{"title":"用于异常检测的无监督混合框架 (HAND) -- 应用于乳房X光筛查","authors":"Zhemin Zhang, Bhavika Patel, Bhavik Patel, Imon Banerjee","doi":"arxiv-2409.11534","DOIUrl":null,"url":null,"abstract":"Out-of-distribution (OOD) detection is crucial for enhancing the\ngeneralization of AI models used in mammogram screening. Given the challenge of\nlimited prior knowledge about OOD samples in external datasets, unsupervised\ngenerative learning is a preferable solution which trains the model to discern\nthe normal characteristics of in-distribution (ID) data. The hypothesis is that\nduring inference, the model aims to reconstruct ID samples accurately, while\nOOD samples exhibit poorer reconstruction due to their divergence from\nnormality. Inspired by state-of-the-art (SOTA) hybrid architectures combining\nCNNs and transformers, we developed a novel backbone - HAND, for detecting OOD\nfrom large-scale digital screening mammogram studies. To boost the learning\nefficiency, we incorporated synthetic OOD samples and a parallel discriminator\nin the latent space to distinguish between ID and OOD samples. Gradient\nreversal to the OOD reconstruction loss penalizes the model for learning OOD\nreconstructions. An anomaly score is computed by weighting the reconstruction\nand discriminator loss. On internal RSNA mammogram held-out test and external\nMayo clinic hand-curated dataset, the proposed HAND model outperformed\nencoder-based and GAN-based baselines, and interestingly, it also outperformed\nthe hybrid CNN+transformer baselines. Therefore, the proposed HAND pipeline\noffers an automated efficient computational solution for domain-specific\nquality checks in external screening mammograms, yielding actionable insights\nwithout direct exposure to the private medical imaging data.","PeriodicalId":501289,"journal":{"name":"arXiv - EE - Image and Video Processing","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2024-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Unsupervised Hybrid framework for ANomaly Detection (HAND) -- applied to Screening Mammogram\",\"authors\":\"Zhemin Zhang, Bhavika Patel, Bhavik Patel, Imon Banerjee\",\"doi\":\"arxiv-2409.11534\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Out-of-distribution (OOD) detection is crucial for enhancing the\\ngeneralization of AI models used in mammogram screening. Given the challenge of\\nlimited prior knowledge about OOD samples in external datasets, unsupervised\\ngenerative learning is a preferable solution which trains the model to discern\\nthe normal characteristics of in-distribution (ID) data. The hypothesis is that\\nduring inference, the model aims to reconstruct ID samples accurately, while\\nOOD samples exhibit poorer reconstruction due to their divergence from\\nnormality. Inspired by state-of-the-art (SOTA) hybrid architectures combining\\nCNNs and transformers, we developed a novel backbone - HAND, for detecting OOD\\nfrom large-scale digital screening mammogram studies. To boost the learning\\nefficiency, we incorporated synthetic OOD samples and a parallel discriminator\\nin the latent space to distinguish between ID and OOD samples. Gradient\\nreversal to the OOD reconstruction loss penalizes the model for learning OOD\\nreconstructions. An anomaly score is computed by weighting the reconstruction\\nand discriminator loss. On internal RSNA mammogram held-out test and external\\nMayo clinic hand-curated dataset, the proposed HAND model outperformed\\nencoder-based and GAN-based baselines, and interestingly, it also outperformed\\nthe hybrid CNN+transformer baselines. Therefore, the proposed HAND pipeline\\noffers an automated efficient computational solution for domain-specific\\nquality checks in external screening mammograms, yielding actionable insights\\nwithout direct exposure to the private medical imaging data.\",\"PeriodicalId\":501289,\"journal\":{\"name\":\"arXiv - EE - Image and Video Processing\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-09-17\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - EE - Image and Video Processing\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2409.11534\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - EE - Image and Video Processing","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.11534","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
摘要
分布外(OOD)检测对于提高乳腺 X 光筛查所用人工智能模型的泛化能力至关重要。鉴于外部数据集中有关 OOD 样本的先验知识有限,无监督生成学习是一种可取的解决方案,它可以训练模型辨别分布内(ID)数据的正常特征。假设在推理过程中,模型的目标是准确重建 ID 样本,而 OOD 样本由于偏离正态性,重建效果较差。受结合了 CNN 和变压器的最先进(SOTA)混合体系结构的启发,我们开发了一种新型骨架--HAND,用于从大规模数字乳腺 X 光筛查研究中检测 OOD。为了提高学习效率,我们在潜空间中加入了合成 OOD 样本和并行判别器,以区分 ID 和 OOD 样本。对 OOD 重建损失的梯度反转对学习 OOD 重建的模型进行惩罚。通过对重构损失和判别损失进行加权,计算出异常得分。在内部 RSNA 乳房 X 射线照片保留测试和外部马约诊所人工合成数据集上,拟议的 HAND 模型优于基于编码器和基于 GAN 的基线,有趣的是,它还优于混合 CNN+ 变换器基线。因此,所提出的 HAND 流水线为外部乳房 X 光筛查中特定领域的质量检查提供了自动化的高效计算解决方案,在不直接接触私人医疗成像数据的情况下产生了可操作的洞察力。
Unsupervised Hybrid framework for ANomaly Detection (HAND) -- applied to Screening Mammogram
Out-of-distribution (OOD) detection is crucial for enhancing the
generalization of AI models used in mammogram screening. Given the challenge of
limited prior knowledge about OOD samples in external datasets, unsupervised
generative learning is a preferable solution which trains the model to discern
the normal characteristics of in-distribution (ID) data. The hypothesis is that
during inference, the model aims to reconstruct ID samples accurately, while
OOD samples exhibit poorer reconstruction due to their divergence from
normality. Inspired by state-of-the-art (SOTA) hybrid architectures combining
CNNs and transformers, we developed a novel backbone - HAND, for detecting OOD
from large-scale digital screening mammogram studies. To boost the learning
efficiency, we incorporated synthetic OOD samples and a parallel discriminator
in the latent space to distinguish between ID and OOD samples. Gradient
reversal to the OOD reconstruction loss penalizes the model for learning OOD
reconstructions. An anomaly score is computed by weighting the reconstruction
and discriminator loss. On internal RSNA mammogram held-out test and external
Mayo clinic hand-curated dataset, the proposed HAND model outperformed
encoder-based and GAN-based baselines, and interestingly, it also outperformed
the hybrid CNN+transformer baselines. Therefore, the proposed HAND pipeline
offers an automated efficient computational solution for domain-specific
quality checks in external screening mammograms, yielding actionable insights
without direct exposure to the private medical imaging data.