from unlabeled medical datasets for transfer learning. However, applying current contrastive learning approaches to medical data without considering its domain-specific anatomical characteristics may lead to visual representations that are inconsistent in appearance and semantics. In this paper, we propose to improve visual representations of medical images via anatomy-aware contrastive learning (AWCL), which incorporates anatomy information to augment the positive/negative pair sampling in a contrastive learning manner. The proposed approach is demonstrated for automated fetal ultrasound imaging tasks, enabling the positive pairs from the same or different ultrasound scans that are anatomically similar to be pulled together and thus improving the representation learning. We empirically investigate the effect of inclusion of anatomy information with coarse- and fine-grained granularity, for contrastive learning and find that learning with fine-grained anatomy information which preserves intra-class difference is more effective than its counterpart. We also analyze the impact of anatomy ratio on our AWCL framework and find that using more distinct but anatomically similar samples to compose positive pairs results in better quality representations. Experiments on a large-scale fetal ultrasound dataset demonstrate that our approach is effective for learning representations that transfer well to three clinical downstream tasks, and achieves superior performance compared to ImageNet supervised and the current state-of-the-art contrastive learning methods. In particular, AWCL outperforms ImageNet supervised method by 13.8% and state-of-the-art contrastive-based method by 7.1% on a cross-domain segmentation task.

Shaheer U. Saeed, Yunguan Fu, Vasilis Stavrinides et al.

In this paper, we consider image quality assessment (IQA) as a measure of how images are amenable with respect to a given downstream task, or task amenability. When the task is performed using machine learning algorithms, such as a neural-network-based task predictor for image classification or segmentation, the performance of the task predictor provides an objective estimate of task amenability. In this work, we use an IQA controller to predict the task amenability which, itself being parameterised by neural networks, can be trained simultaneously with the task predictor. We further develop a meta-reinforcement learning framework to improve the adaptability for both IQA controllers and task predictors, such that they can be fine-tuned efficiently on new datasets or meta-tasks. We demonstrate the efficacy of the proposed task-specific, adaptable IQA approach, using two clinical applications for ultrasound-guided prostate intervention and pneumonia detection on X-ray images.

Qianhui Men, Clare Teng, Lior Drukker et al.

Eye trackers can provide visual guidance to sonographers during ultrasound (US) scanning. Such guidance is potentially valuable for less experienced operators to improve their scanning skills on how to manipulate the probe to achieve the desired plane. In this paper, a multimodal guidance approach (Multimodal-GuideNet) is proposed to capture the stepwise dependency between a real-world US video signal, synchronized gaze, and probe motion within a unified framework. To understand the causal relationship between gaze movement and probe motion, our model exploits multitask learning to jointly learn two related tasks: predicting gaze movements and probe signals that an experienced sonographer would perform in routine obstetric scanning. The two tasks are associated by a modality-aware spatial graph to detect the co-occurrence among the multi-modality inputs and share useful cross-modal information. Instead of a deterministic scanning path, Multimodal-GuideNet allows for scanning diversity by estimating the probability distribution of real scans. Experiments performed with three typical obstetric scanning examinations show that the new approach outperforms single-task learning for both probe motion guidance and gaze movement prediction. Multimodal-GuideNet also provides a visual guidance signal with an error rate of less than 10 pixels for a 224x288 US image.

Yiwen Li, Yunguan Fu, Iani Gayo et al.

The prowess that makes few-shot learning desirable in medical image analysis is the efficient use of the support image data, which are labelled to classify or segment new classes, a task that otherwise requires substantially more training images and expert annotations. This work describes a fully 3D prototypical few-shot segmentation algorithm, such that the trained networks can be effectively adapted to clinically interesting structures that are absent in training, using only a few labelled images from a different institute. First, to compensate for the widely recognised spatial variability between institutions in episodic adaptation of novel classes, a novel spatial registration mechanism is integrated into prototypical learning, consisting of a segmentation head and an spatial alignment module. Second, to assist the training with observed imperfect alignment, support mask conditioning module is proposed to further utilise the annotation available from the support images. Extensive experiments are presented in an application of segmenting eight anatomical structures important for interventional planning, using a data set of 589 pelvic T2-weighted MR images, acquired at seven institutes. The results demonstrate the efficacy in each of the 3D formulation, the spatial registration, and the support mask conditioning, all of which made positive contributions independently or collectively. Compared with the previously proposed 2D alternatives, the few-shot segmentation performance was improved with statistical significance, regardless whether the support data come from the same or different institutes.

Divyanshu Mishra, He Zhao, Pramit Saha et al.

Out-of-distribution (OOD) detection is essential to improve the reliability of machine learning models by detecting samples that do not belong to the training distribution. Detecting OOD samples effectively in certain tasks can pose a challenge because of the substantial heterogeneity within the in-distribution (ID), and the high structural similarity between ID and OOD classes. For instance, when detecting heart views in fetal ultrasound videos there is a high structural similarity between the heart and other anatomies such as the abdomen, and large in-distribution variance as a heart has 5 distinct views and structural variations within each view. To detect OOD samples in this context, the resulting model should generalise to the intra-anatomy variations while rejecting similar OOD samples. In this paper, we introduce dual-conditioned diffusion models (DCDM) where we condition the model on in-distribution class information and latent features of the input image for reconstruction-based OOD detection. This constrains the generative manifold of the model to generate images structurally and semantically similar to those within the in-distribution. The proposed model outperforms reference methods with a 12% improvement in accuracy, 22% higher precision, and an 8% better F1 score.

Jianbo Jiao, Mohammad Alsharid, Lior Drukker et al.

Auditory and visual signals usually present together and correlate with each other, not only in natural environments but also in clinical settings. However, the audio-visual modelling in the latter case can be more challenging, due to the different sources of audio/video signals and the noise (both signal-level and semantic-level) in auditory signals -- usually speech. In this paper, we consider audio-visual modelling in a clinical setting, providing a solution to learn medical representations that benefit various clinical tasks, without human expert annotation. A simple yet effective multi-modal self-supervised learning framework is proposed for this purpose. The proposed approach is able to localise anatomical regions of interest during ultrasound imaging, with only speech audio as a reference. Experimental evaluations on a large-scale clinical multi-modal ultrasound video dataset show that the proposed self-supervised method learns good transferable anatomical representations that boost the performance of automated downstream clinical tasks, even outperforming fully-supervised solutions.

