Breast Ultrasound Classification Under Cross-Dataset Shift: A Multi-Seed Benchmark of Convolutional Networks, a Vision Transformer, and a Self-Supervised Foundation Model

Deep-learning classifiers for breast ultrasound are usually evaluated on images from their training institution, and self-supervised foundation models are increasingly proposed as more transferable than supervised networks; whether ranking ability, decision thresholds, and probability calibration survive a change in institution has rarely been tested under leakage-controlled conditions. We fine-tuned ResNet-50, EfficientNet-B0, a supervised Vision Transformer (ViT-B/16), and self-supervised DINOv2 under one shared protocol on the official BUS-BRA five-fold partitions, each fold with three seeds, and on three case-grouped hold-out splits, and evaluated every trained model unchanged on the Breast Ultrasound Images (BUSI) dataset. The supervised models largely retained their discrimination externally. ResNet-50 had the highest mean area under the receiver operating characteristic curve under the official folds, but no criterion separated it from ViT-B/16, whose small hold-out advantage at the decision threshold did not recur. In every official-fold run, both models labeled a larger share of BUSI than of BUS-BRA images malignant, although prevalence was the same—a shift that temperature scaling cannot correct. Top-label calibration error also understated malignant-class miscalibration. Full-frame DINOv2 did not fit its training data in any run, although exploratory lesion cropping produced good discrimination. External evaluations should report operating-point transport and class-wise calibration alongside discrimination.

Authors

Institutions

Publication Details

Journal
Algorithms
Published
2026-11-18
DOI
https://doi.org/10.3390/a19111011
Primary Topic
AI in cancer detection
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Breast Ultrasound Classification Under Cross-Dataset Shift: A Multi-Seed Benchmark of Convolutional Networks, a Vision Transformer, and a Self-Supervised Foundation Model

Ervin Ramollari, Narasimha Rao Vajjhala, Rajesh Kumar, Khalid Hussain et al.
3 links
Algorithms
AI in cancer detection
article

Breast Ultrasound Classification Under Cross-Dataset Shift: A Multi-Seed Benchmark of Convolutional Networks, a Vision Transformer, and a Self-Supervised Foundation Model

Ervin Ramollari, Narasimha Rao Vajjhala, Rajesh Kumar, Khalid Hussain, Ranjeeta Mahraj, Komal Ramchandani
article en

Abstract

Deep-learning classifiers for breast ultrasound are usually evaluated on images from their training institution, and self-supervised foundation models are increasingly proposed as more transferable than supervised networks; whether ranking ability, decision thresholds, and probability calibration survive a change in institution has rarely been tested under leakage-controlled conditions. We fine-tuned ResNet-50, EfficientNet-B0, a supervised Vision Transformer (ViT-B/16), and self-supervised DINOv2 under one shared protocol on the official BUS-BRA five-fold partitions, each fold with three seeds, and on three case-grouped hold-out splits, and evaluated every trained model unchanged on the Breast Ultrasound Images (BUSI) dataset. The supervised models largely retained their discrimination externally. ResNet-50 had the highest mean area under the receiver operating characteristic curve under the official folds, but no criterion separated it from ViT-B/16, whose small hold-out advantage at the decision threshold did not recur. In every official-fold run, both models labeled a larger share of BUSI than of BUS-BRA images malignant, although prevalence was the same—a shift that temperature scaling cannot correct. Top-label calibration error also understated malignant-class miscalibration. Full-frame DINOv2 did not fit its training data in any run, although exploratory lesion cropping produced good discrimination. External evaluations should report operating-point transport and class-wise calibration alongside discrimination.

AlgorithmsVol. 19(11)
Montclair State University (US), American University in Bulgaria (BG), Chemnitz University of Technology (DE), University of New York Tirana (AL)
Good health and well-being
AI in cancer detection
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon