arXiv 2610.136752026-10-12eess.IVcs.CV

多模态决策模型能否对乳腺超声图像进行分类？在不同提示和数据集上对CLEF与CLEF-Flash进行基准测试

Can Multimodal Decision Models Classify Breast Ultrasound Images? Benchmarking CLEF and CLEF-Flash Across Prompts and Datasets

Research and Development Center, VNSO Technology JSC（VNSO科技股份研发中心）；Department of Information Technology, Ho Chi Minh University of Education（胡志明市教育大学信息技术系）；Faculty of Information Science and Engineering, University of Information Technology（胡志明市信息技术大学信息科学与工程学部）；Vietnam National University, Ho Chi Minh City（胡志明市越南国立大学）

Tran Cao Minh, Hai Tran, Ha Minh Tan

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本研究在三个乳腺超声数据集上对CLEF与CLEF-Flash在三种提示下的分类性能进行基准测试，发现性能依赖于数据集和提示，CLEF-Flash在临床提示下准确率最高，CLEF在BUSI任务引导提示下宏F1值最高，CLEF-Flash延迟更低。

摘要：

多模态决策模型会返回预定义备选方案的概率，为图像分类提供结构化接口。本研究在来自数据集B、BUSI和BUS-BRA的2818张乳腺超声图像上，对CLEF与CLEF-Flash的推理性能进行基准测试。数据集B和BUS-BRA采用良性-恶性二分类，而BUSI包含正常、良性和恶性三类。模型参数保持不变，且提交的原始图像未经过用户定义的预处理。三种提示分别提供最小标签、明确的分类任务或超声检查标准。我们报告了准确率、宏F1值、平衡准确率、宏精确率、吞吐量和延迟。CLEF-Flash在临床提示下在所有三个数据集上均达到最高准确率：数据集B上为74.85%，BUSI上为66.28%，BUS-BRA上为51.63%。然而在BUSI上，CLEF在任务引导提示下达到最高宏F1值（64.51%）和平衡准确率（68.72%）。在所有匹配的比较中，CLEF-Flash的延迟更低。与最小提示相比，临床提示可提高CLEF-Flash的准确率，但会降低CLEF在所有数据集上的准确率。混淆矩阵显示存在频繁的良性到恶性的错误，尤其是在BUS-BRA上，所有配置的准确率均低于多数类预测器的67.63%。这些发现表明性能依赖于数据集和提示，为将多模态决策模型应用于乳腺超声分析，需进行类别特定评估。

英文摘要：

Multimodal decision models return probabilities over predefined alternatives, offering a structured interface for image classification. This study benchmarks inference with CLEF and CLEF-Flash on 2818 breast ultrasound images from Dataset B, BUSI, and BUS-BRA. Dataset B and BUS-BRA use binary benign-malignant classification, whereas BUSI includes normal, benign, and malignant categories. Model parameters remain unchanged, and original images are submitted without user-defined preprocessing. Three prompts provide minimal labels, an explicit classification task, or sonographic criteria. We report accuracy, macro F1, balanced accuracy, macro precision, throughput, and latency. CLEF-Flash with the clinical prompt achieves the highest accuracy on all three datasets: 74.85% on Dataset B, 66.28% on BUSI, and 51.63% on BUS-BRA. On BUSI, however, CLEF with the task-guided prompt achieves the highest macro F1 (64.51%) and balanced accuracy (68.72%). CLEF-Flash has lower latency in every matched comparison. Clinical instructions improve CLEF-Flash accuracy relative to the minimal prompt but reduce CLEF accuracy on all datasets. Confusion matrices reveal frequent benign-to-malignant errors, particularly on BUS-BRA, where all configurations fall below the 67.63% accuracy of a majority-class predictor. These findings demonstrate dataset-and prompt-dependent performance and motivate class-specific evaluation before applying multimodal decision models to breast ultrasound analysis.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。