Xiaoqing Guo, Qianhui Men, J. Alison Noble

We present the first automated multimodal summary generation system, MMSummary, for medical imaging video, particularly with a focus on fetal ultrasound analysis. Imitating the examination process performed by a human sonographer, MMSummary is designed as a three-stage pipeline, progressing from keyframe detection to keyframe captioning and finally anatomy segmentation and measurement. In the keyframe detection stage, an innovative automated workflow is proposed to progressively select a concise set of keyframes, preserving sufficient video information without redundancy. Subsequently, we adapt a large language model to generate meaningful captions for fetal ultrasound keyframes in the keyframe captioning stage. If a keyframe is captioned as fetal biometry, the segmentation and measurement stage estimates biometric parameters by segmenting the region of interest according to the textual prior. The MMSummary system provides comprehensive summaries for fetal ultrasound examinations and based on reported experiments is estimated to reduce scanning time by approximately 31.5%, thereby suggesting the potential to enhance clinical workflow efficiency.

Qianhui Men, Xiaoqing Guo, Aris T. Papageorghiou et al.

3D pose estimation from a 2D cross-sectional view enables healthcare professionals to navigate through the 3D space, and such techniques initiate automatic guidance in many image-guided radiology applications. In this work, we investigate how estimating 3D fetal pose from freehand 2D ultrasound scanning can guide a sonographer to locate a head standard plane. Fetal head pose is estimated by the proposed Pose-GuideNet, a novel 2D/3D registration approach to align freehand 2D ultrasound to a 3D anatomical atlas without the acquisition of 3D ultrasound. To facilitate the 2D to 3D cross-dimensional projection, we exploit the prior knowledge in the atlas to align the standard plane frame in a freehand scan. A semantic-aware contrastive-based approach is further proposed to align the frames that are off standard planes based on their anatomical similarity. In the experiment, we enhance the existing assessment of freehand image localization by comparing the transformation of its estimated pose towards standard plane with the corresponding probe motion, which reflects the actual view change in 3D anatomy. Extensive results on two clinical head biometry tasks show that Pose-GuideNet not only accurately predicts pose but also successfully predicts the direction of the fetal head. Evaluations with probe motions further demonstrate the feasibility of adopting Pose-GuideNet for freehand ultrasound-assisted navigation in a sensor-free environment.

Pramit Saha, Divyanshu Mishra, J. Alison Noble

The most challenging, yet practical, setting of semi-supervised federated learning (SSFL) is where a few clients have fully labeled data whereas the other clients have fully unlabeled data. This is particularly common in healthcare settings where collaborating partners (typically hospitals) may have images but not annotations. The bottleneck in this setting is the joint training of labeled and unlabeled clients as the objective function for each client varies based on the availability of labels. This paper investigates an alternative way for effective training with labeled and unlabeled clients in a federated setting. We propose a novel learning scheme specifically designed for SSFL which we call Isolated Federated Learning (IsoFed) that circumvents the problem by avoiding simple averaging of supervised and semi-supervised models together. In particular, our training approach consists of two parts - (a) isolated aggregation of labeled and unlabeled client models, and (b) local self-supervised pretraining of isolated global models in all clients. We evaluate our model performance on medical image datasets of four different modalities publicly available within the biomedical image classification benchmark MedMNIST. We further vary the proportion of labeled clients and the degree of heterogeneity to demonstrate the effectiveness of the proposed method under varied experimental settings.

Angus Nicolson, Yarin Gal, J. Alison Noble

Concept-based interpretability methods are a popular form of explanation for deep learning models which provide explanations in the form of high-level human interpretable concepts. These methods typically find concept activation vectors (CAVs) using a probe dataset of concept examples. This requires labelled data for these concepts -- an expensive task in the medical domain. We introduce TextCAVs: a novel method which creates CAVs using vision-language models such as CLIP, allowing for explanations to be created solely using text descriptions of the concept, as opposed to image exemplars. This reduced cost in testing concepts allows for many concepts to be tested and for users to interact with the model, testing new ideas as they are thought of, rather than a delay caused by image collection and annotation. In early experimental results, we demonstrate that TextCAVs produces reasonable explanations for a chest x-ray dataset (MIMIC-CXR) and natural images (ImageNet), and that these explanations can be used to debug deep learning-based models.

Pramit Saha, Joshua Strong, Divyanshu Mishra et al.

Federated learning (FL) allows collaborative model training across healthcare sites without sharing sensitive patient data. However, real-world FL deployment is often hindered by complex operational challenges that demand substantial human efforts. This includes: (a) selecting appropriate clients (hospitals), (b) coordinating between the central server and clients, (c) client-level data pre-processing, (d) harmonizing non-standardized data and labels across clients, and (e) selecting FL algorithms based on user instructions and cross-client data characteristics. However, the existing FL works overlook these practical orchestration challenges. These operational bottlenecks motivate the need for autonomous, agent-driven FL systems, where intelligent agents at each hospital client and the central server agent collaboratively manage FL setup and model training with minimal human intervention. To this end, we first introduce an agent-driven FL framework that captures key phases of real-world FL workflows from client selection to training completion and a benchmark dubbed FedAgentBench that evaluates the ability of LLM agents to autonomously coordinate healthcare FL. Our framework incorporates 40 FL algorithms, each tailored to address diverse task-specific requirements and cross-client characteristics. Furthermore, we introduce a diverse set of complex tasks across 201 carefully curated datasets, simulating 6 modality-specific real-world healthcare environments, viz., Dermatoscopy, Ultrasound, Fundus, Histopathology, MRI, and X-Ray. We assess the agentic performance of 14 open-source and 10 proprietary LLMs spanning small, medium, and large model scales. While some agent cores such as GPT-4.1 and DeepSeek V3 can automate various stages of the FL pipeline, our results reveal that more complex, interdependent tasks based on implicit goals remain challenging for even the strongest models.

Yingyu Yang, Qianye Yang, Kangning Cui et al.

The identification of cardiac phase is an essential step for analysis and diagnosis of cardiac function. Automatic methods, especially data-driven methods for cardiac phase detection, typically require extensive annotations, which is time-consuming and labor-intensive. In this paper, we present an unsupervised framework for end-diastole (ED) and end-systole (ES) detection through self-supervised learning of latent cardiac motion trajectories from 4-chamber-view echocardiography videos. Our method eliminates the need for manual annotations, including ED and ES indices, segmentation, or volumetric measurements, by training a reconstruction model to encode interpretable spatiotemporal motion patterns. Evaluated on the EchoNet-Dynamic benchmark, the approach achieves mean absolute error (MAE) of 3 frames (58.3 ms) for ED and 2 frames (38.8 ms) for ES detection, matching state-of-the-art supervised methods. Extended to fetal echocardiography, the model demonstrates robust performance with MAE 1.46 frames (20.7 ms) for ED and 1.74 frames (25.3 ms) for ES, despite the fact that the fetal heart model is built using non-standardized heart views due to fetal heart positioning variability. Our results demonstrate the potential of the proposed latent motion trajectory strategy for cardiac phase detection in adult and fetal echocardiography. This work advances unsupervised cardiac motion analysis, offering a scalable solution for clinical populations lacking annotated data. Code will be released at https://github.com/YingyuYyy/CardiacPhase.

