Papers

Field

Top tools here

Open problems in Image segmentation

Lack of benchmarks for real-world clinical deployment1 paper raises it · 0 address it
Current benchmarks do not reflect how clinicians actually use AI systems, limiting their predictive value for real-world performance.
- Clinician use of language models diverges from how the models are evaluatedraises
Handling gigapixel whole-slide images in segmentation1 paper raises it · 1 addresses it
Existing models operate at patch scale, but clinical supervision and labels are at the whole-slide level, creating a mismatch.
- BigCell: Generating Gigapixel Whole-Slide Imagesraises · addresses
Temporal modeling of multimodal clinical trajectories1 paper raises it · 1 addresses it
Accurate risk prediction requires modeling irregular, heterogeneous clinical data over time, which current approaches often overlook.
- TRACE: Tracking Multimodal Clinical Trajectories for Inpatient Risk Predictionraises · addresses
Resistance to incidental information in clinical reasoning1 paper raises it · 0 address it
Models may misattribute or be distracted by irrelevant information in clinical documentation, leading to errors.
- Large language models are vulnerable to incidental information in clinical documentation and reasoningraises
Learning from deployment experience without retraining1 paper raises it · 1 addresses it
Frozen models cannot adapt to new clinical evidence or updated guidelines without access to weights or additional training.
- Frozen Models, Evolving Expertise: Model-Agnostic Learning from Deployment Experience for Multimodal Medical AIraises · addresses

Named by an AI from these papers' abstracts.

Open problems

Recently funded

Topics in this field

image segmentation · volumetric · medical image409 papers42 this week
radiology report generation · medical vision language · vision language models133 papers35 this week
clinical reasoning · question answering · language models210 papers24 this week
patient history · clinical data · health records112 papers22 this week
eeg foundation models · electroencephalography eeg · eeg foundation158 papers20 this week
electrocardiogram ecg · electrocardiogram · ecgs147 papers14 this week

