Mohammad Yaqub

Karim Lekadir, Aasa Feragen, Abdul Joseph Fofanah et al. · eth-zurich

Despite major advances in artificial intelligence (AI) for medicine and healthcare, the deployment and adoption of AI technologies remain limited in real-world clinical practice. In recent years, concerns have been raised about the technical, clinical, ethical and legal risks associated with medical AI. To increase real world adoption, it is essential that medical AI tools are trusted and accepted by patients, clinicians, health organisations and authorities. This work describes the FUTURE-AI guideline as the first international consensus framework for guiding the development and deployment of trustworthy AI tools in healthcare. The FUTURE-AI consortium was founded in 2021 and currently comprises 118 inter-disciplinary experts from 51 countries representing all continents, including AI scientists, clinicians, ethicists, and social scientists. Over a two-year period, the consortium defined guiding principles and best practices for trustworthy AI through an iterative process comprising an in-depth literature review, a modified Delphi survey, and online consensus meetings. The FUTURE-AI framework was established based on 6 guiding principles for trustworthy AI in healthcare, i.e. Fairness, Universality, Traceability, Usability, Robustness and Explainability. Through consensus, a set of 28 best practices were defined, addressing technical, clinical, legal and socio-ethical dimensions. The recommendations cover the entire lifecycle of medical AI, from design, development and validation to regulation, deployment, and monitoring. FUTURE-AI is a risk-informed, assumption-free guideline which provides a structured approach for constructing medical AI tools that will be trusted, deployed and adopted in real-world practice. Researchers are encouraged to take the recommendations into account in proof-of-concept stages to facilitate future translation towards clinical practice of medical AI.

Hanan Gani, Muzammal Naseer, Mohammad Yaqub

Vision Transformer (ViT), a radically different architecture than convolutional neural networks offers multiple advantages including design simplicity, robustness and state-of-the-art performance on many vision tasks. However, in contrast to convolutional neural networks, Vision Transformer lacks inherent inductive biases. Therefore, successful training of such models is mainly attributed to pre-training on large-scale datasets such as ImageNet with 1.2M or JFT with 300M images. This hinders the direct adaption of Vision Transformer for small-scale datasets. In this work, we show that self-supervised inductive biases can be learned directly from small-scale datasets and serve as an effective weight initialization scheme for fine-tuning. This allows to train these models without large-scale pre-training, changes to model architecture or loss functions. We present thorough experiments to successfully train monolithic and non-monolithic Vision Transformers on five small datasets including CIFAR10/100, CINIC10, SVHN, Tiny-ImageNet and two fine-grained datasets: Aircraft and Cars. Our approach consistently improves the performance of Vision Transformers while retaining their properties such as attention to salient regions and higher robustness. Our codes and pre-trained models are available at: https://github.com/hananshafi/vits-for-small-scale-datasets.

Faris Almalik, Mohammad Yaqub, Karthik Nandakumar

Vision Transformers (ViT) are competing to replace Convolutional Neural Networks (CNN) for various computer vision tasks in medical imaging such as classification and segmentation. While the vulnerability of CNNs to adversarial attacks is a well-known problem, recent works have shown that ViTs are also susceptible to such attacks and suffer significant performance degradation under attack. The vulnerability of ViTs to carefully engineered adversarial samples raises serious concerns about their safety in clinical settings. In this paper, we propose a novel self-ensembling method to enhance the robustness of ViT in the presence of adversarial attacks. The proposed Self-Ensembling Vision Transformer (SEViT) leverages the fact that feature representations learned by initial blocks of a ViT are relatively unaffected by adversarial perturbations. Learning multiple classifiers based on these intermediate feature representations and combining these predictions with that of the final ViT classifier can provide robustness against adversarial attacks. Measuring the consistency between the various predictions can also help detect adversarial samples. Experiments on two modalities (chest X-ray and fundoscopy) demonstrate the efficacy of SEViT architecture to defend against various adversarial attacks in the gray-box (attacker has full knowledge of the target model, but not the defense mechanism) setting. Code: https://github.com/faresmalik/SEViT

Rand Muhtaseb, Mohammad Yaqub

Learning spatiotemporal features is an important task for efficient video understanding especially in medical images such as echocardiograms. Convolutional neural networks (CNNs) and more recent vision transformers (ViTs) are the most commonly used methods with limitations per each. CNNs are good at capturing local context but fail to learn global information across video frames. On the other hand, vision transformers can incorporate global details and long sequences but are computationally expensive and typically require more data to train. In this paper, we propose a method that addresses the limitations we typically face when training on medical video data such as echocardiographic scans. The algorithm we propose (EchoCoTr) utilizes the strength of vision transformers and CNNs to tackle the problem of estimating the left ventricular ejection fraction (LVEF) on ultrasound videos. We demonstrate how the proposed method outperforms state-of-the-art work to-date on the EchoNet-Dynamic dataset with MAE of 3.95 and $R^2$ of 0.82. These results show noticeable improvement compared to all published research. In addition, we show extensive ablations and comparisons with several algorithms, including ViT and BERT. The code is available at https://github.com/BioMedIA-MBZUAI/EchoCoTr.

Ahmed Al Mahrooqi, Dmitrii Medvedev, Rand Muhtaseb et al.

Glaucoma is one of the most severe eye diseases, characterized by rapid progression and leading to irreversible blindness. It is often the case that diagnostics is carried out when one's sight has already significantly degraded due to the lack of noticeable symptoms at early stage of the disease. Regular glaucoma screenings of the population shall improve early-stage detection, however the desirable frequency of etymological checkups is often not feasible due to the excessive load imposed by manual diagnostics on limited number of specialists. Considering the basic methodology to detect glaucoma is to analyze fundus images for the optic-disc-to-optic-cup ratio, Machine Learning algorithms can offer sophisticated methods for image processing and classification. In our work, we propose an advanced image pre-processing technique combined with a multi-view network of deep classification models to categorize glaucoma. Our Glaucoma Automated Retinal Detection Network (GARDNet) has been successfully tested on Rotterdam EyePACS AIROGS dataset with an AUC of 0.92, and then additionally fine-tuned and tested on RIM-ONE DL dataset with an AUC of 0.9308 outperforming the state-of-the-art of 0.9272. Our code is available on https://github.com/ahmed1996said/gardnet

Santosh Sanjeev, Salwa K. Al Khatib, Mai A. Shaaban et al.

Previous deep learning efforts have focused on improving the performance of Pulmonary Embolism(PE) diagnosis from Computed Tomography (CT) scans using Convolutional Neural Networks (CNN). However, the features from CT scans alone are not always sufficient for the diagnosis of PE. CT scans along with electronic heath records (EHR) can provide a better insight into the patients condition and can lead to more accurate PE diagnosis. In this paper, we propose Pulmonary Embolism Detection using Contrastive Learning (PECon), a supervised contrastive pretraining strategy that employs both the patients CT scans as well as the EHR data, aiming to enhance the alignment of feature representations between the two modalities and leverage information to improve the PE diagnosis. In order to achieve this, we make use of the class labels and pull the sample features of the same class together, while pushing away those of the other class. Results show that the proposed work outperforms the existing techniques and achieves state-of-the-art performance on the RadFusion dataset with an F1-score of 0.913, accuracy of 0.90 and an AUROC of 0.943. Furthermore, we also explore the explainability of our approach in comparison to other methods. Our code is publicly available at https://github.com/BioMedIA-MBZUAI/PECon.