Divyanshu Mishra, Mohammadreza Salehi, Pramit Saha et al.

Self-supervised learning (SSL) has achieved major advances in natural images and video understanding, but challenges remain in domains like echocardiography (heart ultrasound) due to subtle anatomical structures, complex temporal dynamics, and the current lack of domain-specific pre-trained models. Existing SSL approaches such as contrastive, masked modeling, and clustering-based methods struggle with high intersample similarity, sensitivity to low PSNR inputs common in ultrasound, or aggressive augmentations that distort clinically relevant features. We present DISCOVR (Distilled Image Supervision for Cross Modal Video Representation), a self-supervised dual branch framework for cardiac ultrasound video representation learning. DISCOVR combines a clustering-based video encoder that models temporal dynamics with an online image encoder that extracts fine-grained spatial semantics. These branches are connected through a semantic cluster distillation loss that transfers anatomical knowledge from the evolving image encoder to the video encoder, enabling temporally coherent representations enriched with fine-grained semantic understanding.Evaluated on six echocardiography datasets spanning fetal, pediatric, and adult populations, DISCOVR outperforms both specialized video anomaly detection methods and state-of-the-art video-SSL baselines in zero-shot and linear probing setups,achieving superior segmentation transfer and strong downstream performance on clinically relevant tasks such as LVEF prediction. Code available at: https://github.com/mdivyanshu97/DISCOVR

Felix Wagner, Pramit Saha, Harry Anthony et al.

Safe deployment of machine learning (ML) models in safety-critical domains such as medical imaging requires detecting inputs with characteristics not seen during training, known as out-of-distribution (OOD) detection, to prevent unreliable predictions. Effective OOD detection after deployment could benefit from access to the training data, enabling direct comparison between test samples and the training data distribution to identify differences. State-of-the-art OOD detection methods, however, either discard the training data after deployment or assume that test samples and training data are centrally stored together, an assumption that rarely holds in real-world settings. This is because shipping the training data with the deployed model is usually impossible due to the size of training databases, as well as proprietary or privacy constraints. We introduce the Isolation Network, an OOD detection framework that quantifies the difficulty of separating a target test sample from the training data by solving a binary classification task. We then propose Decentralized Isolation Networks (DIsoN), which enables the comparison of training and test data when data-sharing is impossible, by exchanging only model parameters between the remote computational nodes of training and deployment. We further extend DIsoN with class-conditioning, comparing a target sample solely with training data of its predicted class. We evaluate DIsoN on four medical imaging datasets (dermatology, chest X-ray, breast ultrasound, histopathology) across 12 OOD detection tasks. DIsoN performs favorably against existing methods while respecting data-privacy. This decentralized OOD detection framework opens the way for a new type of service that ML developers could provide along with their models: providing remote, secure utilization of their training data for OOD detection services. Code: https://github.com/FelixWag/DIsoN

Ziyun Liang, Xiaoqing Guo, Wentian Xu et al.

Unsupervised anomaly detection and segmentation methods train a model to learn the training distribution as `normal'. In the testing phase, they identify patterns that deviate from this normal distribution as `anomalies'. To learn the `normal' distribution, prevailing methods corrupt the images and train a model to reconstruct them. During testing, the model attempts to reconstruct corrupted inputs based on the learned `normal' distribution. Deviations from this distribution lead to high reconstruction errors, which indicate potential anomalies. However, corrupting an input image inevitably causes information loss even in normal regions, leading to suboptimal reconstruction and an increased risk of false positives. To alleviate this, we propose $\rm{IterMask3D}$, an iterative spatial mask-refining strategy designed for 3D brain MRI. We iteratively spatially mask areas of the image as corruption and reconstruct them, then shrink the mask based on reconstruction error. This process iteratively unmasks `normal' areas to the model, whose information further guides reconstruction of `normal' patterns under the mask to be reconstructed accurately, reducing false positives. In addition, to achieve better reconstruction performance, we also propose using high-frequency image content as additional structural information to guide the reconstruction of the masked area. Extensive experiments on the detection of both synthetic and real-world imaging artifacts, as well as segmentation of various pathological lesions across multiple MRI sequences, consistently demonstrate the effectiveness of our proposed method. Code is available at https://github.com/ZiyunLiang/IterMask3D.

Felix Wagner, Wentian Xu, Pramit Saha et al.

Segmentation models for brain lesions in MRI are typically developed for a specific disease and trained on data with a predefined set of MRI modalities. Such models cannot segment the disease using data with a different set of MRI modalities, nor can they segment other types of diseases. Moreover, this training paradigm prevents a model from using the advantages of learning from heterogeneous databases that may contain scans and segmentation labels for different brain pathologies and diverse sets of MRI modalities. Additionally, the confidentiality of patient data often prevents central data aggregation, necessitating a decentralized approach. Is it feasible to use Federated Learning (FL) to train a single model on client databases that contain scans and labels of different brain pathologies and diverse sets of MRI modalities? We demonstrate promising results by combining appropriate, simple, and practical modifications to the model and training strategy: Designing a model with input channels that cover the whole set of modalities available across clients, training with random modality drop, and exploring the effects of feature normalization methods. Evaluation on 7 brain MRI databases with 5 different diseases shows that this FL framework can train a single model achieving very promising results in segmenting all disease types seen during training. Importantly, it can segment these diseases in new databases that contain sets of modalities different from those in training clients. These results demonstrate, for the first time, the feasibility and effectiveness of using FL to train a single 3D segmentation model on decentralised data with diverse brain diseases and MRI modalities, a necessary step towards leveraging heterogeneous real-world databases. Code: https://github.com/FelixWag/FedUniBrain

Joshua Strong, Qianhui Men, Alison Noble

Large language models (LLMs) offer a valuable technology for various applications in healthcare. However, their tendency to hallucinate and the existing reliance on proprietary systems pose challenges in environments concerning critical decision-making and strict data privacy regulations, such as healthcare, where the trust in such systems is paramount. Through combining the strengths and discounting the weaknesses of humans and AI, the field of Human-AI Collaboration (HAIC) presents one front for tackling these challenges and hence improving trust. This paper presents a novel HAIC guided deferral system that can simultaneously parse medical reports for disorder classification, and defer uncertain predictions with intelligent guidance to humans. We develop methodology which builds efficient, effective and open-source LLMs for this purpose, for the real-world deployment in healthcare. We conduct a pilot study which showcases the effectiveness of our proposed system in practice. Additionally, we highlight drawbacks of standard calibration metrics in imbalanced data scenarios commonly found in healthcare, and suggest a simple yet effective solution: the Imbalanced Expected Calibration Error.

Ziyun Liang, Xiaoqing Guo, J. Alison Noble et al.