Topics in this field

Active contributors · 90 days

Active contributors

Activity · 12 weeks

Activity

Recent papers

All papers in Research

arXiv cs.CVSep 27
Can Protein-Derived Knowledge Improve Pathology Foundation Models?
Di Zhang, Zhangpeng Gong, Jiashuai Liu, Zhi Zeng, Jiusong Ge +8
Molecularly guided pathology foundation models (PFMs) exploit transcriptomic or proteomic information to enrich whole-slide image (WSI) representations, yet effectively leveraging large standalone molecular corpora remains challenging. First, existing molecular foundation models encode protein sequences or single-cell states, not the patient-level bulk expression profiles paired with WSIs. Second, because cross-modal supervision is restricted to paired WSI-omics samples, knowledge from standalone molecular corpora reaches the pathology encoder only indirectly, creating a paired-support bottleneck. To address these challenges, we propose a three-stage framework that decouples proteomic knowledge acquisition from cross-modal transfer, yielding ProSlide, a slide-level hierarchical pathology foundation model. First, to close the modality gap, we pretrain a Proteomic Foundation Encoder (PFE) on 12,695 sample-level bulk protein profiles using virtual profile generation and expression-space multi-view pretraining. Second, we pretrain ProSlide, a patch-region-slide encoder, to predict protein expression from paired WSI-protein samples. Third, to relax the paired-support bottleneck, we…
PDF arXiv
arXiv cs.CVSep 27
ABO-Med: Accelerated Bilevel Optimization for Few-Shot Medical Image Classification
Ruoxuan Shi, Sheng Yang, Zhengxing Su, Xiaoyang Hou, Yating Liu
In recent years, bilevel optimization has been widely used in a variety of machine learning tasks. However, prior bilevel optimization algorithms generally require the computation of second-order information, which limits their practical scalability. Only recently has a first-order paradigm for bilevel optimization been established, attaining near-optimal theoretical guarantees for solving bilevel optimization problems. In this paper, we propose ABO-Med, a scalable instantiation of this paradigm for few-shot learning, by incorporating it into the model-agnostic meta-learning (MAML) framework and tailoring it to medical image classification. We also introduce Medical Adaptive RandomAugment (MedRAug), a modality-aware augmentation strategy designed for medical images. Theoretically, ABO-Med establishes the optimality of MAML-type meta-learning approaches. Empirically, ABO-Med outperforms prior baselines on several public medical datasets, with gains of 1.99% to 18.76%, while MedRAug further improves the average accuracy by 2.20% to 6.34%. Additional cross-domain experiments, augmentation ablation studies, backbone ablation studies, and training efficiency analysis further validate…
PDF arXiv
arXiv cs.CVSep 27
FOCUS: Benchmarking Retinal Model Generalization from Foundation Vision Encoders to Multimodal LLMs
David Restrepo, Chenwei Wu, Luis Filipe Nakayama, Miguel L. Martins, Stergios Christodoulidis +2
Progress in AI-based retinal image analysis has advanced with foundation models, yet evaluating their reliability remains challenging. Performance reported on a single dataset does not capture how models behave under dataset shift, across clinical definitions, or for different patient subgroups. This limitation is particularly critical in medical imaging analysis, where robustness, calibration, and fairness are essential for safe deployment. We introduce FOCUS (Foundation Ophthalmic Cross-Dataset Understanding under Shift), a cross-dataset benchmark for evaluating retinal fundus models that considers vision-only encoder models (VM), vision-language dual-encoder models (VLM), and multimodal large language models (MLLM). FOCUS harmonizes binary diabetic retinopathy, referable diabetic retinopathy, and glaucomatous optic neuropathy tasks across ten public datasets spanning diverse geographies, acquisition conditions, and label protocols. The benchmark evaluates models through a unified analysis layer that measures ranking performance, calibration, subgroup disparities, and image-quality robustness. We present a large-scale evaluation covering 532 base configurations and 228 MLLM…
PDF arXiv
arXiv cs.CLSep 27
MedRouter: Demystifying Knowledge Differences Across Medical LLMs for Routing-Based Reasoning
Lang Cao, Binghang Lu, Yuhao Shen, Yue Guo
Medical question answering spans diverse specialties and modalities, and individual medical large language models (LLMs) exhibit distinct strengths across tasks and domains. This heterogeneity suggests that combining specialists may enable broader coverage of medical questions than relying on any single model. However, existing LLM routing methods primarily seek to balance answer quality and inference cost, leaving open how to exploit differences in specialist competence to improve medical reasoning. In this paper, we introduce MedRouter, an agentic system that uses an embedding-based multi-label router to select and query specialist LLMs, then passes their responses to a generator to produce the final answer. We further propose SCALE (Specialist Competence-Aware Learning), a two-stage training framework that first trains the Router with specialist correctness supervision and then optimizes its selections through reinforcement learning. The second stage uses a Performance Gain Reward (PGR) that measures how specialist information affects the generator's answer correctness relative to answering without that information. Experiments on eight text-based and multimodal medical QA…
PDF arXiv
arXiv cs.LGSep 27
ECG-Scroll: A Long-Horizon, Streaming Benchmark and Agent Environment for Interpretation of Ambulatory Electrocardiograms
Haitao Li, Chenglin Li, Zhengyao Ding, Ziyu Li, Yiheng Mao +1
Multimodal large language models (MLLMs) can now interpret a standard ten-second, twelve-lead electrocardiogram (ECG) with clinically grounded, reward-verified reasoning. Real cardiac monitoring is different. Ambulatory (Holter) and telemetry recordings span hours to days and are read as they stream in, and their clinically decisive findings are paroxysmal, brief episodes buried in an otherwise unremarkable trace. Such a recording cannot be held in one context at diagnostic resolution, and its future has not yet happened, so a reader must work online, deciding what to measure now, committing evidence to memory as it passes, and reporting events as they occur. We recast long-duration ECG interpretation as a long-horizon, online (streaming, causal) sequential decision process and introduce ECG-Scroll. As a benchmark, long ambulatory recordings are streamed to an agent chunk by chunk, and it must localize, quantify, and promptly flag paroxysmal events without access to future signal; because the underlying signal is retained, every answer is checkable against objective ground truth, giving rule-based rather than judge-based rewards, and the streaming formulation adds a metric batch…
PDF arXiv
arXiv cs.CVSep 27
Parameter-Efficient 3D Segmentation of Liver and Liver tumors: Depthwise factorization Scales Better Than Dense Convolution with Spatial Dimensionality
Adham M. Alkhadrawi, Mohammed A. B. Mahmoud
Three-dimensional dense convolutional networks are the strongest performers on volumetric medical image segmentation, but their parameter counts scale poorly: moving a dense k x k convolution to k x k x k multiplies its weights by k. We observe that depthwise separable factorization does not share this penalty. Because the cubic kernel term applies only to the depthwise stage while the pointwise projection, which dominates the parameter count, is unchanged, the same architecture grows by 5 % from 2D to 3D where a dense convolutional U-Net grows by 200 %. We exploit this asymmetry to build a 3D U-Net with 536,990 parameters, 24x fewer than an identical dense 3D U-Net. On MSD Task03 Liver (the Medical Segmentation Decathlon liver task, derived from LiTS), evaluated per case under five-fold cross-validation over all 131 public volumes, the model reaches a tumor Dice of 0.577 (95% CI [0.518, 0.633]) and a liver Dice of 0.947 (95% CI [0.941, 0.952]). Its liver Dice exceeds previously reported performance. Its tumor Dice exceeds their low-resolution configuration (0.4701) by 0.107 and their 2D configuration (0.5394), at approximately one twenty-fourth of the parameters and roughly half…
PDF arXiv