Otabek Nazarov, Mohammad Yaqub, Karthik Nandakumar

Chest X-ray is one of the most popular medical imaging modalities due to its accessibility and effectiveness. However, there is a chronic shortage of well-trained radiologists who can interpret these images and diagnose the patient's condition. Therefore, automated radiology report generation can be a very helpful tool in clinical practice. A typical report generation workflow consists of two main steps: (i) encoding the image into a latent space and (ii) generating the text of the report based on the latent image embedding. Many existing report generation techniques use a standard convolutional neural network (CNN) architecture for image encoding followed by a Transformer-based decoder for medical text generation. In most cases, CNN and the decoder are trained jointly in an end-to-end fashion. In this work, we primarily focus on understanding the relative importance of encoder and decoder components. Towards this end, we analyze four different image encoding approaches: direct, fine-grained, CLIP-based, and Cluster-CLIP-based encodings in conjunction with three different decoders on the large-scale MIMIC-CXR dataset. Among these encoders, the cluster CLIP visual encoder is a novel approach that aims to generate more discriminative and explainable representations. CLIP-based encoders produce comparable results to traditional CNN-based encoders in terms of NLP metrics, while fine-grained encoding outperforms all other encoders both in terms of NLP and clinical accuracy metrics, thereby validating the importance of image encoder to effectively extract semantic information. GitHub repository: https://github.com/mudabek/encoding-cxr-report-gen

Aleksandr Matsun, Dana O. Mohamed, Sharon Chokuwa et al.

The domain shift between training and testing data presents a significant challenge for training generalizable deep learning models. As a consequence, the performance of models trained with the independent and identically distributed (i.i.d) assumption deteriorates when deployed in the real world. This problem is exacerbated in the medical imaging context due to variations in data acquisition across clinical centers, medical apparatus, and patients. Domain generalization (DG) aims to address this problem by learning a model that generalizes well to any unseen target domain. Many domain generalization techniques were unsuccessful in learning domain-invariant representations due to the large domain shift. Furthermore, multiple tasks in medical imaging are not yet extensively studied in existing literature when it comes to DG point of view. In this paper, we introduce a DG method that re-establishes the model objective function as a maximization of mutual information with a large pretrained model to the medical imaging field. We re-visit the problem of DG in Diabetic Retinopathy (DR) classification to establish a clear benchmark with a correct model selection strategy and to achieve robust domain-invariant representation for an improved generalization. Moreover, we conduct extensive experiments on public datasets to show that our proposed method consistently outperforms the previous state-of-the-art by a margin of 5.25% in average accuracy and a lower standard deviation. Source code available at https://github.com/BioMedIA-MBZUAI/DGM-DR

Simon Graham, Quoc Dang Vu, Mostafa Jahanifar et al.

Nuclear detection, segmentation and morphometric profiling are essential in helping us further understand the relationship between histology and patient outcome. To drive innovation in this area, we setup a community-wide challenge using the largest available dataset of its kind to assess nuclear segmentation and cellular composition. Our challenge, named CoNIC, stimulated the development of reproducible algorithms for cellular recognition with real-time result inspection on public leaderboards. We conducted an extensive post-challenge analysis based on the top-performing models using 1,658 whole-slide images of colon tissue. With around 700 million detected nuclei per model, associated features were used for dysplasia grading and survival analysis, where we demonstrated that the challenge's improvement over the previous state-of-the-art led to significant boosts in downstream performance. Our findings also suggest that eosinophils and neutrophils play an important role in the tumour microevironment. We release challenge models and WSI-level results to foster the development of further methods for biomarker discovery.

Umaima Rahman, Raza Imam, Mohammad Yaqub et al.

In medical image classification, supervised learning is challenging due to the scarcity of labeled medical images. To address this, we leverage the visual-textual alignment within Vision-Language Models (VLMs) to enable unsupervised learning of a medical image classifier. In this work, we propose \underline{Med}ical \underline{Un}supervised \underline{A}daptation (\texttt{MedUnA}) of VLMs, where the LLM-generated descriptions for each class are encoded into text embeddings and matched with class labels via a cross-modal adapter. This adapter attaches to a visual encoder of \texttt{MedCLIP} and aligns the visual embeddings through unsupervised learning, driven by a contrastive entropy-based loss and prompt tuning. Thereby, improving performance in scenarios where textual information is more abundant than labeled images, particularly in the healthcare domain. Unlike traditional VLMs, \texttt{MedUnA} uses \textbf{unpaired images and text} for learning representations and enhances the potential of VLMs beyond traditional constraints. We evaluate the performance on three chest X-ray datasets and two multi-class datasets (diabetic retinopathy and skin lesions), showing significant accuracy gains over the zero-shot baseline. Our code is available at https://github.com/rumaima/meduna.

Numan Saeed, Ikboljon Sobirov, Roba Al Majzoub et al.

When oncologists estimate cancer patient survival, they rely on multimodal data. Even though some multimodal deep learning methods have been proposed in the literature, the majority rely on having two or more independent networks that share knowledge at a later stage in the overall model. On the other hand, oncologists do not do this in their analysis but rather fuse the information in their brain from multiple sources such as medical images and patient history. This work proposes a deep learning method that mimics oncologists' analytical behavior when quantifying cancer and estimating patient survival. We propose TMSS, an end-to-end Transformer based Multimodal network for Segmentation and Survival prediction that leverages the superiority of transformers that lies in their abilities to handle different modalities. The model was trained and validated for segmentation and prognosis tasks on the training dataset from the HEad & NeCK TumOR segmentation and the outcome prediction in PET/CT images challenge (HECKTOR). We show that the proposed prognostic model significantly outperforms state-of-the-art methods with a concordance index of 0.763+/-0.14 while achieving a comparable dice score of 0.772+/-0.030 to a standalone segmentation model. The code is publicly available.

Numan Saeed, Muhammad Ridzuan, Hussain Alasmawi et al.

The number of studies on deep learning for medical diagnosis is expanding, and these systems are often claimed to outperform clinicians. However, only a few systems have shown medical efficacy. From this perspective, we examine a wide range of deep learning algorithms for the assessment of glioblastoma - a common brain tumor in older adults that is lethal. Surgery, chemotherapy, and radiation are the standard treatments for glioblastoma patients. The methylation status of the MGMT promoter, a specific genetic sequence found in the tumor, affects chemotherapy's effectiveness. MGMT promoter methylation improves chemotherapy response and survival in several cancers. MGMT promoter methylation is determined by a tumor tissue biopsy, which is then genetically tested. This lengthy and invasive procedure increases the risk of infection and other complications. Thus, researchers have used deep learning models to examine the tumor from brain MRI scans to determine the MGMT promoter's methylation state. We employ deep learning models and one of the largest public MRI datasets of 585 participants to predict the methylation status of the MGMT promoter in glioblastoma tumors using MRI scans. We test these models using Grad-CAM, occlusion sensitivity, feature visualizations, and training loss landscapes. Our results show no correlation between these two, indicating that external cohort data should be used to verify these models' performance to assure the accuracy and reliability of deep learning systems in cancer diagnosis.

Fadillah Maani, Asim Ukaye, Nada Saadi et al.