Unsupervised anomaly segmentation approaches to pathology segmentation train a model on images of healthy subjects, that they define as the 'normal' data distribution. At inference, they aim to segment any pathologies in new images as 'anomalies', as they exhibit patterns that deviate from those in 'normal' training data. Prevailing methods follow the 'corrupt-and-reconstruct' paradigm. They intentionally corrupt an input image, reconstruct it to follow the learned 'normal' distribution, and subsequently segment anomalies based on reconstruction error. Corrupting an input image, however, inevitably leads to suboptimal reconstruction even of normal regions, causing false positives. To alleviate this, we propose a novel iterative spatial mask-refining strategy IterMask2. We iteratively mask areas of the image, reconstruct them, and update the mask based on reconstruction error. This iterative process progressively adds information about areas that are confidently normal as per the model. The increasing content guides reconstruction of nearby masked areas, improving reconstruction of normal tissue under these areas, reducing false positives. We also use high-frequency image content as an auxiliary input to provide additional structural information for masked areas. This further improves reconstruction error of normal in comparison to anomalous areas, facilitating segmentation of the latter. We conduct experiments on several brain lesion datasets and demonstrate effectiveness of our method. Code is available at: https://github.com/ZiyunLiang/IterMask2

Pramit Saha, Divyanshu Mishra, Felix Wagner et al.

Multimodal Federated Learning (MMFL) utilizes multiple modalities in each client to build a more powerful Federated Learning (FL) model than its unimodal counterpart. However, the impact of missing modality in different clients, also called modality incongruity, has been greatly overlooked. This paper, for the first time, analyses the impact of modality incongruity and reveals its connection with data heterogeneity across participating clients. We particularly inspect whether incongruent MMFL with unimodal and multimodal clients is more beneficial than unimodal FL. Furthermore, we examine three potential routes of addressing this issue. Firstly, we study the effectiveness of various self-attention mechanisms towards incongruity-agnostic information fusion in MMFL. Secondly, we introduce a modality imputation network (MIN) pre-trained in a multimodal client for modality translation in unimodal clients and investigate its potential towards mitigating the missing modality problem. Thirdly, we assess the capability of client-level and server-level regularization techniques towards mitigating modality incongruity effects. Experiments are conducted under several MMFL settings on two publicly available real-world datasets, MIMIC-CXR and Open-I, with Chest X-Ray and radiology reports.

Pramit Saha, Divyanshu Mishra, Felix Wagner et al.

Large Vision-Language Models typically require large text and image datasets for effective fine-tuning. However, collecting data from various sites, especially in healthcare, is challenging due to strict privacy regulations. An alternative is to fine-tune these models on end-user devices, such as in medical clinics, without sending data to a server. These local clients typically have limited computing power and small datasets, which are not enough for fully fine-tuning large VLMs on their own. A naive solution to these scenarios is to leverage parameter-efficient fine-tuning (PEFT) strategies and apply federated learning (FL) algorithms to combine the learned adapter weights, thereby respecting the resource limitations and data privacy. However, this approach does not fully leverage the knowledge from multiple adapters trained on diverse data distributions and for diverse tasks. The adapters are adversely impacted by data heterogeneity and task heterogeneity across clients resulting in suboptimal convergence. To this end, we propose a novel framework called FedPIA that improves upon the naive combinations of FL and PEFT by introducing Permutation and Integration of the local Adapters in the server and global Adapters in the clients exploiting Wasserstein barycenters for improved blending of client-specific and client-agnostic knowledge. This layerwise permutation helps to bridge the gap in the parameter space of local and global adapters before integration. We conduct over 2000 client-level experiments utilizing 48 medical image datasets across five different medical vision-language FL task settings encompassing visual question answering as well as image and report-based multi-label disease detection. Our experiments involving diverse client settings, ten different modalities, and two VLM backbones demonstrate that FedPIA consistently outperforms the state-of-the-art PEFT-FL baselines.

Angus Nicolson, Lisa Schut, J. Alison Noble et al.

Concept-based explanations translate the internal representations of deep learning models into a language that humans are familiar with: concepts. One popular method for finding concepts is Concept Activation Vectors (CAVs), which are learnt using a probe dataset of concept exemplars. In this work, we investigate three properties of CAVs: (1) inconsistency across layers, (2) entanglement with other concepts, and (3) spatial dependency. Each property provides both challenges and opportunities in interpreting models. We introduce tools designed to detect the presence of these properties, provide insight into how each property can lead to misleading explanations, and provide recommendations to mitigate their impact. To demonstrate practical applications, we apply our recommendations to a melanoma classification task, showing how entanglement can lead to uninterpretable results and that the choice of negative probe set can have a substantial impact on the meaning of a CAV. Further, we show that understanding these properties can be used to our advantage. For example, we introduce spatially dependent CAVs to test if a model is translation invariant with respect to a specific concept and class. Our experiments are performed on natural images (ImageNet), skin lesions (ISIC 2019), and a new synthetic dataset, Elements. Elements is designed to capture a known ground truth relationship between concepts and classes. We release this dataset to facilitate further research in understanding and evaluating interpretability methods.

Pramit Saha, Divyanshu Mishra, Netzahualcoyotl Hernandez-Cruz et al.

Congenital Heart Disease (CHD) is one of the leading causes of fetal mortality, yet the scarcity of labeled CHD data and strict privacy regulations surrounding fetal ultrasound (US) imaging present significant challenges for the development of deep learning-based models for CHD detection. Centralised collection of large real-world datasets for rare conditions, such as CHD, from large populations requires significant co-ordination and resource. In addition, data governance rules increasingly prevent data sharing between sites. To address these challenges, we introduce, for the first time, a novel privacy-preserving, zero-shot CHD detection framework that formulates CHD detection as a normality modeling problem integrated with model merging. In our framework dubbed Sparse Tube Ultrasound Distillation (STUD), each hospital site first trains a sparse video tube-based self-supervised video anomaly detection (VAD) model on normal fetal heart US clips with self-distillation loss. This enables site-specific models to independently learn the distribution of healthy cases. To aggregate knowledge across the decentralized models while maintaining privacy, we propose a Divergence Vector-Guided Model Merging approach, DivMerge, that combines site-specific models into a single VAD model without data exchange. Our approach preserves domain-agnostic rich spatio-temporal representations, ensuring generalization to unseen CHD cases. We evaluated our approach on real-world fetal US data collected from 5 hospital sites. Our merged model outperformed site-specific models by 23.77% and 30.13% in accuracy and F1-score respectively on external test sets.

Yiwen Li, Yunguan Fu, Iani J. M. B. Gayo et al.

