Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Related Concept Videos

Mouse Models of Cancer Study

Mouse Models of Cancer Study

Mice have long served as models for studying human biology and pathology because of their phylogenetic and physiological similarity with humans. They are also easy to maintain and breed in the laboratory, and hence, many inbred strains are now available for research. Studies on mice have contributed immeasurably to our understanding of cancer biology.
The development of transgenic, knockout, and knock-in mice has led to an exponential increase in their use as model organisms in research,...

Mouse Models of Cancer Study

Mouse Models of Cancer Study

Mice have long served as models for studying human biology and pathology because of their phylogenetic and physiological similarity with humans. They are also easy to maintain and breed in the laboratory, and hence, many inbred strains are now available for research. Studies on mice have contributed immeasurably to our understanding of cancer biology.
The development of transgenic, knockout, and knock-in mice has led to an exponential increase in their use as model organisms in research,...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Tightly coupled equivariant flow matching for molecular docking with multimodal physical constraints.

Molecular diversity·2026

Same author

Machine learning in the diagnosis of Hirschsprung disease: a systematic review and meta-analysis.

Pediatric surgery international·2026

Same author

Correction of aberrant splicing caused by intronic CAPN3 pathogenic variants using RNA-targeted therapeutic strategies in limb-girdle muscular dystrophy type R1.

Orphanet journal of rare diseases·2026

Same author

Iatrogenic coronal-sagittal coupling driven by a 12.4° rotational mismatch in manual total knee arthroplasty and precise decoupling with robotic assistance: a radiographic retrospective cohort study.

Arthroplasty (London, England)·2026

Same author

Promoting Generalization for Exact Combinatorial Solvers via Adversarial Instance Augmentation.

IEEE transactions on pattern analysis and machine intelligence·2026

Same author

Pretreatment intratumoral mature TLSs in non-clear cell renal cell carcinoma are associated with response to immunotherapy rechallenge.

Journal for immunotherapy of cancer·2026

Search research articles

Related Experiment Video

Updated: Jun 23, 2026

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

Transformer-based Deep Learning Models with Shape Guidance for Predicting Breast Cancer in Mammography Images.

Kengo Takahashi¹, Yuwen Zeng², Zhang Zhang³

¹Department of Radiological Imaging and Informatics, Tohoku University Graduate School of Medicine, 2-1 Seiryo-Machi, Aoba-Ku, Sendai, Miyagi, 980-8575, Japan.

Journal of Imaging Informatics in Medicine

|December 19, 2025

Summary

This summary is machine-generated.

This study shows that guiding Vision Transformer (ViT) models with breast masks at specific layers improves breast cancer detection. Masked ViT models achieved higher accuracy than conventional methods, aiding diagnosis.

Keywords:

Breast cancer Mammography image Shape guidance Swin Transformer Vision Transformer

More Related Videos

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

Related Experiment Videos

Last Updated: Jun 23, 2026

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

Area of Science:

Medical Imaging
Artificial Intelligence
Computer Vision

Background:

Breast cancer research increasingly uses Vision Transformer (ViT) models.
Attention guidance in ViTs focuses on anatomical structures and tumor heterogeneity.
Optimal transformer encoder layer stages for attention guidance remain unclear.

Purpose of the Study:

To evaluate shape-guidance strategies in ViT models for breast cancer detection.
To determine the best encoder layer combinations for attention guidance.
To compare proposed Masked Transformer models against conventional methods.

Main Methods:

Applied breast masks to the attention mechanism of ViT models to emphasize spatial dependencies.
Compared Masked ViT models with ResNet50, ViT, and SwinT V2.
Utilized 2,436 mammography images from the Chinese Mammography Database.
Employed three-fold cross-validation with a 70% training and 30% validation split.

Main Results:

Masked ViT models applying guidance at shallow and deep stages achieved the highest Area Under the Receiver-Operating Characteristic Curve (AUROC).
The best Masked ViT model reached an AUROC of 0.885, outperforming conventional models.
Achieved a sensitivity of 0.876 and specificity of 0.802.

Conclusions:

Shape-guidance strategies, particularly when applied to specific transformer encoder layers, enhance representation learning in ViT models.
Masked ViT models show significant potential as decision-support tools for breast cancer diagnosis.
Optimizing attention guidance in ViTs can improve diagnostic accuracy for medical imaging tasks.