Echocardiography has become an indispensable clinical imaging modality for general heart health assessment. From calculating biomarkers such as ejection fraction to the probability of a patient's heart failure, accurate segmentation of the heart structures allows doctors to assess the heart's condition and devise treatments with greater precision and accuracy. However, achieving accurate and reliable left ventricle segmentation is time-consuming and challenging due to different reasons. Hence, clinicians often rely on segmenting the left ventricular (LV) in two specific echocardiogram frames to make a diagnosis. This limited coverage in manual LV segmentation poses a challenge for developing automatic LV segmentation with high temporal consistency, as the resulting dataset is typically annotated sparsely. In response to this challenge, this work introduces SimLVSeg, a novel paradigm that enables video-based networks for consistent LV segmentation from sparsely annotated echocardiogram videos. SimLVSeg consists of self-supervised pre-training with temporal masking, followed by weakly supervised learning tailored for LV segmentation from sparse annotations. We demonstrate how SimLVSeg outperforms the state-of-the-art solutions by achieving a 93.32% (95%CI 93.21-93.43%) dice score on the largest 2D+time echocardiography dataset (EchoNet-Dynamic) while being more efficient. SimLVSeg is compatible with two types of video segmentation networks: 2D super image and 3D segmentation. To show the effectiveness of our approach, we provide extensive ablation studies, including pre-training settings and various deep learning backbones. We further conduct an out-of-distribution test to showcase SimLVSeg's generalizability on unseen distribution (CAMUS dataset). The code is publicly available at https://github.com/fadamsyah/SimLVSeg.

Ikboljon Sobirov, Cheng Xie, Muhammad Siddique et al.

Since the emergence of convolutional neural networks (CNNs) and, later, vision transformers (ViTs), deep learning architectures have predominantly relied on identical block types with varying hyperparameters. We propose a novel block alternation strategy to leverage the complementary strengths of different architectural designs, assembling structurally distinct components similar to Lego blocks. We introduce LegoNet, a deep learning framework that alternates CNN-based and SwinViT-based blocks to enhance feature learning for medical image segmentation. We investigate three variations of LegoNet and apply this concept to a previously unexplored clinical problem: the segmentation of the internal mammary artery (IMA), aorta, and perivascular adipose tissue (PVAT) from computed tomography angiography (CTA) scans. These PVAT regions have been shown to possess prognostic value in assessing cardiovascular risk and primary clinical outcomes. We evaluate LegoNet on large datasets, achieving superior performance to other leading architectures. Furthermore, we assess the model's generalizability on external testing cohorts, where an expert clinician corrects the model's segmentations, achieving DSC > 0.90 across various external, international, and public cohorts. To further validate the model's clinical reliability, we perform intra- and inter-observer variability analysis, demonstrating strong agreement with human annotations. The proposed methodology has significant implications for diagnostic cardiovascular management and early prognosis, offering a robust, automated solution for vascular and perivascular segmentation and risk assessment in clinical practice, paving the way for personalised medicine.

Mohammad Areeb Qazi, Mohammed Talha Alam, Ibrahim Almakky et al.

Precise estimation of fetal biometry parameters from ultrasound images is vital for evaluating fetal growth, monitoring health, and identifying potential complications reliably. However, the automated computerized segmentation of the fetal head, abdomen, and femur from ultrasound images, along with the subsequent measurement of fetal biometrics, remains challenging. In this work, we propose a multi-task learning approach to classify the region into head, abdomen and femur as well as estimate the associated parameters. We were able to achieve a mean absolute error (MAE) of 1.08 mm on head circumference, 1.44 mm on abdomen circumference and 1.10 mm on femur length with a classification accuracy of 99.91\% on a dataset of fetal Ultrasound images. To achieve this, we leverage a weighted joint classification and segmentation loss function to train a U-Net architecture with an added classification head. The code can be accessed through \href{https://github.com/BioMedIA-MBZUAI/Multi-Task-Learning-Approach-for-Unified-Biometric-Estimation-from-Fetal-Ultrasound-Anomaly-Scans.git}{\texttt{Github}

Sayed Hashim, Karthik Nandakumar, Mohammad Yaqub

We have gained access to vast amounts of multi-omics data thanks to Next Generation Sequencing. However, it is challenging to analyse this data due to its high dimensionality and much of it not being annotated. Lack of annotated data is a significant problem in machine learning, and Self-Supervised Learning (SSL) methods are typically used to deal with limited labelled data. However, there is a lack of studies that use SSL methods to exploit inter-omics relationships on unlabelled multi-omics data. In this work, we develop a novel and efficient pre-training paradigm that consists of various SSL components, including but not limited to contrastive alignment, data recovery from corrupted samples, and using one type of omics data to recover other omic types. Our pre-training paradigm improves performance on downstream tasks with limited labelled data. We show that our approach outperforms the state-of-the-art method in cancer type classification on the TCGA pan-cancer dataset in semi-supervised setting. Moreover, we show that the encoders that are pre-trained using our approach can be used as powerful feature extractors even without fine-tuning. Our ablation study shows that the method is not overly dependent on any pretext task component. The network architectures in our approach are designed to handle missing omic types and multiple datasets for pre-training and downstream training. Our pre-training paradigm can be extended to perform zero-shot classification of rare cancers.

Hussam Azzuni, Muhammad Ridzuan, Min Xu et al.

Nuclei segmentation and classification is the first and most crucial step that is utilized for many different microscopy medical analysis applications. However, it suffers from many issues such as the segmentation of small objects, imbalance, and fine-grained differences between types of nuclei. In this paper, multiple different contributions were done tackling these problems present. Firstly, the recently released "ConvNeXt" was used as the encoder for HoVer-Net model since it leverages the key components of transformers that make them perform well. Secondly, to enhance the visual differences between nuclei, a multi-channel color space-based approach is used to aid the model in extracting distinguishing features. Thirdly, Unified Focal loss (UFL) was used to tackle the background-foreground imbalance. Finally, Sharpness-Aware Minimization (SAM) was used to ensure generalizability of the model. Overall, we were able to outperform the current state-of-the-art (SOTA), HoVer-Net, on the preliminary test set of the CoNiC Challenge 2022 by 12.489% mPQ+.

Mohammad Areeb Qazi, Munachiso S Nwadike, Ibrahim Almakky et al.

Foundational models are trained on extensive datasets to capture the general trends of a domain. However, in medical imaging, the scarcity of data makes pre-training for every domain, modality, or task challenging. Instead of building separate models, we propose MAFM^3 (Modular Adaptation of Foundation Models for Multi-Modal Medical AI), a framework that enables a single foundation model to expand into diverse domains, tasks, and modalities through lightweight modular components. These components serve as specialized skill sets that allow the system to flexibly activate the appropriate capability at the inference time, depending on the input type or clinical objective. Unlike conventional adaptation methods that treat each new task or modality in isolation, MAFM^3 provides a unified and expandable framework for efficient multitask and multimodality adaptation. Empirically, we validate our approach by adapting a chest CT foundation model initially trained for classification into prognosis and segmentation modules. Our results show improved performance on both tasks. Furthermore, by incorporating PET scans, MAFM^3 achieved an improvement in the Dice score 5% compared to the respective baselines. These findings establish that foundation models, when equipped with modular components, are not inherently constrained to their initial training scope but can evolve into multitask, multimodality systems for medical imaging. The code implementation of this work can be found at https://github.com/Areeb2735/CTscan_prognosis_VLM

Raza Imam, Ibrahim Almakky, Salma Alrashdi et al.