For training registration networks, weak supervision from segmented corresponding regions-of-interest (ROIs) have been proven effective for (a) supplementing unsupervised methods, and (b) being used independently in registration tasks in which unsupervised losses are unavailable or ineffective. This correspondence-informing supervision entails cost in annotation that requires significant specialised effort. This paper describes a semi-weakly-supervised registration pipeline that improves the model performance, when only a small corresponding-ROI-labelled dataset is available, by exploiting unlabelled image pairs. We examine two types of augmentation methods by perturbation on network weights and image resampling, such that consistency-based unsupervised losses can be applied on unlabelled data. The novel WarpDDF and RegCut approaches are proposed to allow commutative perturbation between an image pair and the predicted spatial transformation (i.e. respective input and output of registration networks), distinct from existing perturbation methods for classification or segmentation. Experiments using 589 male pelvic MR images, labelled with eight anatomical ROIs, show the improvement in registration performance and the ablated contributions from the individual strategies. Furthermore, this study attempts to construct one of the first computational atlases for pelvic structures, enabled by registering inter-subject MRs, and quantifies the significant differences due to the proposed semi-weak supervision with a discussion on the potential clinical use of example atlas-derived statistics.

Joshua Strong, Pramit Saha, Yasin Ibrahim et al.

Learning to Defer (L2D) trains autonomous systems to handle straightforward cases while deferring uncertain ones to human experts. Recent advancements in this field have introduced methods that offer flexibility to unseen experts at test time. However, we find these approaches struggle to generalise to experts with behaviours not seen during training, require extensive human annotation, and lack mechanisms for incorporating prior knowledge of expert capabilities. To address these challenges, we introduce Expert-Agnostic Learning to Defer (EA-L2D), a novel L2D framework that employs a Bayesian approach to model expert behaviour in an \textit{expert-agnostic} fashion. Across benchmark medical imaging datasets (HAM10000, Blood Cells, Retinal OCT, and Liver Tumours), EA-L2D significantly outperforms prior methods on unseen experts, achieving up to a 28\% relative improvement, while also matching or exceeding state-of-the-art performance on seen experts.

Jiayuan Zhu, Junde Wu, Cheng Ouyang et al.

Medical image segmentation data inherently contain uncertainty. This can stem from both imperfect image quality and variability in labeling preferences on ambiguous pixels, which depend on annotator expertise and the clinical context of the annotations. For instance, a boundary pixel might be labeled as tumor in diagnosis to avoid under-estimation of severity, but as normal tissue in radiotherapy to prevent damage to sensitive structures. As segmentation preferences vary across downstream applications, it is often desirable for an image segmentation model to offer user-adaptable predictions rather than a fixed output. While prior uncertainty-aware and interactive methods offer adaptability, they are inefficient at test time: uncertainty-aware models require users to choose from numerous similar outputs, while interactive models demand significant user input through click or box prompts to refine segmentation. To address these challenges, we propose \textbf{SPA}, a new \textbf{S}egmentation \textbf{P}reference \textbf{A}lignment framework that efficiently adapts to diverse test-time preferences with minimal human interaction. By presenting users with a select few, distinct segmentation candidates that best capture uncertainties, it reduces the user workload to reach the preferred segmentation. To accommodate user preference, we introduce a probabilistic mechanism that leverages user feedback to adapt a model's segmentation preference. The proposed framework is evaluated on several medical image segmentation tasks: color fundus images, lung lesion and kidney CT scans, MRI scans of brain and prostate. SPA shows 1) a significant reduction in user time and effort compared to existing interactive segmentation approaches, 2) strong adaptability based on human feedback, and 3) state-of-the-art image segmentation performance across different imaging modalities and semantic labels.

Pramit Saha, Felix Wagner, Divyanshu Mishra et al.

Effective training of large Vision-Language Models (VLMs) on resource-constrained client devices in Federated Learning (FL) requires the usage of parameter-efficient fine-tuning (PEFT) strategies. To this end, we demonstrate the impact of two factors \textit{viz.}, client-specific layer importance score that selects the most important VLM layers for fine-tuning and inter-client layer diversity score that encourages diverse layer selection across clients for optimal VLM layer selection. We first theoretically motivate and leverage the principal eigenvalue magnitude of layerwise Neural Tangent Kernels and show its effectiveness as client-specific layer importance score. Next, we propose a novel layer updating strategy dubbed F$^3$OCUS that jointly optimizes the layer importance and diversity factors by employing a data-free, multi-objective, meta-heuristic optimization on the server. We explore 5 different meta-heuristic algorithms and compare their effectiveness for selecting model layers and adapter layers towards PEFT-FL. Furthermore, we release a new MedVQA-FL dataset involving overall 707,962 VQA triplets and 9 modality-specific clients and utilize it to train and evaluate our method. Overall, we conduct more than 10,000 client-level experiments on 6 Vision-Language FL task settings involving 58 medical image datasets and 4 different VLM architectures of varying sizes to demonstrate the effectiveness of the proposed method.

Can Peng, Yuyuan Liu, Yingyu Yang et al.

Federated Learning (FL) enables collaborative model training across distributed clients while preserving data privacy. However, the performance of deep learning often deteriorates in FL due to decentralized and heterogeneous data. This challenge is further amplified in multi-label scenarios, where data exhibit complex characteristics such as label co-occurrence, inter-label dependency, and discrepancies between local and global label relationships. While most existing FL research primarily focuses on single-label classification, many real-world applications, particularly in domains such as medical imaging, often involve multi-label settings. In this paper, we address this important yet underexplored scenario in FL, where clients hold multi-label data with skewed label distributions. Neural Collapse (NC) describes a geometric structure in the latent feature space where features of each class collapse to their class mean with vanishing intra-class variance, and the class means form a maximally separated configuration. Motivated by this theory, we propose a method to align feature distributions across clients and to learn high-quality, well-clustered representations. To make the NC-structure applicable to multi-label settings, where image-level features may contain multiple semantic concepts, we introduce a feature disentanglement module that extracts semantically specific features. The clustering of these disentangled class-wise features is guided by a predefined shared NC structure, which mitigates potential conflicts between client models due to diverse local data distributions. In addition, we design regularisation losses to encourage compact clustering in the latent feature space. Experiments conducted on four benchmark datasets across eight diverse settings demonstrate that our approach outperforms existing methods, validating its effectiveness in this challenging FL scenario.

Divyanshu Mishra, Pramit Saha, He Zhao et al.

Accurate standard plane acquisition in fetal ultrasound (US) videos is crucial for fetal growth assessment, anomaly detection, and adherence to clinical guidelines. However, manually selecting standard frames is time-consuming and prone to intra- and inter-sonographer variability. Existing methods primarily rely on image-based approaches that capture standard frames and then classify the input frames across different anatomies. This ignores the dynamic nature of video acquisition and its interpretation. To address these challenges, we introduce Multi-Tier Class-Aware Token Transformer (MCAT), a visual query-based video clip localization (VQ-VCL) method, to assist sonographers by enabling them to capture a quick US sweep. By then providing a visual query of the anatomy they wish to analyze, MCAT returns the video clip containing the standard frames for that anatomy, facilitating thorough screening for potential anomalies. We evaluate MCAT on two ultrasound video datasets and a natural image VQ-VCL dataset based on Ego4D. Our model outperforms state-of-the-art methods by 10% and 13% mIoU on the ultrasound datasets and by 5.35% mIoU on the Ego4D dataset, using 96% fewer tokens. MCAT's efficiency and accuracy have significant potential implications for public health, especially in low- and middle-income countries (LMICs), where it may enhance prenatal care by streamlining standard plane acquisition, simplifying US-based screening, diagnosis and allowing sonographers to examine more patients.

