Beyond Accuracy: Quantifying Pulmonary Attribution in Anatomy-Guided Chest X-Ray Classification Under Domain Shift

• URL: http://arxiv.org/abs/2608.30467v1
• Date: Mon, 31 Aug 2026 08:54:44 GMT
• Title: Beyond Accuracy: Quantifying Pulmonary Attribution in Anatomy-Guided Chest X-Ray Classification Under Domain Shift
• Abstract summary: Deep-learning models can achieve strong chest X-ray classification performance without establishing whether their predictions rely on pulmonary image content.<n>This study evaluates pulmonary attribution containment as an anatomy-related reliability property distinct from diagnostic performance.
• Score: 0.6213771671016098
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Deep-learning models can achieve strong chest X-ray (CXR) classification performance without establishing whether their predictions predominantly rely on pulmonary image content. This study evaluates pulmonary attribution containment as an anatomy-related reliability property distinct from diagnostic performance. We propose DBCA-SegNet-MGAP, a multi-task anatomy-guided CNN-Transformer framework that combines complementary feature representations through bidirectional cross-backbone attention, predicts a soft lung mask, and incorporates this anatomical prior directly into classification through Mask-Guided Adaptive Global Average Pooling (MGAP). Pulmonary attribution containment is quantified using the Anatomical Local Energy Ratio (ALR) and high-intensity cumulative ALR (cALR@0.9). Experiments were repeated across three training seeds using the COVID-19 Radiography Database for four-class internal testing and a locked Shenzhen-to-Montgomery protocol for zero-shot external tuberculosis testing. On COVID-19, the proposed model achieved a weighted F1 of $0.9615 \pm 0.0015$ and macro ROC-AUC of $0.9906 \pm 0.0007$. In an architecture-matched dual-bridge comparison, replacing conventional GAP with MGAP increased ALR from $0.3878 \pm 0.0098$ to $0.7086 \pm 0.0104$ and cALR@0.9 from $0.5265 \pm 0.0101$ to $0.9905 \pm 0.0018$, while weighted F1 remained essentially unchanged ($0.9618 \pm 0.0015$ vs. $0.9615 \pm 0.0015$). Under locked external transfer to Montgomery, ROC-AUC remained $0.9080 \pm 0.0043$ and pulmonary ALR remained $0.6466 \pm 0.0081$, whereas weighted F1 decreased to $0.7528 \pm 0.0080$ and ECE increased to $0.1683 \pm 0.0055$. These findings show that diagnostic discrimination, calibration, and pulmonary attribution containment are distinct model properties and support their joint evaluation under internal testing and external domain shift.