Deep Learning methods have recently seen increased adoption in medical imaging applications. However, elevated vulnerabilities have been explored in recent Deep Learning solutions, which can hinder future adoption. Particularly, the vulnerability of Vision Transformer (ViT) to adversarial, privacy, and confidentiality attacks raise serious concerns about their reliability in medical settings. This work aims to enhance the robustness of self-ensembling ViTs for the tuberculosis chest x-ray classification task. We propose Self-Ensembling ViT with defensive Distillation and Adversarial training (SEDA). SEDA utilizes efficient CNN blocks to learn spatial features with various levels of abstraction from feature representations extracted from intermediate ViT blocks, that are largely unaffected by adversarial perturbations. Furthermore, SEDA leverages adversarial training in combination with defensive distillation for improved robustness against adversaries. Training using adversarial examples leads to better model generalizability and improves its ability to handle perturbations. Distillation using soft probabilities introduces uncertainty and variation into the output probabilities, making it more difficult for adversarial and privacy attacks. Extensive experiments performed with the proposed architecture and training paradigm on publicly available Tuberculosis x-ray dataset shows SOTA efficacy of SEDA compared to SEViT in terms of computational efficiency with 70x times lighter framework and enhanced robustness of +9%.

Ivo Gollini Navarrete, Mohammad Yaqub

Lung cancer is a leading cause of death worldwide. Early-stage detection of lung cancer is essential for a more favorable prognosis. Radiogenomics is an emerging discipline that combines medical imaging and genomics features for modeling patient outcomes non-invasively. This study presents a radiogenomics pipeline that has: 1) a novel mixed architecture (RA-Seg) to segment lung cancer through attention and recurrent blocks; and 2) deep feature classifiers to distinguish Epidermal Growth Factor Receptor (EGFR) mutation status. We evaluate the proposed algorithm on multiple public datasets to assess its generalizability and robustness. We demonstrate how the proposed segmentation and classification methods outperform existing baseline and SOTA approaches (73.54 Dice and 93 F1 scores).

Ikboljon Sobirov, Numan Saeed, Mohammad Yaqub

In medical imaging analysis, deep learning has shown promising results. We frequently rely on volumetric data to segment medical images, necessitating the use of 3D architectures, which are commended for their capacity to capture interslice context. However, because of the 3D convolutions, max pooling, up-convolutions, and other operations utilized in these networks, these architectures are often more inefficient in terms of time and computation than their 2D equivalents. Furthermore, there are few 3D pretrained model weights, and pretraining is often difficult. We present a simple yet effective 2D method to handle 3D data while efficiently embedding the 3D knowledge during training. We propose transforming volumetric data into 2D super images and segmenting with 2D networks to solve these challenges. Our method generates a super-resolution image by stitching slices side by side in the 3D image. We expect deep neural networks to capture and learn these properties spatially despite losing depth information. This work aims to present a novel perspective when dealing with volumetric data, and we test the hypothesis using CNN and ViT networks as well as self-supervised pretraining. While attaining equal, if not superior, results to 3D networks utilizing only 2D counterparts, the model complexity is reduced by around threefold. Because volumetric data is relatively scarce, we anticipate that our approach will entice more studies, particularly in medical imaging analysis.

Alya Almsouti, Ainur Khamitova, Darya Taratynova et al.

Assessing the severity of artifacts in pediatric brain Magnetic Resonance Imaging (MRI) is critical for diagnostic accuracy, especially in low-field systems where the signal-to-noise ratio is reduced. Manual quality assessment is time-consuming and subjective, motivating the need for robust automated solutions. In this work, we propose BRIQA (Balanced Reweighting in Image Quality Assessment), which addresses class imbalance in artifact severity levels. BRIQA uses gradient-based loss reweighting to dynamically adjust per-class contributions and employs a rotating batching scheme to ensure consistent exposure to underrepresented classes. Through experiments, no single architecture performs best across all artifact types, emphasizing the importance of architectural diversity. The rotating batching configuration improves performance across metrics by promoting balanced learning when combined with cross-entropy loss. BRIQA improves average macro F1 score from 0.659 to 0.706, with notable gains in Noise (0.430), Zipper (0.098), Positioning (0.097), Contrast (0.217), Motion (0.022), and Banding (0.012) artifact severity classification. The code is available at https://github.com/BioMedIA-MBZUAI/BRIQA.

Shahad Hardan, Hussain Alasmawi, Xiangjian Hou et al.

Vestibular schwannoma (VS) is a non-cancerous tumor located next to the ear that can cause hearing loss. Most brain MRI images acquired from patients are contrast-enhanced T1 (ceT1), with a growing interest in high-resolution T2 images (hrT2) to replace ceT1, which involves the use of a contrast agent. As hrT2 images are currently scarce, it is less likely to train robust machine learning models to segment VS or other brain structures. In this work, we propose a weakly supervised machine learning approach that learns from only ceT1 scans and adapts to segment two structures from hrT2 scans: the VS and the cochlea from the crossMoDA dataset. Our model 1) generates fake hrT2 scans from ceT1 images and segmentation masks, 2) is trained using the fake hrT2 scans, 3) predicts the augmented real hrT2 scans, and 4) is retrained again using both the fake and real hrT2. The final result of this model has been computed on an unseen testing dataset provided by the 2022 crossMoDA challenge organizers. The mean dice score and average symmetric surface distance (ASSD) are 0.78 and 0.46, respectively. The predicted segmentation masks achieved a dice score of 0.83 and an ASSD of 0.56 on the VS, and a dice score of 0.74 and an ASSD of 0.35 on the cochleas.

Sevim Cengiz, Ibrahim Almakky, Mohammad Yaqub

Deep learning models have been effective for various fetal ultrasound segmentation tasks. However, generalization to new unseen data has raised questions about their effectiveness for clinical adoption. Normally, a transition to new unseen data requires time-consuming and costly quality assurance processes to validate the segmentation performance post-transition. Segmentation quality assessment efforts have focused on natural images, where the problem has been typically formulated as a dice score regression task. In this paper, we propose a simplified Fetal Ultrasound Segmentation Quality Assessment (FUSQA) model to tackle the segmentation quality assessment when no masks exist to compare with. We formulate the segmentation quality assessment process as an automated classification task to distinguish between good and poor-quality segmentation masks for more accurate gestational age estimation. We validate the performance of our proposed approach on two datasets we collect from two hospitals using different ultrasound machines. We compare different architectures, with our best-performing architecture achieving over 90% classification accuracy on distinguishing between good and poor-quality segmentation masks from an unseen dataset. Additionally, there was only a 1.45-day difference between the gestational age reported by doctors and estimated based on CRL measurements using well-segmented masks. On the other hand, this difference increased and reached up to 7.73 days when we calculated CRL from the poorly segmented masks. As a result, AI-based approaches can potentially aid fetal ultrasound segmentation quality assessment and might detect poor segmentation in real-time screening in the future.

Asma Ahmed Hashmi, Aigerim Zhumabayeva, Nikita Kotelevskii et al.

The success of Deep Neural Network (DNN) models significantly depends on the quality of provided annotations. In medical image segmentation, for example, having multiple expert annotations for each data point is common to minimize subjective annotation bias. Then, the goal of estimation is to filter out the label noise and recover the ground-truth masks, which are not explicitly given. This paper proposes a probabilistic model for noisy observations that allows us to build a confident classification and segmentation models. To accomplish it, we explicitly model label noise and introduce a new information-based regularization that pushes the network to recover the ground-truth labels. In addition, for segmentation task we adjust the loss function by prioritizing learning in high-confidence regions where all the annotators agree on labeling. We evaluate the proposed method on a series of classification tasks such as noisy versions of MNIST, CIFAR-10, Fashion-MNIST datasets as well as CIFAR-10N, which is real-world dataset with noisy human annotations. Additionally, for segmentation task, we consider several medical imaging datasets, such as, LIDC and RIGA that reflect real-world inter-variability among multiple annotators. Our experiments show that our algorithm outperforms state-of-the-art solutions for the considered classification and segmentation problems.

