Anatomical Grounding and Leakage-Aware Multimodal Contrastive Learning for Alzheimer's Disease Classification from Structural MRI

• URL: http://arxiv.org/abs/2609.15888v1
• Date: Mon, 14 Sep 2026 17:13:12 GMT
• Title: Anatomical Grounding and Leakage-Aware Multimodal Contrastive Learning for Alzheimer's Disease Classification from Structural MRI
• Abstract summary: Deep networks trained on structural MRI for Alzheimer's disease often reach reasonable accuracy while attending to anatomically irrelevant regions.<n>We use FastSurfer segmentations as an anatomical reference.<n>We adapt a CLIP-style image and organize ADNIMERGE variables along a label-leakage spectrum.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Deep networks trained on structural MRI for Alzheimer's disease (AD) staging often reach reasonable accuracy while attending to anatomically irrelevant regions, and multimodal models that add clinical tables frequently rely on variables that were used to assign the diagnostic label in the first place. We study both issues with a deliberately lightweight slice-based encoder (ResNet18 with a one-layer Transformer over slices) on 1,075 baseline T1-weighted scans from ADNI-1. First, we use FastSurfer segmentations as an anatomical reference: YOLOv8 models trained on segmentation-derived labels localize Alzheimer-relevant structures with mAP_50 above 0.96, and a Grad-CAM comparison shows that the image-only classifier frequently attends to the skull, orbits and background. Second, we adapt a CLIP-style image - tabular contrastive framework and organize ADNIMERGE variables along a label-leakage spectrum. Fusion with cognitive scores yields 87.3% three-way accuracy, which we treat as a leakage-driven upper bound rather than an imaging result; fusion with regional volumes yields 73.0%. We observe that the choice of contrastive target changes what the image encoder learns: on MCI vs. CN, the image-only head reaches 52.4% when the encoder is aligned to cognitive scores and 73.8\% when aligned to volumes, although no tabular input is used at inference. Third, restricting the input to a per-subject crop of the medial temporal lobe raises image-only three-way accuracy from 58.7% to 65.1%. All results come from single runs on a small balanced test set, and we report confidence intervals and the protocol differences that prevent direct comparison with published numbers.

Related papers

• Bridging Vision Foundation Model Priors with CLIP for Spatial-aware Few-shot Anomaly Detection in Medical Images [48.646448077284276]
  We propose Spatial-FAD, a framework for effective few-shot medical anomaly detection.<n>It improves lesion localization by combining VFM spatial priors with CLIP semantics.<n>Our method achieves an average improvement of over 11.4% in Dice score and 1.8% in AUC.
  arXiv  Detail & Related papers  (2026-09-11T05:21:01Z)
• DINO-Med: A Unified Patch-Based Adaptation Framework for Multi-Modal Medical Image Analysis Applied to Liver Fibrosis Staging [13.33205970973723]
  We present a unified, patch-based framework that processes raw multimodal imaging through training-free registration, automated localization, and mask-filtered patch extraction.<n>We evaluate four patch-level feature representations: handcrafted Radiomics features, learned ResNet features, pre-trained foundation model SAM-Med2D features, frozen cirrhosis DINOv3 features.
  arXiv  Detail & Related papers  (2026-09-10T11:16:26Z)
• Self-supervision drives representational convergence in medical foundation models more than clinical supervision [2.8356496885865017]
  We present a controlled dissection across 18 image and 7 text encoders, all open-weight and run locally, spanning 7M to 27B parameters and five imaging modalities.<n>We train encoders that vary only the objective under fixed data, architecture, and scale, and reproduce the effect in a synthetic model.<n> Convergence is modest but above a random floor, driven by the self-supervised objective, not clinical supervision.
  arXiv  Detail & Related papers  (2026-07-22T15:25:05Z)
• Cross-modal linkage risk in clinical vision-language models [1.828619151598757]
  Vision-language models (VLMs) trained on paired chest radiographs and radiology reports learn a shared embedding space that can preserve instance-level image-report correspondence.<n>This poses a privacy risk in settings where radiographs and reports are deliberately kept separate after acquisition.<n>We formalized this as image-to-report retrieval and used public paired cohorts to audit the risk rather than as the privacy scenario.
  arXiv  Detail & Related papers  (2026-06-01T14:01:46Z)
• Medical thinking with multiple images [44.04557445622649]
  We introduce MedThinkVQA, an expert-annotated benchmark for thinking with multiple images.<n>The dataset contains 8,067 cases, including 720 test cases, with an average of 6.62 images per case.<n>On the test set, the best closed-source models, Claude-4.6-Opus, Gemini-3-Pro, and GPT-5.2-xhigh, reach only 57.2%, 55.3%, and 54.9% accuracy.
  arXiv  Detail & Related papers  (2026-04-14T18:51:07Z)
• Brain Tumor Classification on MRI in Light of Molecular Markers [56.99710477905796]
  Co-deletion of the 1p/19q gene is associated with clinical outcomes in low-grade gliomas.<n>This study aims to utilize a specially MRI-based convolutional neural network for brain cancer detection.
  arXiv  Detail & Related papers  (2024-09-29T07:04:26Z)
• Acute ischemic stroke lesion segmentation in non-contrast CT images using 3D convolutional neural networks [0.0]
  We propose an automatic algorithm aimed at volumetric segmentation of acute ischemic stroke lesion in non-contrast computed tomography brain 3D images. Our deep-learning approach is based on the popular 3D U-Net convolutional neural network architecture.
  arXiv  Detail & Related papers  (2023-01-17T10:39:39Z)
• Lightweight 3D Convolutional Neural Network for Schizophrenia diagnosis using MRI Images and Ensemble Bagging Classifier [1.487444917213389]
  This paper proposed a lightweight 3D convolutional neural network (CNN) based framework for schizophrenia diagnosis using MRI images. The model achieves the highest accuracy 92.22%, sensitivity 94.44%, specificity 90%, precision 90.43%, recall 94.44%, F1-score 92.39% and G-mean 92.19% as compared to the current state-of-the-art techniques.
  arXiv  Detail & Related papers  (2022-11-05T10:27:37Z)
• Corneal endothelium assessment in specular microscopy images with Fuchs' dystrophy via deep regression of signed distance maps [48.498376125522114]
  This paper proposes a UNet-based segmentation approach that requires minimal post-processing. It achieves reliable CE morphometric assessment and guttae identification across all degrees of Fuchs' dystrophy.
  arXiv  Detail & Related papers  (2022-10-13T15:34:20Z)
• Osteoporosis Prescreening using Panoramic Radiographs through a Deep Convolutional Neural Network with Attention Mechanism [65.70943212672023]
  Deep convolutional neural network (CNN) with an attention module can detect osteoporosis on panoramic radiographs. dataset of 70 panoramic radiographs (PRs) from 70 different subjects of age between 49 to 60 was used.
  arXiv  Detail & Related papers  (2021-10-19T00:03:57Z)
• Automatic size and pose homogenization with spatial transformer network to improve and accelerate pediatric segmentation [51.916106055115755]
  We propose a new CNN architecture that is pose and scale invariant thanks to the use of Spatial Transformer Network (STN) Our architecture is composed of three sequential modules that are estimated together during training. We test the proposed method in kidney and renal tumor segmentation on abdominal pediatric CT scanners.
  arXiv  Detail & Related papers  (2021-07-06T14:50:03Z)

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.