Zeyu Fu, Jianbo Jiao, Michael Suttie et al.

Fetal alcohol syndrome (FAS) caused by prenatal alcohol exposure can result in a series of cranio-facial anomalies, and behavioral and neurocognitive problems. Current diagnosis of FAS is typically done by identifying a set of facial characteristics, which are often obtained by manual examination. Anatomical landmark detection, which provides rich geometric information, is important to detect the presence of FAS associated facial anomalies. This imaging application is characterized by large variations in data appearance and limited availability of labeled data. Current deep learning-based heatmap regression methods designed for facial landmark detection in natural images assume availability of large datasets and are therefore not wellsuited for this application. To address this restriction, we develop a new regularized transfer learning approach that exploits the knowledge of a network learned on large facial recognition datasets. In contrast to standard transfer learning which focuses on adjusting the pre-trained weights, the proposed learning approach regularizes the model behavior. It explicitly reuses the rich visual semantics of a domain-similar source model on the target task data as an additional supervisory signal for regularizing landmark detection optimization. Specifically, we develop four regularization constraints for the proposed transfer learning, including constraining the feature outputs from classification and intermediate layers, as well as matching activation attention maps in both spatial and channel levels. Experimental evaluation on a collected clinical imaging dataset demonstrate that the proposed approach can effectively improve model generalizability under limited training samples, and is advantageous to other approaches in the literature.

Yuan Gao, Lok Hin Lee, Richard Droste et al.

This paper presents a novel approach to automatic fetal brain biometry motivated by needs in low- and medium- income countries. Specifically, we leverage high-end (HE) ultrasound images to build a biometry solution for low-cost (LC) point-of-care ultrasound images. We propose a novel unsupervised domain adaptation approach to train deep models to be invariant to significant image distribution shift between the image types. Our proposed method, which employs a Dual Adversarial Calibration (DAC) framework, consists of adversarial pathways which enforce model invariance to; i) adversarial perturbations in the feature space derived from LC images, and ii) appearance domain discrepancy. Our Dual Adversarial Calibration method estimates transcerebellar diameter and head circumference on images from low-cost ultrasound devices with a mean absolute error (MAE) of 2.43mm and 1.65mm, compared with 7.28 mm and 5.65 mm respectively for SOTA.

Shaheer U. Saeed, Yunguan Fu, Vasilis Stavrinides et al.

The performance of many medical image analysis tasks are strongly associated with image data quality. When developing modern deep learning algorithms, rather than relying on subjective (human-based) image quality assessment (IQA), task amenability potentially provides an objective measure of task-specific image quality. To predict task amenability, an IQA agent is trained using reinforcement learning (RL) with a simultaneously optimised task predictor, such as a classification or segmentation neural network. In this work, we develop transfer learning or adaptation strategies to increase the adaptability of both the IQA agent and the task predictor so that they are less dependent on high-quality, expert-labelled training data. The proposed transfer learning strategy re-formulates the original RL problem for task amenability in a meta-reinforcement learning (meta-RL) framework. The resulting algorithm facilitates efficient adaptation of the agent to different definitions of image quality, each with its own Markov decision process environment including different images, labels and an adaptable task predictor. Our work demonstrates that the IQA agents pre-trained on non-expert task labels can be adapted to predict task amenability as defined by expert task labels, using only a small set of expert labels. Using 6644 clinical ultrasound images from 249 prostate cancer patients, our results for image classification and segmentation tasks show that the proposed IQA method can be adapted using data with as few as respective 19.7% and 29.6% expert-reviewed consensus labels and still achieve comparable IQA and task performance, which would otherwise require a training dataset with 100% expert labels.

Lok Hin Lee, Yuan Gao, J. Alison Noble

Deep learning models with large learning capacities often overfit to medical imaging datasets. This is because training sets are often relatively small due to the significant time and financial costs incurred in medical data acquisition and labelling. Data augmentation is therefore often used to expand the availability of training data and to increase generalization. However, augmentation strategies are often chosen on an ad-hoc basis without justification. In this paper, we present an augmentation policy search method with the goal of improving model classification performance. We include in the augmentation policy search additional transformations that are often used in medical image analysis and evaluate their performance. In addition, we extend the augmentation policy search to include non-linear mixed-example data augmentation strategies. Using these learned policies, we show that principled data augmentation for medical image model training can lead to significant improvements in ultrasound standard plane detection, with an an average F1-score improvement of 7.0% overall over naive data augmentation strategies in ultrasound fetal standard plane classification. We find that the learned representations of ultrasound images are better clustered and defined with optimized data augmentation.

Zeyu Fu, Jianbo Jiao, Michael Suttie et al.

Recently, there is an increasing demand for automatically detecting anatomical landmarks which provide rich structural information to facilitate subsequent medical image analysis. Current methods related to this task often leverage the power of deep neural networks, while a major challenge in fine tuning such models in medical applications arises from insufficient number of labeled samples. To address this, we propose to regularize the knowledge transfer across source and target tasks through cross-task representation learning. The proposed method is demonstrated for extracting facial anatomical landmarks which facilitate the diagnosis of fetal alcohol syndrome. The source and target tasks in this work are face recognition and landmark detection, respectively. The main idea of the proposed method is to retain the feature representations of the source model on the target task data, and to leverage them as an additional source of supervisory signals for regularizing the target model learning, thereby improving its performance under limited training samples. Concretely, we present two approaches for the proposed representation learning by constraining either final or intermediate model features on the target model. Experimental results on a clinical face image dataset demonstrate that the proposed approach works well with few labeled data, and outperforms other compared approaches.

Jianbo Jiao, Ana I. L. Namburete, Aris T. Papageorghiou et al.