Abdelrahman Elsayed, Ahmed Jaheen, Mohammad Yaqub

Automated brain tumor segmentation in multi-parametric MRI remains challenging in resource-constrained settings where deep 3D networks are computationally prohibitive. We propose MMRINet, a lightweight architecture that replaces quadratic-complexity attention with linear-complexity Mamba state-space models for efficient volumetric context modeling. Novel Dual-Path Feature Refinement (DPFR) modules maximize feature diversity without additional data requirements, while Progressive Feature Aggregation (PFA) enables effective multi-scale fusion. In the BraTS-Lighthouse SSA 2025, our model achieves strong performance with an average Dice score of (0.752) and an average HD95 of (12.23) with only ~2.5M parameters, demonstrating efficient and accurate segmentation suitable for low-resource clinical environments. Our GitHub repository can be accessed here: github.com/BioMedIA-MBZUAI/MMRINet.

Xiangjian Hou, Sarit Khirirat, Mohammad Yaqub et al.

Federated learning (FL) is a distributed machine learning (ML) framework where multiple clients collaborate to train a model without exposing their private data. FL involves cycles of local computations and bi-directional communications between the clients and server. To bolster data security during this process, FL algorithms frequently employ a differential privacy (DP) mechanism that introduces noise into each client's model updates before sharing. However, while enhancing privacy, the DP mechanism often hampers convergence performance. In this paper, we posit that an optimal balance exists between the number of local steps and communication rounds, one that maximizes the convergence performance within a given privacy budget. Specifically, we present a proof for the optimal number of local steps and communication rounds that enhance the convergence bounds of the DP version of the ScaffNew algorithm. Our findings reveal a direct correlation between the optimal number of local steps, communication rounds, and a set of variables, e.g the DP privacy budget and other problem parameters, specifically in the context of strongly convex optimization. We furthermore provide empirical evidence to validate our theoretical findings.

Raza Imam, Hu Wang, Dwarikanath Mahapatra et al.

In medical imaging, vision-language models face a critical duality: pretrained networks offer broad robustness but lack subtle, modality-specific characteristics, while fine-tuned expert models achieve high in-distribution accuracy yet falter under modality shift. Existing model-merging techniques, designed for natural-image benchmarks, are simple and efficient but fail to deliver consistent gains across diverse medical modalities; their static interpolation limits reliability in varied clinical tasks. To address this, we introduce Test-Time Task adaptive merging (T^3), a backpropagation-free framework that computes per-sample interpolation coefficients via the Jensen-Shannon divergence between the two models' output distributions. T^3 dynamically preserves local precision when models agree and defers to generalist robustness under drift. To overcome the inference costs of sample-wise merging, we further propose a batch-wise extension, T^3_B, that computes a merging coefficient across a batch of samples, dramatically reducing computational bottleneck. Recognizing the lack of a standardized medical-merging benchmark, we present a rigorous cross-evaluation protocol spanning in-domain, base-to-novel, and corruptions across four modalities. Empirically, T^3 sets new state-of-the-art in Top-1 accuracy and error reduction, outperforming strong baselines while maintaining efficiency, paving the way for adaptive MVLM deployment in clinical settings. Our code is available at https://github.com/Razaimam45/TCube.

Jieyun Bai, Yitong Tang, Zihao Zhou et al.

Accurate segmentation of cervical structures in transvaginal ultrasound (TVS) is critical for assessing the risk of spontaneous preterm birth (PTB), yet the scarcity of labeled data limits the performance of supervised learning approaches. This paper introduces the Fetal Ultrasound Grand Challenge (FUGC), the first benchmark for semi-supervised learning in cervical segmentation, hosted at ISBI 2025. FUGC provides a dataset of 890 TVS images, including 500 training images, 90 validation images, and 300 test images. Methods were evaluated using the Dice Similarity Coefficient (DSC), Hausdorff Distance (HD), and runtime (RT), with a weighted combination of 0.4/0.4/0.2. The challenge attracted 10 teams with 82 participants submitting innovative solutions. The best-performing methods for each individual metric achieved 90.26\% mDSC, 38.88 mHD, and 32.85 ms RT, respectively. FUGC establishes a standardized benchmark for cervical segmentation, demonstrates the efficacy of semi-supervised methods with limited labeled data, and provides a foundation for AI-assisted clinical PTB risk assessment.

Jieyun Bai, Zihao Zhou, Yitong Tang et al.

A substantial proportion (45\%) of maternal deaths, neonatal deaths, and stillbirths occur during the intrapartum phase, with a particularly high burden in low- and middle-income countries. Intrapartum biometry plays a critical role in monitoring labor progression; however, the routine use of ultrasound in resource-limited settings is hindered by a shortage of trained sonographers. To address this challenge, the Intrapartum Ultrasound Grand Challenge (IUGC), co-hosted with MICCAI 2024, was launched. The IUGC introduces a clinically oriented multi-task automatic measurement framework that integrates standard plane classification, fetal head-pubic symphysis segmentation, and biometry, enabling algorithms to exploit complementary task information for more accurate estimation. Furthermore, the challenge releases the largest multi-center intrapartum ultrasound video dataset to date, comprising 774 videos (68,106 frames) collected from three hospitals, providing a robust foundation for model training and evaluation. In this study, we present a comprehensive overview of the challenge design, review the submissions from eight participating teams, and analyze their methods from five perspectives: preprocessing, data augmentation, learning strategy, model architecture, and post-processing. In addition, we perform a systematic analysis of the benchmark results to identify key bottlenecks, explore potential solutions, and highlight open challenges for future research. Although encouraging performance has been achieved, our findings indicate that the field remains at an early stage, and further in-depth investigation is required before large-scale clinical deployment. All benchmark solutions and the complete dataset have been publicly released to facilitate reproducible research and promote continued advances in automatic intrapartum ultrasound biometry.

Mothilal Asokan, Joseph Geo Benjamin, Mohammad Yaqub et al.

Adapting foundation models for medical image analysis requires finetuning them on a considerable amount of data because of extreme distribution shifts between natural (source) data used for pretraining and medical (target) data. However, collecting task-specific medical data for such finetuning at a central location raises many privacy concerns. Although Federated learning (FL) provides an effective means for training on private decentralized data, communication costs in federating large foundation models can quickly become a significant bottleneck, impacting the solution's scalability. In this work, we address this problem of efficient communication while ensuring effective learning in FL by combining the strengths of Parameter-Efficient Fine-tuning (PEFT) with FL. Specifically, we study plug-and-play Low-Rank Adapters (LoRA) in a federated manner to adapt the Segment Anything Model (SAM) for 3D medical image segmentation. Unlike prior works that utilize LoRA and finetune the entire decoder, we critically analyze the contribution of each granular component of SAM on finetuning performance. Thus, we identify specific layers to be federated that are very efficient in terms of communication cost while producing on-par accuracy. Our experiments show that retaining the parameters of the SAM model (including most of the decoder) in their original state during adaptation is beneficial because fine-tuning on small datasets tends to distort the inherent capabilities of the underlying foundation model. On Fed-KiTS, our approach decreases communication cost (~48x) compared to full fine-tuning while increasing performance (~6% Dice score) in 3D segmentation tasks. Our approach performs similar to SAMed while achieving ~2.8x reduction in communication and parameters to be finetuned. We further validate our approach with experiments on Fed-IXI and Prostate MRI datasets.

