Benchmarking Foundation Models and Parameter-Efficient Fine-Tuning for Prognosis Prediction in Medical Imaging

• URL: http://arxiv.org/abs/2506.18434v2
• Date: Wed, 05 Nov 2025 09:33:35 GMT
• Title: Benchmarking Foundation Models and Parameter-Efficient Fine-Tuning for Prognosis Prediction in Medical Imaging
• Authors: Filippo Ruffini, Elena Mulero Ayllon, Linlin Shen, Paolo Soda, Valerio Guarrasi,
• Abstract summary: This study introduces the first structured benchmark to assess the robustness and efficiency of transfer learning strategies for Foundation Models.<n>Four publicly available COVID-19 chest X-ray datasets were used, covering mortality, severity, and admission.<n>CNNs pretrained on ImageNet and FMs pretrained on general or biomedical datasets were adapted using full finetuning, linear probing, and parameter-efficient methods.
• Score: 40.35825564674249
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Despite the significant potential of Foundation Models (FMs) in medical imaging, their application to prognosis prediction remains challenging due to data scarcity, class imbalance, and task complexity, which limit their clinical adoption. This study introduces the first structured benchmark to assess the robustness and efficiency of transfer learning strategies for FMs compared with convolutional neural networks (CNNs) in predicting COVID-19 patient outcomes from chest X-rays. The goal is to systematically compare finetuning strategies, both classical and parameter efficient, under realistic clinical constraints related to data scarcity and class imbalance, offering empirical guidance for AI deployment in clinical workflows. Four publicly available COVID-19 chest X-ray datasets were used, covering mortality, severity, and ICU admission, with varying sample sizes and class imbalances. CNNs pretrained on ImageNet and FMs pretrained on general or biomedical datasets were adapted using full finetuning, linear probing, and parameter-efficient methods. Models were evaluated under full data and few shot regimes using the Matthews Correlation Coefficient (MCC) and Precision Recall AUC (PR-AUC), with cross validation and class weighted losses. CNNs with full fine-tuning performed robustly on small, imbalanced datasets, while FMs with Parameter-Efficient Fine-Tuning (PEFT), particularly LoRA and BitFit, achieved competitive results on larger datasets. Severe class imbalance degraded PEFT performance, whereas balanced data mitigated this effect. In few-shot settings, FMs showed limited generalization, with linear probing yielding the most stable results. No single fine-tuning strategy proved universally optimal: CNNs remain dependable for low-resource scenarios, whereas FMs benefit from parameter-efficient methods when data are sufficient.

Related papers

• A Federated and Parameter-Efficient Framework for Large Language Model Training in Medicine [59.78991974851707]
  Large language models (LLMs) have demonstrated strong performance on medical benchmarks, including question answering and diagnosis.<n>Most medical LLMs are trained on data from a single institution, which faces limitations in generalizability and safety in heterogeneous systems.<n>We introduce the model-agnostic and parameter-efficient federated learning framework for adapting LLMs to medical applications.
  arXiv  Detail & Related papers  (2026-01-29T18:48:21Z)
• Pretraining Transformer-Based Models on Diffusion-Generated Synthetic Graphs for Alzheimer's Disease Prediction [0.0]
  We propose a Transformer-based diagnostic framework that combines synthetic data generation with graph representation learning and transfer learning.<n>A class-conditional denoising diffusion probabilistic model (DDPM) is trained on the real-world NACC dataset to generate a large synthetic cohort.<n> Modality-specific Graph Transformer encoders are first pretrained on this synthetic data to learn robust, class-discriminative representations.
  arXiv  Detail & Related papers  (2025-11-24T19:34:53Z)
• PathBench: A comprehensive comparison benchmark for pathology foundation models towards precision oncology [33.51485504161335]
  We present PathBench, the first comprehensive benchmark for pathology foundation models (PFMs)<n>Our framework incorporates large-scale data, enabling objective comparison of PFMs.<n>We have collected 15,888 WSIs from 8,549 patients across 10 hospitals, encompassing over 64 diagnosis and prognosis tasks.
  arXiv  Detail & Related papers  (2025-05-26T16:42:22Z)
• Enhancing Treatment Effect Estimation via Active Learning: A Counterfactual Covering Perspective [61.284843894545475]
  Complex algorithms for treatment effect estimation are ineffective when handling insufficiently labeled training sets.<n>We propose FCCM, which transforms the optimization objective into the textitFactual and textitCounterfactual Coverage Maximization to ensure effective radius reduction during data acquisition.<n> benchmarking FCCM against other baselines demonstrates its superiority across both fully synthetic and semi-synthetic datasets.