Fetal brain magnetic resonance imaging (MRI) offers exquisite images of the developing brain but is not suitable for second-trimester anomaly screening, for which ultrasound (US) is employed. Although expert sonographers are adept at reading US images, MR images which closely resemble anatomical images are much easier for non-experts to interpret. Thus in this paper we propose to generate MR-like images directly from clinical US images. In medical image analysis such a capability is potentially useful as well, for instance for automatic US-MRI registration and fusion. The proposed model is end-to-end trainable and self-supervised without any external annotations. Specifically, based on an assumption that the US and MRI data share a similar anatomical latent space, we first utilise a network to extract the shared latent features, which are then used for MRI synthesis. Since paired data is unavailable for our study (and rare in practice), pixel-level constraints are infeasible to apply. We instead propose to enforce the distributions to be statistically indistinguishable, by adversarial learning in both the image domain and feature space. To regularise the anatomical structures between US and MRI during synthesis, we further propose an adversarial structural constraint. A new cross-modal attention technique is proposed to utilise non-local spatial information, by encouraging multi-modal knowledge fusion and propagation. We extend the approach to consider the case where 3D auxiliary information (e.g., 3D neighbours and a 3D location index) from volumetric data is also available, and show that this improves image synthesis. The proposed approach is evaluated quantitatively and qualitatively with comparison to real fetal MR images and other approaches to synthesis, demonstrating its feasibility of synthesising realistic MR images.

Jianbo Jiao, Yifan Cai, Mohammad Alsharid et al.

In medical imaging, manual annotations can be expensive to acquire and sometimes infeasible to access, making conventional deep learning-based models difficult to scale. As a result, it would be beneficial if useful representations could be derived from raw data without the need for manual annotations. In this paper, we propose to address the problem of self-supervised representation learning with multi-modal ultrasound video-speech raw data. For this case, we assume that there is a high correlation between the ultrasound video and the corresponding narrative speech audio of the sonographer. In order to learn meaningful representations, the model needs to identify such correlation and at the same time understand the underlying anatomical features. We designed a framework to model the correspondence between video and audio without any kind of human annotations. Within this framework, we introduce cross-modal contrastive learning and an affinity-aware self-paced learning scheme to enhance correlation modelling. Experimental evaluations on multi-modal fetal ultrasound video and audio show that the proposed approach is able to learn strong representations and transfers well to downstream tasks of standard plane detection and eye-gaze prediction.

Richard Droste, Lior Drukker, Aris T. Papageorghiou et al.

We present the first system that provides real-time probe movement guidance for acquiring standard planes in routine freehand obstetric ultrasound scanning. Such a system can contribute to the worldwide deployment of obstetric ultrasound scanning by lowering the required level of operator expertise. The system employs an artificial neural network that receives the ultrasound video signal and the motion signal of an inertial measurement unit (IMU) that is attached to the probe, and predicts a guidance signal. The network termed US-GuideNet predicts either the movement towards the standard plane position (goal prediction), or the next movement that an expert sonographer would perform (action prediction). While existing models for other ultrasound applications are trained with simulations or phantoms, we train our model with real-world ultrasound video and probe motion data from 464 routine clinical scans by 17 accredited sonographers. Evaluations for 3 standard plane types show that the model provides a useful guidance signal with an accuracy of 88.8% for goal prediction and 90.9% for action prediction.

Jianbo Jiao, Richard Droste, Lior Drukker et al.

Recent advances in deep learning have achieved promising performance for medical image analysis, while in most cases ground-truth annotations from human experts are necessary to train the deep model. In practice, such annotations are expensive to collect and can be scarce for medical imaging applications. Therefore, there is significant interest in learning representations from unlabelled raw data. In this paper, we propose a self-supervised learning approach to learn meaningful and transferable representations from medical imaging video without any type of human annotation. We assume that in order to learn such a representation, the model should identify anatomical structures from the unlabelled data. Therefore we force the model to address anatomy-aware tasks with free supervision from the data itself. Specifically, the model is designed to correct the order of a reshuffled video clip and at the same time predict the geometric transformation applied to the video clip. Experiments on fetal ultrasound video show that the proposed approach can effectively learn meaningful and strong representations, which transfer well to downstream tasks like standard plane detection and saliency prediction.

Richard Droste, Pierre Chatelain, Lior Drukker et al.

Anatomical landmarks are a crucial prerequisite for many medical imaging tasks. Usually, the set of landmarks for a given task is predefined by experts. The landmark locations for a given image are then annotated manually or via machine learning methods trained on manual annotations. In this paper, in contrast, we present a method to automatically discover and localize anatomical landmarks in medical images. Specifically, we consider landmarks that attract the visual attention of humans, which we term visually salient landmarks. We illustrate the method for fetal neurosonographic images. First, full-length clinical fetal ultrasound scans are recorded with live sonographer gaze-tracking. Next, a convolutional neural network (CNN) is trained to predict the gaze point distribution (saliency map) of the sonographers on scan video frames. The CNN is then used to predict saliency maps of unseen fetal neurosonographic images, and the landmarks are extracted as the local maxima of these saliency maps. Finally, the landmarks are matched across images by clustering the landmark CNN features. We show that the discovered landmarks can be used within affine image registration, with average landmark alignment errors between 4.1% and 10.9% of the fetal head long axis length.

Omar S. Al-Kadi, Daniel Y. F. Chung, Constantin C. Coussios et al.

Assessing tumor tissue heterogeneity via ultrasound has recently been suggested for predicting early response to treatment. The ultrasound backscattering characteristics can assist in better understanding the tumor texture by highlighting local concentration and spatial arrangement of tissue scatterers. However, it is challenging to quantify the various tissue heterogeneities ranging from fine-to-coarse of the echo envelope peaks in tumor texture. Local parametric fractal features extracted via maximum likelihood estimation from five well-known statistical model families are evaluated for the purpose of ultrasound tissue characterization. The fractal dimension (self-similarity measure) was used to characterize the spatial distribution of scatterers, while the Lacunarity (sparsity measure) was applied to determine scatterer number density. Performance was assessed based on 608 cross-sectional clinical ultrasound RF images of liver tumors (230 and 378 demonstrating respondent and non-respondent cases, respectively). Crossvalidation via leave-one-tumor-out and with different k-folds methodologies using a Bayesian classifier were employed for validation. The fractal properties of the backscattered echoes based on the Nakagami model (Nkg) and its extend four-parameter Nakagami-generalized inverse Gaussian (NIG) distribution achieved best results - with nearly similar performance - for characterizing liver tumor tissue. Accuracy, sensitivity and specificity for the Nkg/NIG were: 85.6%/86.3%, 94.0%/96.0%, and 73.0%/71.0%, respectively. Other statistical models, such as the Rician, Rayleigh, and K-distribution were found to not be as effective in characterizing the subtle changes in tissue texture as an indication of response to treatment. Employing the most relevant and practical statistical model could have potential consequences for the design of an early and effective clinical therapy.

Jianbo Jiao, Ana I. L. Namburete, Aris T. Papageorghiou et al.

