Weight Space Correlation Analysis: Quantifying Feature Utilization in Deep Learning Models

• URL: http://arxiv.org/abs/2512.13144v1
• Date: Mon, 15 Dec 2025 09:52:46 GMT
• Title: Weight Space Correlation Analysis: Quantifying Feature Utilization in Deep Learning Models
• Authors: Chun Kit Wong, Paraskevas Pegios, Nina Weng, Emilie Pi Fogtmann Sejer, Martin Grønnebæk Tolsgaard, Anders Nymark Christensen, Aasa Feragen,
• Abstract summary: We introduce Weight Space Correlation Analysis, an interpretable methodology that quantifies feature utilization.<n>We first validate our method by successfully detecting artificially induced shortcut learning.<n>We then apply it to probe the feature utilization of an SA-SonoNet model trained for Spontaneous Preterm Birth.
• Score: 7.637026905961675
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Deep learning models in medical imaging are susceptible to shortcut learning, relying on confounding metadata (e.g., scanner model) that is often encoded in image embeddings. The crucial question is whether the model actively utilizes this encoded information for its final prediction. We introduce Weight Space Correlation Analysis, an interpretable methodology that quantifies feature utilization by measuring the alignment between the classification heads of a primary clinical task and auxiliary metadata tasks. We first validate our method by successfully detecting artificially induced shortcut learning. We then apply it to probe the feature utilization of an SA-SonoNet model trained for Spontaneous Preterm Birth (sPTB) prediction. Our analysis confirmed that while the embeddings contain substantial metadata, the sPTB classifier's weight vectors were highly correlated with clinically relevant factors (e.g., birth weight) but decoupled from clinically irrelevant acquisition factors (e.g. scanner). Our methodology provides a tool to verify model trustworthiness, demonstrating that, in the absence of induced bias, the clinical model selectively utilizes features related to the genuine clinical signal.

Related papers

• Clinical semantics for lung cancer prediction [1.6744500686720596]
  Existing clinical prediction models often represent patient data using features that ignore semantic relationships between clinical concepts.<n>This study integrates domain-specific semantic information by mapping the SNOMED medical term hierarchy into a low-dimensional hyperbolic space.
  arXiv  Detail & Related papers  (2025-08-20T11:29:47Z)
• AUTOCT: Automating Interpretable Clinical Trial Prediction with LLM Agents [47.640779069547534]
  AutoCT is a novel framework that combines the reasoning capabilities of large language models with the explainability of classical machine learning.<n>We show that AutoCT performs on par with or better than SOTA methods on clinical trial prediction tasks within only a limited number of self-refinement iterations.
  arXiv  Detail & Related papers  (2025-06-04T11:50:55Z)
• Uncertainty-Error correlations in Evidential Deep Learning models for biomedical segmentation [0.0]
  Evidential Deep Learning is applied in the context of biomedical image segmentation. We found that Evidential Deep Learning models with U-Net backbones generally yielded superior correlations between prediction errors and uncertainties. These superior features of EDL models render them well-suited for segmentation tasks that warrant a critical sensitivity in detecting large model errors.
  arXiv  Detail & Related papers  (2024-10-24T06:16:04Z)
• A Comprehensive Dataset and Automated Pipeline for Nailfold Capillary Analysis [24.8934927577986]
  We present a pioneering effort in constructing a comprehensive nailfold capillary dataset-321 images, 219 videos from 68 subjects, with clinic reports and expert annotations. We finetuned three deep learning models with expert annotations as supervised labels and integrated them into a novel end-to-end nailfold capillary analysis pipeline. Experiment results showed that our automated pipeline achieves an average of sub-pixel level precision in measurements and 89.9% accuracy in identifying morphological abnormalities.
  arXiv  Detail & Related papers  (2023-12-10T16:33:41Z)
• An interpretable deep learning method for bearing fault diagnosis [12.069344716912843]
  We utilize a convolutional neural network (CNN) with Gradient-weighted Class Activation Mapping (Grad-CAM) visualizations to form an interpretable Deep Learning (DL) method for classifying bearing faults. During the model evaluation process, the proposed approach retrieves prediction basis samples from the health library according to the similarity of the feature importance.
  arXiv  Detail & Related papers  (2023-08-20T15:22:08Z)
• DeepTechnome: Mitigating Unknown Bias in Deep Learning Based Assessment of CT Images [44.62475518267084]
  We debias deep learning models during training against unknown bias. We use control regions as surrogates that carry information regarding the bias. Applying the proposed method to learn from data exhibiting a strong bias, it near-perfectly recovers the classification performance observed when training with corresponding unbiased data.
  arXiv  Detail & Related papers  (2022-05-26T12:18:48Z)
• Ensembling Handcrafted Features with Deep Features: An Analytical Study for Classification of Routine Colon Cancer Histopathological Nuclei Images [13.858624044986815]
  We have used F1-measure, Precision, Recall, AUC, and Cross-Entropy Loss to analyse the performance of our approaches. We observed from the results that the DL features ensemble bring a marked improvement in the overall performance of the model.
  arXiv  Detail & Related papers  (2022-02-22T06:48:50Z)
• A multi-stage machine learning model on diagnosis of esophageal manometry [50.591267188664666]
  The framework includes deep-learning models at the swallow-level stage and feature-based machine learning models at the study-level stage. This is the first artificial-intelligence-style model to automatically predict CC diagnosis of HRM study from raw multi-swallow data.
  arXiv  Detail & Related papers  (2021-06-25T20:09:23Z)
• Now You See It, Now You Dont: Adversarial Vulnerabilities in Computational Pathology [2.1577322127603407]
  We show that a highly accurate model for classification of tumour patches in pathology images can easily be attacked with minimal perturbations. Our analytical results show that it is possible to generate single-instance white-box attacks on specific input images with high success rate and low perturbation energy. We systematically analyze the relationship between perturbation energy of an adversarial attack, its impact on morphological constructs of clinical significance, their perceptibility by a trained pathologist and saliency maps obtained using deep learning models.
  arXiv  Detail & Related papers  (2021-06-14T14:33:24Z)
• Select-ProtoNet: Learning to Select for Few-Shot Disease Subtype Prediction [55.94378672172967]
  We focus on few-shot disease subtype prediction problem, identifying subgroups of similar patients. We introduce meta learning techniques to develop a new model, which can extract the common experience or knowledge from interrelated clinical tasks. Our new model is built upon a carefully designed meta-learner, called Prototypical Network, that is a simple yet effective meta learning machine for few-shot image classification.
  arXiv  Detail & Related papers  (2020-09-02T02:50:30Z)
• Semi-supervised Medical Image Classification with Relation-driven Self-ensembling Model [71.80319052891817]
  We present a relation-driven semi-supervised framework for medical image classification. It exploits the unlabeled data by encouraging the prediction consistency of given input under perturbations. Our method outperforms many state-of-the-art semi-supervised learning methods on both single-label and multi-label image classification scenarios.
  arXiv  Detail & Related papers  (2020-05-15T06:57:54Z)
• Self-Training with Improved Regularization for Sample-Efficient Chest X-Ray Classification [80.00316465793702]
  We present a deep learning framework that enables robust modeling in challenging scenarios. Our results show that using 85% lesser labeled data, we can build predictive models that match the performance of classifiers trained in a large-scale data setting.
  arXiv  Detail & Related papers  (2020-05-03T02:36:00Z)

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.