  arXiv  Detail & Related papers  (2025-05-08T13:42:00Z)
• Continually Evolved Multimodal Foundation Models for Cancer Prognosis [50.43145292874533]
  Cancer prognosis is a critical task that involves predicting patient outcomes and survival rates.<n>Previous studies have integrated diverse data modalities, such as clinical notes, medical images, and genomic data, leveraging their complementary information.<n>Existing approaches face two major limitations. First, they struggle to incorporate newly arrived data with varying distributions into training, such as patient records from different hospitals.<n>Second, most multimodal integration methods rely on simplistic concatenation or task-specific pipelines, which fail to capture the complex interdependencies across modalities.
  arXiv  Detail & Related papers  (2025-01-30T06:49:57Z)
• Development and Comparative Analysis of Machine Learning Models for Hypoxemia Severity Triage in CBRNE Emergency Scenarios Using Physiological and Demographic Data from Medical-Grade Devices [0.0]
  Gradient Boosting Models (GBMs) outperformed sequential models in terms of training speed, interpretability, and reliability. A 5-minute prediction window was chosen for timely intervention, with minute-levels standardizing the data. This study highlights ML's potential to improve triage and reduce alarm fatigue.
  arXiv  Detail & Related papers  (2024-10-30T23:24:28Z)
• How Deep is your Guess? A Fresh Perspective on Deep Learning for Medical Time-Series Imputation [6.547981908229007]
  We show how architectural and framework biases combine to influence model performance.<n>Experiments show imputation performance variations of up to 20% based on preprocessing and implementation choices.<n>We identify critical gaps between current deep imputation methods and medical requirements.
  arXiv  Detail & Related papers  (2024-07-11T12:33:28Z)
• Comparison of fine-tuning strategies for transfer learning in medical image classification [2.271776292902496]
  Despite availability of advanced pre-trained models, their direct application to medical imaging often falls short due to unique characteristics of medical data. This study provides a comprehensive analysis on the performance of various fine-tuning methods applied to pre-trained models across a spectrum of medical imaging domains.
  arXiv  Detail & Related papers  (2024-06-14T14:00:02Z)
• Rethinking model prototyping through the MedMNIST+ dataset collection [0.11999555634662634]
  This work introduces a comprehensive benchmark for the MedMNIST+ dataset collection.<n>We reassess commonly used Convolutional Neural Networks (CNNs) and Vision Transformer (ViT) architectures across distinct medical datasets.<n>Our findings suggest that computationally efficient training schemes and modern foundation models offer viable alternatives to costly end-to-end training.
  arXiv  Detail & Related papers  (2024-04-24T10:19:25Z)
• DrFuse: Learning Disentangled Representation for Clinical Multi-Modal Fusion with Missing Modality and Modal Inconsistency [18.291267748113142]
  We propose DrFuse to achieve effective clinical multi-modal fusion. We address the missing modality issue by disentangling the features shared across modalities and those unique within each modality. We validate the proposed method using real-world large-scale datasets, MIMIC-IV and MIMIC-CXR.
  arXiv  Detail & Related papers  (2024-03-10T12:41:34Z)
• Few-shot learning for COVID-19 Chest X-Ray Classification with Imbalanced Data: An Inter vs. Intra Domain Study [49.5374512525016]
  Medical image datasets are essential for training models used in computer-aided diagnosis, treatment planning, and medical research. Some challenges are associated with these datasets, including variability in data distribution, data scarcity, and transfer learning issues when using models pre-trained from generic images. We propose a methodology based on Siamese neural networks in which a series of techniques are integrated to mitigate the effects of data scarcity and distribution imbalance.
  arXiv  Detail & Related papers  (2024-01-18T16:59:27Z)
• MCRAGE: Synthetic Healthcare Data for Fairness [3.0089659534785853]
  We propose Minority Class Rebalancing through Augmentation by Generative modeling (MCRAGE) to augment imbalanced datasets. MCRAGE involves training a Denoising Diffusion Probabilistic Model (CDDPM) capable of generating high-quality synthetic EHR samples from underrepresented classes. We use this synthetic data to augment the existing imbalanced dataset, resulting in a more balanced distribution across all classes.