Mohammad Areeb Qazi, Anees Ur Rehman Hashmi, Santosh Sanjeev et al.

Deep Learning has shown great success in reshaping medical imaging, yet it faces numerous challenges hindering widespread application. Issues like catastrophic forgetting and distribution shifts in the continuously evolving data stream increase the gap between research and applications. Continual Learning offers promise in addressing these hurdles by enabling the sequential acquisition of new knowledge without forgetting previous learnings in neural networks. In this survey, we comprehensively review the recent literature on continual learning in the medical domain, highlight recent trends, and point out the practical issues. Specifically, we survey the continual learning studies on classification, segmentation, detection, and other tasks in the medical domain. Furthermore, we develop a taxonomy for the reviewed studies, identify the challenges, and provide insights to overcome them. We also critically discuss the current state of continual learning in medical imaging, including identifying open problems and outlining promising future directions. We hope this survey will provide researchers with a useful overview of the developments in the field and will further increase interest in the community. To keep up with the fast-paced advancements in this field, we plan to routinely update the repository with the latest relevant papers at https://github.com/BioMedIA-MBZUAI/awesome-cl-in-medical .

Mohammad Areeb Qazi, Ibrahim Almakky, Anees Ur Rehman Hashmi et al.

Continual learning, the ability to acquire knowledge from new data while retaining previously learned information, is a fundamental challenge in machine learning. Various approaches, including memory replay, knowledge distillation, model regularization, and dynamic network expansion, have been proposed to address this issue. Thus far, dynamic network expansion methods have achieved state-of-the-art performance at the cost of incurring significant computational overhead. This is due to the need for additional model buffers, which makes it less feasible in resource-constrained settings, particularly in the medical domain. To overcome this challenge, we propose Dynamic Model Merging, DynaMMo, a method that merges multiple networks at different stages of model training to achieve better computational efficiency. Specifically, we employ lightweight learnable modules for each task and combine them into a unified model to minimize computational overhead. DynaMMo achieves this without compromising performance, offering a cost-effective solution for continual learning in medical applications. We evaluate DynaMMo on three publicly available datasets, demonstrating its effectiveness compared to existing approaches. DynaMMo offers around 10-fold reduction in GFLOPS with a small drop of 2.76 in average accuracy when compared to state-of-the-art dynamic-based approaches. The code implementation of this work will be available upon the acceptance of this work at https://github.com/BioMedIA-MBZUAI/DynaMMo.

Ibrahim Almakky, Santosh Sanjeev, Anees Ur Rehman Hashmi et al.

Transfer learning has become a powerful tool to initialize deep learning models to achieve faster convergence and higher performance. This is especially useful in the medical imaging analysis domain, where data scarcity limits possible performance gains for deep learning models. Some advancements have been made in boosting the transfer learning performance gain by merging models starting from the same initialization. However, in the medical imaging analysis domain, there is an opportunity to merge models starting from different initializations, thus combining the features learned from different tasks. In this work, we propose MedMerge, a method whereby the weights of different models can be merged, and their features can be effectively utilized to boost performance on a new task. With MedMerge, we learn kernel-level weights that can later be used to merge the models into a single model, even when starting from different initializations. Testing on various medical imaging analysis tasks, we show that our merged model can achieve significant performance gains, with up to 7% improvement on the F1 score. The code implementation of this work is available at github.com/BioMedIA-MBZUAI/MedMerge.

Sarim Hashmi, Juan Lugo, Abdelrahman Elsayed et al.

Identifying key pathological features in brain MRIs is crucial for the long-term survival of glioma patients. However, manual segmentation is time-consuming, requiring expert intervention and is susceptible to human error. Therefore, significant research has been devoted to developing machine learning methods that can accurately segment tumors in 3D multimodal brain MRI scans. Despite their progress, state-of-the-art models are often limited by the data they are trained on, raising concerns about their reliability when applied to diverse populations that may introduce distribution shifts. Such shifts can stem from lower quality MRI technology (e.g., in sub-Saharan Africa) or variations in patient demographics (e.g., children). The BraTS-2024 challenge provides a platform to address these issues. This study presents our methodology for segmenting tumors in the BraTS-2024 SSA and Pediatric Tumors tasks using MedNeXt, comprehensive model ensembling, and thorough postprocessing. Our approach demonstrated strong performance on the unseen validation set, achieving an average Dice Similarity Coefficient (DSC) of 0.896 on the BraTS-2024 SSA dataset and an average DSC of 0.830 on the BraTS Pediatric Tumor dataset. Additionally, our method achieved an average Hausdorff Distance (HD95) of 14.682 on the BraTS-2024 SSA dataset and an average HD95 of 37.508 on the BraTS Pediatric dataset. Our GitHub repository can be accessed here: Project Repository : https://github.com/python-arch/BioMbz-Optimizing-Brain-Tumor-Segmentation-with-MedNeXt-BraTS-2024-SSA-and-Pediatrics

Mai A. Shaaban, Adnan Khan, Mohammad Yaqub

Chest X-ray images are commonly used for predicting acute and chronic cardiopulmonary conditions, but efforts to integrate them with structured clinical data face challenges due to incomplete electronic health records (EHR). This paper introduces MedPromptX, the first clinical decision support system that integrates multimodal large language models (MLLMs), few-shot prompting (FP) and visual grounding (VG) to combine imagery with EHR data for chest X-ray diagnosis. A pre-trained MLLM is utilized to complement the missing EHR information, providing a comprehensive understanding of patients' medical history. Additionally, FP reduces the necessity for extensive training of MLLMs while effectively tackling the issue of hallucination. Nevertheless, the process of determining the optimal number of few-shot examples and selecting high-quality candidates can be burdensome, yet it profoundly influences model performance. Hence, we propose a new technique that dynamically refines few-shot data for real-time adjustment to new patient scenarios. Moreover, VG narrows the search area in X-ray images, thereby enhancing the identification of abnormalities. We also release MedPromptX-VQA, a new in-context visual question answering dataset encompassing interleaved images and EHR data derived from MIMIC-IV and MIMIC-CXR-JPG databases. Results demonstrate the SOTA performance of MedPromptX, achieving an 11% improvement in F1-score compared to the baselines. Code and data are publicly available on https://github.com/BioMedIA-MBZUAI/MedPromptX.

Numan Saeed, Muhammad Ridzuan, Fadillah Adamsyah Maani et al.

