For patients with thyroid nodules, the ability to detect and diagnose a malignant nodule is the key to creating an appropriate treatment plan. However, assessments of ultrasound images do not accurately represent malignancy, and often require a biopsy to confirm the diagnosis. Deep learning techniques can classify thyroid nodules from ultrasound images, but current methods depend on manually annotated nodule segmentations. Furthermore, the heterogeneity in the level of magnification across ultrasound images presents a significant obstacle to existing methods. We developed a multi-scale, attention-based multiple-instance learning model which fuses both global and local features of different ultrasound frames to achieve patient-level malignancy classification. Our model demonstrates improved performance with an AUROC of 0.785 (p<0.05) and AUPRC of 0.539, significantly surpassing the baseline model trained on clinical features with an AUROC of 0.667 and AUPRC of 0.444. Improved classification performance better triages the need for biopsy.

Download full-text PDF

Source
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC10785838PMC

Publication Analysis

Top Keywords

ultrasound images
12
thyroid nodules
8
ultrasound
5
patient-level thyroid
4
thyroid cancer
4
cancer classification
4
classification attention
4
attention multiple
4
multiple instance
4
instance learning
4

Similar Publications

This systematic review and meta-analysis aimed to assess the accuracy and success rate of ultrasound in determining fetal sex. A search was conducted on Medline, Cochrane Library, and EMBASE databases, and the reference lists of selected studies were also reviewed. Meta-analyses were performed using Revman 5.

View Article and Find Full Text PDF

Purpose Of Review: This review aims to explore the complex interplay between atrial functional mitral regurgitation (AFMR), atrial fibrillation (AF), and heart failure with preserved ejection fraction (HFpEF). The goal is to define these conditions, examine their underlying mechanisms, and discuss treatment perspectives, particularly addressing diagnostic challenges.

Recent Findings: Recent research highlights the rising prevalence of AFMR, now accounting for nearly one-third of significant mitral regurgitation cases.

View Article and Find Full Text PDF

Background: Recent technological advances have introduced novel methods for measuring body composition, each with unique benefits and limitations. The choice of method often depends on the trade-offs between accuracy, cost, participant burden, and the ability to measure specific body composition compartments.

Objective: To review the considerations of cost, accuracy, portability, and participant burden in reference and emerging body composition assessment methods, and to evaluate their clinical applicability.

View Article and Find Full Text PDF

Are there atypical sites of IgG4 related disease in head and neck region? Personal experience and literature review.

Eur Arch Otorhinolaryngol

January 2025

Department of Otolaryngology and Head and Neck Surgery, IRCSS AOU San Martino, University of Genoa, Largo Rosanna Benzi 10, 16132, Genoa, Italy.

Purpose: Immunoglobulin G4-related disease (IgG4-RD) is a complex systemic fibroinflammatory condition with different clinical manifestations affecting multiple organ systems. Despite its rarity, the disease presents diagnostic and therapeutic challenges due to its mimicry of malignancies and other immune-mediated disorders. The 2019 American College of Rheumatology/European League Against Rheumatism Classification Criteria for IgG4-Related Disease is the current state of art to confirm the diagnosis of IgG4-RD even in the absence of histological analysis.

View Article and Find Full Text PDF

Objectives: The aim is to assess the feasibility and accuracy of a novel quantitative ultrasound (US) method based on global speed-of-sound (g-SoS) measurement using conventional US machines, for breast density assessment in comparison to mammographic ACR (m-ACR) categories.

Materials And Methods: In a prospective study, g-SoS was assessed in the upper-outer breast quadrant of 100 women, with 92 of them also having m-ACR assessed by two radiologists across the entire breast. For g-SoS, ultrasonic waves were transmitted from varying transducer locations and the image misalignments between these were then related analytically to breast SoS.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!