  arXiv  Detail & Related papers  (2023-10-27T19:02:22Z)
• The effect of data augmentation and 3D-CNN depth on Alzheimer's Disease detection [51.697248252191265]
  This work summarizes and strictly observes best practices regarding data handling, experimental design, and model evaluation. We focus on Alzheimer's Disease (AD) detection, which serves as a paradigmatic example of challenging problem in healthcare. Within this framework, we train predictive 15 models, considering three different data augmentation strategies and five distinct 3D CNN architectures.
  arXiv  Detail & Related papers  (2023-09-13T10:40:41Z)
• Improving Clinical Decision Support through Interpretable Machine Learning and Error Handling in Electronic Health Records [6.594072648536156]
  Trust-MAPS translates clinical domain knowledge into high-dimensional, mixed-integer programming models.<n>Trust-scores emerge as clinically meaningful features that not only boost predictive performance for clinical decision support tasks, but also lend interpretability to ML models.
  arXiv  Detail & Related papers  (2023-08-21T15:14:49Z)
• Density-Aware Personalized Training for Risk Prediction in Imbalanced Medical Data [89.79617468457393]
  Training models with imbalance rate (class density discrepancy) may lead to suboptimal prediction. We propose a framework for training models for this imbalance issue. We demonstrate our model's improved performance in real-world medical datasets.
  arXiv  Detail & Related papers  (2022-07-23T00:39:53Z)
• Learning brain MRI quality control: a multi-factorial generalization problem [0.0]
  This work aimed at evaluating the performances of the MRIQC pipeline on various large-scale datasets. We focused our analysis on the MRIQC preprocessing steps and tested the pipeline with and without them. We concluded that a model trained with data from a heterogeneous population, such as the CATI dataset, provides the best scores on unseen data.
  arXiv  Detail & Related papers  (2022-05-31T15:46:44Z)
• Multi-task fusion for improving mammography screening data classification [3.7683182861690843]
  We propose a pipeline approach, where we first train a set of individual, task-specific models. We then investigate the fusion thereof, which is in contrast to the standard model ensembling strategy. Our fusion approaches improve AUC scores significantly by up to 0.04 compared to standard model ensembling.
  arXiv  Detail & Related papers  (2021-12-01T13:56:27Z)
• Clinical Outcome Prediction from Admission Notes using Self-Supervised Knowledge Integration [55.88616573143478]
  Outcome prediction from clinical text can prevent doctors from overlooking possible risks. Diagnoses at discharge, procedures performed, in-hospital mortality and length-of-stay prediction are four common outcome prediction targets. We propose clinical outcome pre-training to integrate knowledge about patient outcomes from multiple public sources.
  arXiv  Detail & Related papers  (2021-02-08T10:26:44Z)
• Adversarial Sample Enhanced Domain Adaptation: A Case Study on Predictive Modeling with Electronic Health Records [57.75125067744978]
  We propose a data augmentation method to facilitate domain adaptation. adversarially generated samples are used during domain adaptation. Results confirm the effectiveness of our method and the generality on different tasks.
  arXiv  Detail & Related papers  (2021-01-13T03:20:20Z)
• An Optimal Control Approach to Learning in SIDARTHE Epidemic model [67.22168759751541]
  We propose a general approach for learning time-variant parameters of dynamic compartmental models from epidemic data. We forecast the epidemic evolution in Italy and France.
  arXiv  Detail & Related papers  (2020-10-28T10:58:59Z)
• UNITE: Uncertainty-based Health Risk Prediction Leveraging Multi-sourced Data [81.00385374948125]
  We present UNcertaInTy-based hEalth risk prediction (UNITE) model. UNITE provides accurate disease risk prediction and uncertainty estimation leveraging multi-sourced health data. We evaluate UNITE on real-world disease risk prediction tasks: nonalcoholic fatty liver disease (NASH) and Alzheimer's disease (AD) UNITE achieves up to 0.841 in F1 score for AD detection, up to 0.609 in PR-AUC for NASH detection, and outperforms various state-of-the-art baselines by up to $19%$ over the best baseline.
  arXiv  Detail & Related papers  (2020-10-22T02:28:11Z)
• Predicting Clinical Diagnosis from Patients Electronic Health Records Using BERT-based Neural Networks [62.9447303059342]
  We show the importance of this problem in medical community. We present a modification of Bidirectional Representations from Transformers (BERT) model for classification sequence. We use a large-scale Russian EHR dataset consisting of about 4 million unique patient visits.
  arXiv  Detail & Related papers  (2020-07-15T09:22:55Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.