Predicting the likelihood of survival is of paramount importance for individuals diagnosed with cancer as it provides invaluable information regarding prognosis at an early stage. This knowledge enables the formulation of effective treatment plans that lead to improved patient outcomes. In the past few years, deep learning models have provided a feasible solution for assessing medical images, electronic health records, and genomic data to estimate cancer risk scores. However, these models often fall short of their potential because they struggle to learn regression-aware feature representations. In this study, we propose Survival Rank-N Contrast (SurvRNC) method, which introduces a loss function as a regularizer to obtain an ordered representation based on the survival times. This function can handle censored data and can be incorporated into any survival model to ensure that the learned representation is ordinal. The model was extensively evaluated on a HEad \& NeCK TumOR (HECKTOR) segmentation and the outcome-prediction task dataset. We demonstrate that using the SurvRNC method for training can achieve higher performance on different deep survival models. Additionally, it outperforms state-of-the-art methods by 3.6% on the concordance index. The code is publicly available on https://github.com/numanai/SurvRNC

Hussain Alasmawi, Leanne Bricker, Mohammad Yaqub

Ultrasound is the primary imaging modality in clinical practice during pregnancy. More than 140M fetuses are born yearly, resulting in numerous scans. The availability of a large volume of fetal ultrasound scans presents the opportunity to train robust machine learning models. However, the abundance of scans also has its challenges, as manual labeling of each image is needed for supervised methods. Labeling is typically labor-intensive and requires expertise to annotate the images accurately. This study presents an unsupervised approach for automatically clustering ultrasound images into a large range of fetal views, reducing or eliminating the need for manual labeling. Our Fetal Ultrasound Semantic Clustering (FUSC) method is developed using a large dataset of 88,063 images and further evaluated on an additional unseen dataset of 8,187 images achieving over 92% clustering purity. The result of our investigation hold the potential to significantly impact the field of fetal ultrasound imaging and pave the way for more advanced automated labeling solutions. Finally, we make the code and the experimental setup publicly available to help advance the field.

Hu Wang, Congbo Ma, Ian Reid et al.

The advantage function is a central concept in RL that helps reduce variance in policy gradient estimates. Recently, for language modeling, Group Relative Policy Optimization (GRPO) was proposed to compute the advantage for each output by subtracting the mean reward, as the baseline, for all outputs in the group. However, it can lead to high variance when the reward advantage is inaccurately predicted. In this work, we propose Kalman Filter Enhanced Group Relative Policy Optimization (KRPO) model, by using lightweight Kalman filtering to dynamically estimate the latent reward baseline and uncertainty. This filtering technique replaces the naive group mean, enabling more adaptive advantage normalization. Our method does not require additional learned parameters over GRPO. This approach offers a simple yet effective way to incorporate multiple outputs of GRPO into advantage estimation, improving policy optimization in settings where highly dynamic reward signals are difficult to model for language models. Through the accuracies and rewards obtained from math question answering and reasoning, we show that using a more adaptive advantage estimation model, KRPO can improve the stability and performance of GRPO. The code is available at https://github.com/billhhh/KRPO_LLMs_RL.

Fadillah Maani, Anees Ur Rehman Hashmi, Numan Saeed et al.

Brain tumor segmentation is a fundamental step in assessing a patient's cancer progression. However, manual segmentation demands significant expert time to identify tumors in 3D multimodal brain MRI scans accurately. This reliance on manual segmentation makes the process prone to intra- and inter-observer variability. This work proposes a brain tumor segmentation method as part of the BraTS-GoAT challenge. The task is to segment tumors in brain MRI scans automatically from various populations, such as adults, pediatrics, and underserved sub-Saharan Africa. We employ a recent CNN architecture for medical image segmentation, namely MedNeXt, as our baseline, and we implement extensive model ensembling and postprocessing for inference. Our experiments show that our method performs well on the unseen validation set with an average DSC of 85.54% and HD95 of 27.88. The code is available on https://github.com/BioMedIA-MBZUAI/BraTS2024_BioMedIAMBZ.

Santosh Sanjeev, Nuren Zhaksylyk, Ibrahim Almakky et al.

The scarcity of well-annotated medical datasets requires leveraging transfer learning from broader datasets like ImageNet or pre-trained models like CLIP. Model soups averages multiple fine-tuned models aiming to improve performance on In-Domain (ID) tasks and enhance robustness against Out-of-Distribution (OOD) datasets. However, applying these methods to the medical imaging domain faces challenges and results in suboptimal performance. This is primarily due to differences in error surface characteristics that stem from data complexities such as heterogeneity, domain shift, class imbalance, and distributional shifts between training and testing phases. To address this issue, we propose a hierarchical merging approach that involves local and global aggregation of models at various levels based on models' hyperparameter configurations. Furthermore, to alleviate the need for training a large number of models in the hyperparameter search, we introduce a computationally efficient method using a cyclical learning rate scheduler to produce multiple models for aggregation in the weight space. Our method demonstrates significant improvements over the model souping approach across multiple datasets (around 6% gain in HAM10000 and CheXpert datasets) while maintaining low computational costs for model generation and selection. Moreover, we achieve better results on OOD datasets than model soups. The code is available at https://github.com/BioMedIA-MBZUAI/FissionFusion.

Fadillah Adamsyah Maani, Numan Saeed, Aleksandr Matsun et al.

Deep learning (DL) models have been advancing automatic medical image analysis on various modalities, including echocardiography, by offering a comprehensive end-to-end training pipeline. This approach enables DL models to regress ejection fraction (EF) directly from 2D+time echocardiograms, resulting in superior performance. However, the end-to-end training pipeline makes the learned representations less explainable. The representations may also fail to capture the continuous relation among echocardiogram clips, indicating the existence of spurious correlations, which can negatively affect the generalization. To mitigate this issue, we propose CoReEcho, a novel training framework emphasizing continuous representations tailored for direct EF regression. Our extensive experiments demonstrate that CoReEcho: 1) outperforms the current state-of-the-art (SOTA) on the largest echocardiography dataset (EchoNet-Dynamic) with MAE of 3.90 & R2 of 82.44, and 2) provides robust and generalizable features that transfer more effectively in related downstream tasks. The code is publicly available at https://github.com/fadamsyah/CoReEcho.

Hu Wang, Congbo Ma, Ibrahim Almakky et al.

Model merging, particularly through weight averaging, has shown surprising effectiveness in saving computations and improving model performance without any additional training. However, the interpretability of why and how this technique works remains unclear. In this work, we reinterpret weight-averaged model merging through the lens of interpretability and provide empirical insights into the underlying mechanisms that govern its behavior. We approach the problem from three perspectives: (1) we analyze the learned weight structures and demonstrate that model weights encode structured representations that help explain the compatibility of weight averaging; (2) we compare averaging in weight space and feature space across diverse model architectures (CNNs and ViTs) and datasets, aiming to expose under which circumstances what combination paradigm will work more effectively; (3) we study the effect of parameter scaling on prediction stability, highlighting how weight averaging acts as a form of regularization that contributes to robustness. By framing these analyses in an interpretability context, our work contributes to a more transparent and systematic understanding of model merging for stakeholders interested in the safety and reliability of untrained model combination methods. The code is available at https://github.com/billhhh/Rethink-Merge.

Hussain Alasmawi, Numan Saeed, Mohammad Yaqub

The growing demand for prenatal ultrasound imaging has intensified a global shortage of trained sonographers, creating barriers to essential fetal health monitoring. Deep learning has the potential to enhance sonographers' efficiency and support the training of new practitioners. Vision-Language Models (VLMs) are particularly promising for ultrasound interpretation, as they can jointly process images and text to perform multiple clinical tasks within a single framework. However, despite the expansion of VLMs, no standardized benchmark exists to evaluate their performance in fetal ultrasound imaging. This gap is primarily due to the modality's challenging nature, operator dependency, and the limited public availability of datasets. To address this gap, we present Fetal-Gauge, the first and largest visual question answering benchmark specifically designed to evaluate VLMs across various fetal ultrasound tasks. Our benchmark comprises over 42,000 images and 93,000 question-answer pairs, spanning anatomical plane identification, visual grounding of anatomical structures, fetal orientation assessment, clinical view conformity, and clinical diagnosis. We systematically evaluate several state-of-the-art VLMs, including general-purpose and medical-specific models, and reveal a substantial performance gap: the best-performing model achieves only 55\% accuracy, far below clinical requirements. Our analysis identifies critical limitations of current VLMs in fetal ultrasound interpretation, highlighting the urgent need for domain-adapted architectures and specialized training approaches. Fetal-Gauge establishes a rigorous foundation for advancing multimodal deep learning in prenatal care and provides a pathway toward addressing global healthcare accessibility challenges. Our benchmark will be publicly available once the paper gets accepted.