Papers

Topics in this field

image segmentation · volumetric · medical image409 papers42 this week
radiology report generation · medical vision language · vision language models133 papers35 this week
clinical reasoning · question answering · language models210 papers24 this week
patient history · clinical data · health records112 papers22 this week
eeg foundation models · electroencephalography eeg · eeg foundation158 papers20 this week
electrocardiogram ecg · electrocardiogram · ecgs147 papers14 this week

Active contributors · 90 days

Recent papers

All papers in Research

arXiv cs.CVSep 27
Can Protein-Derived Knowledge Improve Pathology Foundation Models?
Di Zhang, Zhangpeng Gong, Jiashuai Liu, Zhi Zeng, Jiusong Ge +8
Molecularly guided pathology foundation models (PFMs) exploit transcriptomic or proteomic information to enrich whole-slide image (WSI) representations, yet effectively leveraging large standalone molecular corpora remains challenging. First, existing molecular foundation models encode protein sequences or single-cell states, not the patient-level bulk expression profiles paired with WSIs. Second, because cross-modal supervision is restricted to paired WSI-omics samples, knowledge from standalone molecular corpora reaches the pathology encoder only indirectly, creating a paired-support bottleneck. To address these challenges, we propose a three-stage framework that decouples proteomic knowledge acquisition from cross-modal transfer, yielding ProSlide, a slide-level hierarchical pathology foundation model. First, to close the modality gap, we pretrain a Proteomic Foundation Encoder (PFE) on 12,695 sample-level bulk protein profiles using virtual profile generation and expression-space multi-view pretraining. Second, we pretrain ProSlide, a patch-region-slide encoder, to predict protein expression from paired WSI-protein samples. Third, to relax the paired-support bottleneck, we…
PDF arXiv
arXiv cs.CVSep 27
ABO-Med: Accelerated Bilevel Optimization for Few-Shot Medical Image Classification
Ruoxuan Shi, Sheng Yang, Zhengxing Su, Xiaoyang Hou, Yating Liu
In recent years, bilevel optimization has been widely used in a variety of machine learning tasks. However, prior bilevel optimization algorithms generally require the computation of second-order information, which limits their practical scalability. Only recently has a first-order paradigm for bilevel optimization been established, attaining near-optimal theoretical guarantees for solving bilevel optimization problems. In this paper, we propose ABO-Med, a scalable instantiation of this paradigm for few-shot learning, by incorporating it into the model-agnostic meta-learning (MAML) framework and tailoring it to medical image classification. We also introduce Medical Adaptive RandomAugment (MedRAug), a modality-aware augmentation strategy designed for medical images. Theoretically, ABO-Med establishes the optimality of MAML-type meta-learning approaches. Empirically, ABO-Med outperforms prior baselines on several public medical datasets, with gains of 1.99% to 18.76%, while MedRAug further improves the average accuracy by 2.20% to 6.34%. Additional cross-domain experiments, augmentation ablation studies, backbone ablation studies, and training efficiency analysis further validate…
PDF arXiv
arXiv cs.CVSep 27
FOCUS: Benchmarking Retinal Model Generalization from Foundation Vision Encoders to Multimodal LLMs
David Restrepo, Chenwei Wu, Luis Filipe Nakayama, Miguel L. Martins, Stergios Christodoulidis +2
Progress in AI-based retinal image analysis has advanced with foundation models, yet evaluating their reliability remains challenging. Performance reported on a single dataset does not capture how models behave under dataset shift, across clinical definitions, or for different patient subgroups. This limitation is particularly critical in medical imaging analysis, where robustness, calibration, and fairness are essential for safe deployment. We introduce FOCUS (Foundation Ophthalmic Cross-Dataset Understanding under Shift), a cross-dataset benchmark for evaluating retinal fundus models that considers vision-only encoder models (VM), vision-language dual-encoder models (VLM), and multimodal large language models (MLLM). FOCUS harmonizes binary diabetic retinopathy, referable diabetic retinopathy, and glaucomatous optic neuropathy tasks across ten public datasets spanning diverse geographies, acquisition conditions, and label protocols. The benchmark evaluates models through a unified analysis layer that measures ranking performance, calibration, subgroup disparities, and image-quality robustness. We present a large-scale evaluation covering 532 base configurations and 228 MLLM…