Fetal brain magnetic resonance imaging (MRI) offers exquisite images of the developing brain but is not suitable for anomaly screening. For this ultrasound (US) is employed. While expert sonographers are adept at reading US images, MR images are much easier for non-experts to interpret. Hence in this paper we seek to produce images with MRI-like appearance directly from clinical US images. Our own clinical motivation is to seek a way to communicate US findings to patients or clinical professionals unfamiliar with US, but in medical image analysis such a capability is potentially useful, for instance, for US-MRI registration or fusion. Our model is self-supervised and end-to-end trainable. Specifically, based on an assumption that the US and MRI data share a similar anatomical latent space, we first utilise an extractor to determine shared latent features, which are then used for data synthesis. Since paired data was unavailable for our study (and rare in practice), we propose to enforce the distributions to be similar instead of employing pixel-wise constraints, by adversarial learning in both the image domain and latent space. Furthermore, we propose an adversarial structural constraint to regularise the anatomical structures between the two modalities during the synthesis. A cross-modal attention scheme is proposed to leverage non-local spatial correlations. The feasibility of the approach to produce realistic looking MR images is demonstrated quantitatively and with a qualitative evaluation compared to real fetal MR images.

Huan Qi, Sally Collins, J. Alison Noble

Semantic contour detection is a challenging problem that is often met in medical imaging, of which placental image analysis is a particular example. In this paper, we investigate utero-placental interface (UPI) detection in 2D placental ultrasound images by formulating it as a semantic contour detection problem. As opposed to natural images, placental ultrasound images contain specific anatomical structures thus have unique geometry. We argue it would be beneficial for UPI detectors to incorporate global context modelling in order to reduce unwanted false positive UPI predictions. Our approach, namely UPI-Net, aims to capture long-range dependencies in placenta geometry through lightweight global context modelling and effective multi-scale feature aggregation. We perform a subject-level 10-fold nested cross-validation on a placental ultrasound database (4,871 images with labelled UPI from 49 scans). Experimental results demonstrate that, without introducing considerable computational overhead, UPI-Net yields the highest performance in terms of standard contour detection metrics, compared to other competitive benchmarks.

Tapabrata Chakraborti, Arijit Patra, Alison Noble

Was it fair that Harry was hired but not Barry? Was it fair that Pam was fired instead of Sam? How can one ensure fairness when an intelligent algorithm takes these decisions instead of a human? How can one ensure that the decisions were taken based on merit and not on protected attributes like race or sex? These are the questions that must be answered now that many decisions in real life can be made through machine learning. However research in fairness of algorithms has focused on the counterfactual questions "what if?" or "why?", whereas in real life most subjective questions of consequence are contrastive: "why this but not that?". We introduce concepts and mathematical tools using causal inference to address contrastive fairness in algorithmic decision-making with illustrative examples.

Richard Droste, Yifan Cai, Harshita Sharma et al.

Image representations are commonly learned from class labels, which are a simplistic approximation of human image understanding. In this paper we demonstrate that transferable representations of images can be learned without manual annotations by modeling human visual attention. The basis of our analyses is a unique gaze tracking dataset of sonographers performing routine clinical fetal anomaly screenings. Models of sonographer visual attention are learned by training a convolutional neural network (CNN) to predict gaze on ultrasound video frames through visual saliency prediction or gaze-point regression. We evaluate the transferability of the learned representations to the task of ultrasound standard plane detection in two contexts. Firstly, we perform transfer learning by fine-tuning the CNN with a limited number of labeled standard plane images. We find that fine-tuning the saliency predictor is superior to training from random initialization, with an average F1-score improvement of 9.6% overall and 15.3% for the cardiac planes. Secondly, we train a simple softmax regression on the feature activations of each CNN layer in order to evaluate the representations independently of transfer learning hyper-parameters. We find that the attention models derive strong representations, approaching the precision of a fully-supervised baseline model for all but the last layer.

Arijit Patra, J. A. Noble

Fetal heart motion is an important diagnostic indicator for structural detection and functional assessment of congenital heart disease. We propose an approach towards integrating deep convolutional and recurrent architectures that utilize localized spatial and temporal features of different anatomical substructures within a global spatiotemporal context for interpretation of fetal echocardiography videos. We formulate our task as a cardiac structure localization problem with convolutional architectures for aggregating global spatial context and detecting anatomical structures on spatial region proposals. This information is aggregated temporally by recurrent architectures to quantify the progressive motion patterns. We experimentally show that the resulting architecture combines anatomical landmark detection at the frame-level over multiple video sequences-with temporal progress of the associated anatomical motions to encode local spatiotemporal fetal heart dynamics and is validated on a real-world clinical dataset.

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.

Davis M. Vigneault, Weidi Xie, Carolyn Y. Ho et al.

Pixelwise segmentation of the left ventricular (LV) myocardium and the four cardiac chambers in 2-D steady state free precession (SSFP) cine sequences is an essential preprocessing step for a wide range of analyses. Variability in contrast, appearance, orientation, and placement of the heart between patients, clinical views, scanners, and protocols makes fully automatic semantic segmentation a notoriously difficult problem. Here, we present $Ω$-Net (Omega-Net): a novel convolutional neural network (CNN) architecture for simultaneous localization, transformation into a canonical orientation, and semantic segmentation. First, an initial segmentation is performed on the input image, second, the features learned during this initial segmentation are used to predict the parameters needed to transform the input image into a canonical orientation, and third, a final segmentation is performed on the transformed image. In this work, $Ω$-Nets of varying depths were trained to detect five foreground classes in any of three clinical views (short axis, SA, four-chamber, 4C, two-chamber, 2C), without prior knowledge of the view being segmented. The architecture was trained on a cohort of patients with hypertrophic cardiomyopathy and healthy control subjects. Network performance as measured by weighted foreground intersection-over-union (IoU) was substantially improved in the best-performing $Ω$- Net compared with U-Net segmentation without localization or orientation. In addition, Ω-Net was retrained from scratch on the 2017 MICCAI ACDC dataset, and achieves state-of-the-art results on the LV and RV bloodpools, and performed slightly worse in segmentation of the LV myocardium. We conclude this architecture represents a substantive advancement over prior approaches, with implications for biomedical image segmentation more generally.

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.

Weilin Huang, Christopher P. Bridge, J. Alison Noble et al.

We present an automatic method to describe clinically useful information about scanning, and to guide image interpretation in ultrasound (US) videos of the fetal heart. Our method is able to jointly predict the visibility, viewing plane, location and orientation of the fetal heart at the frame level. The contributions of the paper are three-fold: (i) a convolutional neural network architecture is developed for a multi-task prediction, which is computed by sliding a 3x3 window spatially through convolutional maps. (ii) an anchor mechanism and Intersection over Union (IoU) loss are applied for improving localization accuracy. (iii) a recurrent architecture is designed to recursively compute regional convolutional features temporally over sequential frames, allowing each prediction to be conditioned on the whole video. This results in a spatial-temporal model that precisely describes detailed heart parameters in challenging US videos. We report results on a real-world clinical dataset, where our method achieves performance on par with expert annotations.