Related papers

• OmniMed-FL: A Robust Multimodal Federated Learning Framework for Clinical Diagnosis [17.411567668075577]
  We present OmniMed-FL, a controlled systems study of multimodal learning for five-class clinical condition classification.<n>Our proxy corpus pairs 3,000 public chest radiographs with 3,000 class-FOLD synthetic notes, matched by class, not by patient.<n>As all notes are synthetic and pairing is not patient-level, these are descriptive proxy comparisons, not estimates of diagnostic performance or deployment readiness.
  arXiv  Detail & Related papers  (2026-09-09T15:56:26Z)
• Cascade Classification of Dermoscopic Images of Skin Neoplasms with Controllable Sensitivity and External Clinical Validation [36.94429692322632]
  Four architectures were compared in three schemes: binary (malignant/benign), single-stage four-class (benign, MEL, SCC, BCC), and a two-stage cascade (binary triage, then three-class differentiation MEL/ SCC/BCC)<n>All models used ImageNet-pretrained weights and a single augmentation protocol on aggregated open ISIC Archive data.<n>On ISIC MILK10k, direct 11-class classification yields mean-class sensitivity 0.525.
  arXiv  Detail & Related papers  (2026-06-11T09:55:57Z)
• CT-to-X-ray Distillation Under Tiny Paired Cohorts: An Evidence-Bounded Reproducible Pilot Study [11.316541559874864]
  This study is not a validated CT-to-X-ray architecture, but a reproducible and evidence-bounded pilot protocol that makes the exact task definition, failure modes, ranking instability, and the minimum requirements for future credible CT-to-X-ray transfer claims explicit.
  arXiv  Detail & Related papers  (2026-03-31T02:25:33Z)
• SA-CycleGAN-2.5D: Self-Attention CycleGAN with Tri-Planar Context for Multi-Site MRI Harmonization [3.5109108807229403]
  We propose SA-CycleGAN-2.5D, a domain adaptation framework motivated by the $HH$-divergence bound of Ben-David et al.<n>By bridging 2D efficiency and 3D consistency, our framework yields voxel-level harmonized images that preserve tumor pathophysiology.
  arXiv  Detail & Related papers  (2026-03-17T23:49:46Z)
• Pillar-0: A New Frontier for Radiology Foundation Models [41.640120966890954]
  We introduce Pillar-0, a radiology foundation model pretrained on 42,990 abdomen-pelvis CTs, 86,411 chest CTs, 14,348 head CTs, and 11,543 breast MRIs.<n>Pillar-0 achieves mean AUROCs of 86.4, 88.0, 90.1, and 82.9, outperforming MedGemma (Google), MedImageInsight (Microsoft), Lingshu (Alibaba), and Merlin (Stanford) by 7.8-15.8 AUROC points and ranking best in 87.2% (319/366) tasks.
  arXiv  Detail & Related papers  (2025-11-21T21:50:34Z)
• Localizing Anomalies via Multiscale Score Matching Analysis [13.898576482792173]
  This paper introduces Spatial-MSMA, a novel unsupervised method for anomaly localization in brain MRIs. We employ a flexible normalizing flow model conditioned on patch positions and global image features to estimate patch-wise anomaly scores. The method is evaluated on a dataset of 1,650 T1- and T2-weighted brain MRIs from typically developing children.
  arXiv  Detail & Related papers  (2024-06-28T17:57:12Z)
• CIMIL-CRC: a clinically-informed multiple instance learning framework for patient-level colorectal cancer molecular subtypes classification from H\&E stained images [42.771819949806655]
  We introduce CIMIL-CRC', a framework that solves the MSI/MSS MIL problem by efficiently combining a pre-trained feature extraction model with principal component analysis (PCA) to aggregate information from all patches. We assessed our CIMIL-CRC method using the average area under the curve (AUC) from a 5-fold cross-validation experimental setup for model development on the TCGA-CRC-DX cohort.
  arXiv  Detail & Related papers  (2024-01-29T12:56:11Z)
• Fibroglandular Tissue Segmentation in Breast MRI using Vision Transformers -- A multi-institutional evaluation [0.06499822663743766]
  We developed and evaluated a transformer-based neural network for breast segmentation (TraBS) and compared its performance to the well established convolutional network nnUNet. TraBS and nnUNet were trained tested on 200 internal and 40 external breast MRI examinations using manual segmentation generated by experienced human readers. Our study demonstrates that neural networks improve the quality of fibroglandular tissue segmentation compared to transformer-based models like nnUNet.
  arXiv  Detail & Related papers  (2023-04-18T13:12:35Z)
• Attention-based Saliency Maps Improve Interpretability of Pneumothorax Classification [52.77024349608834]
  To investigate chest radiograph (CXR) classification performance of vision transformers (ViT) and interpretability of attention-based saliency. ViTs were fine-tuned for lung disease classification using four public data sets: CheXpert, Chest X-Ray 14, MIMIC CXR, and VinBigData. ViTs had comparable CXR classification AUCs compared with state-of-the-art CNNs.
  arXiv  Detail & Related papers  (2023-03-03T12:05:41Z)
• Controlling False Positive/Negative Rates for Deep-Learning-Based Prostate Cancer Detection on Multiparametric MR images [58.85481248101611]
  We propose a novel PCa detection network that incorporates a lesion-level cost-sensitive loss and an additional slice-level loss based on a lesion-to-slice mapping function. Our experiments based on 290 clinical patients concludes that 1) The lesion-level FNR was effectively reduced from 0.19 to 0.10 and the lesion-level FPR was reduced from 1.03 to 0.66 by changing the lesion-level cost.
  arXiv  Detail & Related papers  (2021-06-04T09:51:27Z)
• Developing Univariate Neurodegeneration Biomarkers with Low-Rank and Sparse Subspace Decomposition [37.62559407222313]
  Cognitive decline due to Alzheimer's disease (AD) is closely associated with brain structure alterations captured by structural magnetic resonance imaging (sMRI) We propose a novel low-rank and sparse subspace decomposition method capable of stably quantifying the morphological changes induced by AD dementia (ADD) Our experimental results outperform traditional hippocampal volume measures and suggest the application of UMI as a potential UNB.
  arXiv  Detail & Related papers  (2020-10-26T23:42:43Z)
• Lymph Node Gross Tumor Volume Detection in Oncology Imaging via Relationship Learning Using Graph Neural Network [37.88742249495087]
  We propose a unified LN appearance and inter-LN relationship learning framework to detect the true GTV$_LN$. The proposed method significantly improves over the state-of-the-art (SOTA) LN classification method by $5.5%$ and $13.1%$ in F1 score.
  arXiv  Detail & Related papers  (2020-08-29T16:59:23Z)
• Automated Quantification of CT Patterns Associated with COVID-19 from Chest CT [48.785596536318884]
  The proposed method takes as input a non-contrasted chest CT and segments the lesions, lungs, and lobes in three dimensions. The method outputs two combined measures of the severity of lung and lobe involvement, quantifying both the extent of COVID-19 abnormalities and presence of high opacities. Evaluation of the algorithm is reported on CTs of 200 participants (100 COVID-19 confirmed patients and 100 healthy controls) from institutions from Canada, Europe and the United States.
  arXiv  Detail & Related papers  (2020-04-02T21:49:14Z)

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.