PDF arXiv
arXiv cs.CLSep 27
MedRouter: Demystifying Knowledge Differences Across Medical LLMs for Routing-Based Reasoning
Lang Cao, Binghang Lu, Yuhao Shen, Yue Guo
Medical question answering spans diverse specialties and modalities, and individual medical large language models (LLMs) exhibit distinct strengths across tasks and domains. This heterogeneity suggests that combining specialists may enable broader coverage of medical questions than relying on any single model. However, existing LLM routing methods primarily seek to balance answer quality and inference cost, leaving open how to exploit differences in specialist competence to improve medical reasoning. In this paper, we introduce MedRouter, an agentic system that uses an embedding-based multi-label router to select and query specialist LLMs, then passes their responses to a generator to produce the final answer. We further propose SCALE (Specialist Competence-Aware Learning), a two-stage training framework that first trains the Router with specialist correctness supervision and then optimizes its selections through reinforcement learning. The second stage uses a Performance Gain Reward (PGR) that measures how specialist information affects the generator's answer correctness relative to answering without that information. Experiments on eight text-based and multimodal medical QA…
PDF arXiv
arXiv cs.LGSep 27
ECG-Scroll: A Long-Horizon, Streaming Benchmark and Agent Environment for Interpretation of Ambulatory Electrocardiograms
Haitao Li, Chenglin Li, Zhengyao Ding, Ziyu Li, Yiheng Mao +1
Multimodal large language models (MLLMs) can now interpret a standard ten-second, twelve-lead electrocardiogram (ECG) with clinically grounded, reward-verified reasoning. Real cardiac monitoring is different. Ambulatory (Holter) and telemetry recordings span hours to days and are read as they stream in, and their clinically decisive findings are paroxysmal, brief episodes buried in an otherwise unremarkable trace. Such a recording cannot be held in one context at diagnostic resolution, and its future has not yet happened, so a reader must work online, deciding what to measure now, committing evidence to memory as it passes, and reporting events as they occur. We recast long-duration ECG interpretation as a long-horizon, online (streaming, causal) sequential decision process and introduce ECG-Scroll. As a benchmark, long ambulatory recordings are streamed to an agent chunk by chunk, and it must localize, quantify, and promptly flag paroxysmal events without access to future signal; because the underlying signal is retained, every answer is checkable against objective ground truth, giving rule-based rather than judge-based rewards, and the streaming formulation adds a metric batch…
PDF arXiv
arXiv cs.CVSep 27
Parameter-Efficient 3D Segmentation of Liver and Liver tumors: Depthwise factorization Scales Better Than Dense Convolution with Spatial Dimensionality
Adham M. Alkhadrawi, Mohammed A. B. Mahmoud
Three-dimensional dense convolutional networks are the strongest performers on volumetric medical image segmentation, but their parameter counts scale poorly: moving a dense k x k convolution to k x k x k multiplies its weights by k. We observe that depthwise separable factorization does not share this penalty. Because the cubic kernel term applies only to the depthwise stage while the pointwise projection, which dominates the parameter count, is unchanged, the same architecture grows by 5 % from 2D to 3D where a dense convolutional U-Net grows by 200 %. We exploit this asymmetry to build a 3D U-Net with 536,990 parameters, 24x fewer than an identical dense 3D U-Net. On MSD Task03 Liver (the Medical Segmentation Decathlon liver task, derived from LiTS), evaluated per case under five-fold cross-validation over all 131 public volumes, the model reaches a tumor Dice of 0.577 (95% CI [0.518, 0.633]) and a liver Dice of 0.947 (95% CI [0.941, 0.952]). Its liver Dice exceeds previously reported performance. Its tumor Dice exceeds their low-resolution configuration (0.4701) by 0.107 and their 2D configuration (0.5394), at approximately one twenty-fourth of the parameters and roughly half…
PDF arXiv

