← All publications

conference · 2024 · Published (e-poster)

Enhancing mammography screening sensitivity with AI-assistance: Evidence from a Vietnamese study cohort

Adarsh Bhandary Panambur · Anh Hoang Dinh · L. Chung · L. T. Le · Andreas Maier · Alejandro Rodriguez Ruiz · M. Andrews · B. Schmitt · Siming Bayer

DOI ↗Poster ↗
01 / SUMMARY

What this paper does

Mammographic sensitivity drops in dense breasts, and dense tissue is common in Vietnamese screening populations. This retrospective study compared radiologist reads with and without commercial AI support on a Hanoi Medical University Hospital cohort collected between November 2022 and April 2023, from a base of 1,119 eligible patients. The AI-assisted read reported AUC 75.2% (95% CI 75.0–75.5) and sensitivity 61.5% (95% CI 60.8–61.6), against AUC 73.2% and sensitivity 57.5% for the unassisted read — roughly a four-point absolute sensitivity gain. It is a retrospective e-poster study, not a prospective trial.

02 / METHOD

How

Retrospective analysis of a Hanoi Medical University Hospital screening cohort with radiologist BI-RADS assessment performed with and without commercial AI support, compared by AUC and sensitivity with 95% confidence intervals.

03 / FINDINGS

Reported results

  • AI-assisted read: AUC 75.2% (95% CI 75.0–75.5), sensitivity 61.5% (95% CI 60.8–61.6).
  • Unassisted read: AUC 73.2% (95% CI 72.9–73.2), sensitivity 57.5% (95% CI 57.2–58.1).
  • Study base of 1,119 eligible patients, Hanoi Medical University Hospital, November 2022 to April 2023.
  • The exhibit concludes that AI assistance increased screening sensitivity in this cohort.
04 / WHY IT MATTERS

Relevance

Method papers are usually validated on public datasets. This is the same question asked inside a real screening workflow, in a population where the underlying problem — dense tissue masking lesions — is most acute.

05 / LIMITATION

What it does not show

Retrospective single-centre design published as a conference e-poster rather than a full peer-reviewed article; it is not a prospective randomized comparison.

06 / REFERENCE

Citation

A. Bhandary Panambur, A. Hoang Dinh, L. Chung, L. T. Le, A. Maier, A. Rodriguez Ruiz, M. Andrews, B. Schmitt, S. Bayer. Enhancing mammography screening sensitivity with AI-assistance: Evidence from a Vietnamese study cohort. ECR 2024, EPOS C-16214. doi:10.26044/ecr2024/C-16214

View original abstract

Retrospective reader study of AI-assisted mammography screening in a Vietnamese cohort with a high prevalence of dense breast tissue. Reads with and without commercial AI support were compared on BI-RADS assessment and ROC-based sensitivity at Hanoi Medical University Hospital.

Related work

Connected publications

BE-WISE architecture: pre-contrast, post-contrast and subtraction MRI volumes stacked into a (32, 3, 224, 224) tensor, encoded by a shared Swin-Tiny slice encoder into (32, 768) embeddings, then split into a slice-level lesion predictor producing a per-slice probability curve and a breast-level classifier producing normal, benign or malignant

Bildverarbeitung für die Medizin (BVM) 2026, 10–17

BE-WISE: Breast MRI Evaluation with Weakly-Informed Slice-Level Explanation

Published

Breast MRI models usually need lesions outlined voxel by voxel, which does not scale. BE-WISE asks how far one annotated slice per exam gets you. Pre-contrast, post-contrast and subtraction volumes are stacked and encoded slice by slice with a shared Swin-Tiny transformer. One branch predicts a per-slice lesion probability against a Gaussian target centred on the radiologist's slice; the other pools the same embeddings by attention into a normal / benign / malignant decision. On the multicenter ODELIA dataset this reached test AUC 0.8683 and an ODELIA score of 0.7098, against 0.7968 and 0.5695 for a medical slice transformer baseline.

Explainable AIMedical ImagingDetection / Localization
1 min read → Paper ↗
BE-WISER architecture: pre-contrast, post-contrast and subtraction inputs go through a SwinV2-Tiny slice encoder into slice embeddings, which split into a breast-level classifier module with ABMIL attention, feature aggregation, global head and global logits, and a weak slice-level lesion localization branch with a slice head, slice logits and per-slice class probabilities over normal, benign and malignant

Research Square preprint (in review)

BE-WISER: Class-Aware Weak Slice-Level Supervision for Breast MRI

Preprint · in review

BE-WISE could say which slice contained something; BE-WISER says what that something is. Each slice inside the Gaussian region around the radiologist's marked slice inherits the exam's class — benign or malignant — while the rest are labelled normal, so the slice head becomes three-class instead of binary. On the latest multicenter ODELIA release (926 / 230 / 292 examinations from six institutions) this reached breast-level AUC 0.864 and an ODELIA score of 0.699, against 0.807 and 0.591 for a breast-level baseline. Localization also improved: Hit@1 rose from 0.554 to 0.632 and the mean slice distance fell from 2.838 to 2.505 slices.

Explainable AIDetection / LocalizationMedical Imaging
1 min read → Paper ↗
conference 2026 Vision-Language Models · Foundation Models

European Congress of Radiology (ECR) 2026, scientific exhibit C-15780

Lesion-Aware AI for Mammography: Multi-Dataset Pretraining with ROI-Guided Contrastive Learning and Clinical Image Retrieval

Accepted

Global contrastive pretraining lets a mammography model align on breast density or view rather than on the lesion, because those cues are enough to match an image to its report. This work adds region-of-interest guidance to multi-dataset contrastive pretraining, so localized pathology has to contribute to the representation, and then tests the embeddings on clinical image retrieval — finding visually and diagnostically similar prior cases. Accepted as a scientific exhibit at ECR 2026; quantitative results are not yet publicly available.

Vision-Language ModelsFoundation ModelsMedical Imaging
Details → DOI ↗