Muhammad Ridzuan, Numan Saeed, Fadillah Adamsyah Maani et al.

Survival analysis plays a crucial role in estimating the likelihood of future events for patients by modeling time-to-event data, particularly in healthcare settings where predictions about outcomes such as death and disease recurrence are essential. However, this analysis poses challenges due to the presence of censored data, where time-to-event information is missing for certain data points. Yet, censored data can offer valuable insights, provided we appropriately incorporate the censoring time during modeling. In this paper, we propose SurvCORN, a novel method utilizing conditional ordinal ranking networks to predict survival curves directly. Additionally, we introduce SurvMAE, a metric designed to evaluate the accuracy of model predictions in estimating time-to-event outcomes. Through empirical evaluation on two real-world cancer datasets, we demonstrate SurvCORN's ability to maintain accurate ordering between patient outcomes while improving individual time-to-event predictions. Our contributions extend recent advancements in ordinal regression to survival analysis, offering valuable insights into accurate prognosis in healthcare settings.

Nuren Zhaksylyk, Ibrahim Almakky, Jay Paranjape et al.

Accurate surgical instrument segmentation is essential in cataract surgery for tasks such as skill assessment and workflow optimization. However, limited annotated data makes it difficult to develop fully automatic models. Prompt-based methods like SAM2 offer flexibility yet remain highly sensitive to the point prompt placement, often leading to inconsistent segmentations. We address this issue by introducing RP-SAM2, which incorporates a novel shift block and a compound loss function to stabilize point prompts. Our approach reduces annotator reliance on precise point positioning while maintaining robust segmentation capabilities. Experiments on the Cataract1k dataset demonstrate that RP-SAM2 improves segmentation accuracy, with a 2% mDSC gain, a 21.36% reduction in mHD95, and decreased variance across random single-point prompt results compared to SAM2. Additionally, on the CaDIS dataset, pseudo masks generated by RP-SAM2 for fine-tuning SAM2's mask decoder outperformed those generated by SAM2. These results highlight RP-SAM2 as a practical, stable and reliable solution for semi-automatic instrument segmentation in data-constrained medical settings. The code is available at https://github.com/BioMedIA-MBZUAI/RP-SAM2.

Abdelrahman Elsayed, Sarim Hashmi, Mohammed Elseiagy et al.

The complex nature of medical image segmentation calls for models that are specifically designed to capture detailed, domain-specific features. Large foundation models offer considerable flexibility, yet the cost of fine-tuning these models remains a significant barrier. Parameter-Efficient Fine-Tuning (PEFT) methods, such as Low-Rank Adaptation (LoRA), efficiently update model weights with low-rank matrices but may suffer from underfitting when the chosen rank is insufficient to capture domain-specific nuances. Conversely, full-rank Singular Value Decomposition (SVD) based methods provide comprehensive updates by modifying all singular values, yet they often lack flexibility and exhibit variable performance across datasets. We propose SALT (Singular Value Adaptation with Low-Rank Transformation), a method that selectively adapts the most influential singular values using trainable scale and shift parameters while complementing this with a low-rank update for the remaining subspace. This hybrid approach harnesses the advantages of both LoRA and SVD, enabling effective adaptation without relying on increasing model size or depth. Evaluated on 5 challenging medical datasets, ranging from as few as 20 samples to 1000, SALT outperforms state-of-the-art PEFT (LoRA and SVD) by 2% to 5% in Dice with only 3.9% trainable parameters, demonstrating robust adaptation even in low-resource settings. The code for SALT is available at: https://github.com/BioMedIA-MBZUAI/SALT

Hu Wang, Salma Hassan, Yuyuan Liu et al.

In multi-modal learning, some modalities are more influential than others, and their absence can have a significant impact on classification/segmentation accuracy. Addressing this challenge, we propose a novel approach called Meta-learned Modality-weighted Knowledge Distillation (MetaKD), which enables multi-modal models to maintain high accuracy even when key modalities are missing. MetaKD adaptively estimates the importance weight of each modality through a meta-learning process. These learned importance weights guide a pairwise modality-weighted knowledge distillation process, allowing high-importance modalities to transfer knowledge to lower-importance ones, resulting in robust performance despite missing inputs. Unlike previous methods in the field, which are often task-specific and require significant modifications, our approach is designed to work in multiple tasks (e.g., segmentation and classification) with minimal adaptation. Experimental results on five prevalent datasets, including three Brain Tumor Segmentation datasets (BraTS2018, BraTS2019 and BraTS2020), the Alzheimer's Disease Neuroimaging Initiative (ADNI) classification dataset and the Audiovision-MNIST classification dataset, demonstrate the proposed model is able to outperform the compared models by a large margin. The code is available at https://github.com/billhhh/MetaKD.

Santosh Sanjeev, Fadillah Adamsyah Maani, Arsen Abzhanov et al.

With the emergence of vision language models in the medical imaging domain, numerous studies have focused on two dominant research activities: (1) report generation from Chest X-rays (CXR), and (2) synthetic scan generation from text or reports. Despite some research incorporating multi-view CXRs into the generative process, prior patient scans and reports have been generally disregarded. This can inadvertently lead to the leaving out of important medical information, thus affecting generation quality. To address this, we propose TiBiX: Leveraging Temporal information for Bidirectional X-ray and Report Generation. Considering previous scans, our approach facilitates bidirectional generation, primarily addressing two challenging problems: (1) generating the current image from the previous image and current report and (2) generating the current report based on both the previous and current images. Moreover, we extract and release a curated temporal benchmark dataset derived from the MIMIC-CXR dataset, which focuses on temporal data. Our comprehensive experiments and ablation studies explore the merits of incorporating prior CXRs and achieve state-of-the-art (SOTA) results on the report generation task. Furthermore, we attain on-par performance with SOTA image generation efforts, thus serving as a new baseline in longitudinal bidirectional CXR-to-report generation. The code is available at https://github.com/BioMedIA-MBZUAI/TiBiX.

Joseph Geo Benjamin, Mothilal Asokan, Amna Alhosani et al.

Self-supervised learning (SSL) methods are popular since they can address situations with limited annotated data by directly utilising the underlying data distribution. However, the adoption of such methods is not explored enough in ultrasound (US) imaging, especially for fetal assessment. We investigate the potential of dual-encoder SSL in utilizing unlabelled US video data to improve the performance of challenging downstream Standard Fetal Cardiac Planes (SFCP) classification using limited labelled 2D US images. We study 7 SSL approaches based on reconstruction, contrastive loss, distillation, and information theory and evaluate them extensively on a large private US dataset. Our observations and findings are consolidated from more than 500 downstream training experiments under different settings. Our primary observation shows that for SSL training, the variance of the dataset is more crucial than its size because it allows the model to learn generalisable representations, which improve the performance of downstream tasks. Overall, the BarlowTwins method shows robust performance, irrespective of the training settings and data variations, when used as an initialisation for downstream tasks. Notably, full fine-tuning with 1% of labelled data outperforms ImageNet initialisation by 12% in F1-score and outperforms other SSL initialisations by at least 4% in F1-score, thus making it a promising candidate for transfer learning from US video to image data.