Activity · 12 weeks

Open problems in Image segmentation

Lack of benchmarks for real-world clinical deployment1 paper raises it · 0 address it
Current benchmarks do not reflect how clinicians actually use AI systems, limiting their predictive value for real-world performance.
- Clinician use of language models diverges from how the models are evaluatedraises
Handling gigapixel whole-slide images in segmentation1 paper raises it · 1 addresses it
Existing models operate at patch scale, but clinical supervision and labels are at the whole-slide level, creating a mismatch.
- BigCell: Generating Gigapixel Whole-Slide Imagesraises · addresses
Temporal modeling of multimodal clinical trajectories1 paper raises it · 1 addresses it
Accurate risk prediction requires modeling irregular, heterogeneous clinical data over time, which current approaches often overlook.
- TRACE: Tracking Multimodal Clinical Trajectories for Inpatient Risk Predictionraises · addresses
Resistance to incidental information in clinical reasoning1 paper raises it · 0 address it
Models may misattribute or be distracted by irrelevant information in clinical documentation, leading to errors.
- Large language models are vulnerable to incidental information in clinical documentation and reasoningraises
Learning from deployment experience without retraining1 paper raises it · 1 addresses it
Frozen models cannot adapt to new clinical evidence or updated guidelines without access to weights or additional training.
- Frozen Models, Evolving Expertise: Model-Agnostic Learning from Deployment Experience for Multimodal Medical AIraises · addresses

Named by an AI from these papers' abstracts.

Fields · AI · updated 49m ago

Image segmentation

Found by the topic model: image segmentation, clinical reasoning, eeg foundation models.

157papers this week−25.2% vs last week

All papers · 1.2k With code · 4

Newest Most relevant

989 papers this month