[{"data":1,"prerenderedAt":396},["ShallowReactive",2],{"content-query-cA3e1DF6kZ":3},{"_path":4,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":8,"description":9,"date":10,"cover":11,"type":12,"category":13,"body":14,"_type":390,"_id":391,"_source":392,"_file":393,"_stem":394,"_extension":395},"/technology-blogs/zh/3575","zh",false,"","昇思MindSpore原生论文 | 基于MobileNetV2的家用睡眠呼吸障碍早筛模型","MobileNetV2: A lightweight classification model for home-based sleep apnea screening","2025-01-08","https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2025/01/10/34ed0a8069ed47e6b40ab0e170af9f16.png","technology-blogs","开发者分享",{"type":15,"children":16,"toc":387},"root",[17,25,31,35,40,45,50,61,66,76,81,86,97,105,114,119,124,131,136,141,149,154,159,164,172,180,188,193,201,209,214,219,224,229,236,241,246,251,256,261,266,273,278,285,290,295,300,305,310,318,323,328,335,340,347,352,357,364,372,377,382],{"type":18,"tag":19,"props":20,"children":22},"element","h1",{"id":21},"昇思mindspore原生论文-基于mobilenetv2的家用睡眠呼吸障碍早筛模型",[23],{"type":24,"value":8},"text",{"type":18,"tag":26,"props":27,"children":28},"p",{},[29],{"type":24,"value":30},"论文标题",{"type":18,"tag":26,"props":32,"children":33},{},[34],{"type":24,"value":9},{"type":18,"tag":26,"props":36,"children":37},{},[38],{"type":24,"value":39},"论文来源",{"type":18,"tag":26,"props":41,"children":42},{},[43],{"type":24,"value":44},"arXiv",{"type":18,"tag":26,"props":46,"children":47},{},[48],{"type":24,"value":49},"论文链接",{"type":18,"tag":26,"props":51,"children":52},{},[53],{"type":18,"tag":54,"props":55,"children":59},"a",{"href":56,"rel":57},"http://arxiv.org/abs/2412.19967",[58],"nofollow",[60],{"type":24,"value":56},{"type":18,"tag":26,"props":62,"children":63},{},[64],{"type":24,"value":65},"代码链接",{"type":18,"tag":26,"props":67,"children":68},{},[69],{"type":18,"tag":54,"props":70,"children":73},{"href":71,"rel":72},"https://github.com/mindspore-lab/models/tree/master/research/arxiv%5C_papers/Easy-MobileNetV2",[58],[74],{"type":24,"value":75},"https://github.com/mindspore-lab/models/tree/master/research/arxiv\\_papers/Easy-MobileNetV2",{"type":18,"tag":26,"props":77,"children":78},{},[79],{"type":24,"value":80},"昇思MindSpore作为开源的AI框架，为开发人员带来端边云全场景协同、极简开发、极致性能的体验，支持国内高校/科研机构发表1700+篇AI顶会论文。为鼓励基于昇思MindSpore进行原生创新，昇思开源社区转载、解读系列原生arXiv论文，本文为昇思MindSpore AI arXiv论文系列第6篇。",{"type":18,"tag":26,"props":82,"children":83},{},[84],{"type":24,"value":85},"作者：潘慧",{"type":18,"tag":26,"props":87,"children":88},{},[89,91],{"type":24,"value":90},"感谢各位专家教授与同学的投稿，更多精彩的论文精读文章和开源代码实现请访问Models。更多内容请访问： ",{"type":18,"tag":54,"props":92,"children":95},{"href":93,"rel":94},"https://gitee.com/mindspore/community/issues/I9W2Z3",[58],[96],{"type":24,"value":93},{"type":18,"tag":26,"props":98,"children":99},{},[100],{"type":18,"tag":101,"props":102,"children":104},"img",{"alt":7,"src":103},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2025/01/10/c4da37b75d444057b0359be007c43a5b.png",[],{"type":18,"tag":26,"props":106,"children":107},{},[108],{"type":18,"tag":109,"props":110,"children":111},"strong",{},[112],{"type":24,"value":113},"研究背景",{"type":18,"tag":26,"props":115,"children":116},{},[117],{"type":24,"value":118},"睡眠呼吸障碍（Sleep Apnea，简称SA）是一种常见但容易被忽视的健康问题，其特征是在睡眠期间发生呼吸暂停或低通气。未及时诊断和干预的睡眠呼吸障碍可能会导致心血管疾病、代谢紊乱以及其他健康问题，因此，早期筛查至关重要。",{"type":18,"tag":26,"props":120,"children":121},{},[122],{"type":24,"value":123},"目前，睡眠呼吸障碍的检测主要依赖于多导睡眠图（PSG）测试，这是一种复杂且昂贵的检测方法，需要在专业医疗机构中进行。然而，受限于检测成本和设备可及性，许多患者未能获得及时的诊断和治疗。因此，开发一种基于家用设备的高效、低成本筛查方法具有重要意义。",{"type":18,"tag":26,"props":125,"children":126},{},[127],{"type":18,"tag":101,"props":128,"children":130},{"alt":7,"src":129},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2025/01/10/6f336c03345149b096208ecdc4f36b9a.png",[],{"type":18,"tag":26,"props":132,"children":133},{},[134],{"type":24,"value":135},"图1 PSG采集信号分析图",{"type":18,"tag":26,"props":137,"children":138},{},[139],{"type":24,"value":140},"本研究基于轻量级深度学习模型MobileNetV2，提出了一种高效的家用睡眠呼吸障碍早筛方案。通过利用家用监测设备采集的数据，结合MindSpore框架进行模型构建和优化，该方法能够以较低的计算资源需求实现对睡眠呼吸障碍的准确预测，为广泛普及和推广提供了技术支持。MobileNetV2是一种经典的轻量级神经网络模型，以其效率和高性能在移动设备上得到广泛应用。其核心特点包括深度可分离卷积（Depthwise Separable Convolution）和倒残差结构（Inverted Residuals）。这些设计大幅减少了模型参数量和计算复杂度，使其成为资源受限环境中的理想选择。",{"type":18,"tag":26,"props":142,"children":143},{},[144],{"type":18,"tag":109,"props":145,"children":146},{},[147],{"type":24,"value":148},"作者介绍",{"type":18,"tag":26,"props":150,"children":151},{},[152],{"type":24,"value":153},"论文第一作者为潘慧，研究方向为生理信号处理、医疗电子， 就读于深圳大学医学部，对MindSpore和PyTorch这两个流行的深度学习框架有丰富的实践经验。",{"type":18,"tag":26,"props":155,"children":156},{},[157],{"type":24,"value":158},"论文第二作者，叶继伦教授，深圳大学医学部生物医学教授，中国医疗器械杂志编委，曾担任中国迈瑞公司的项目研发总监，主要从事生命信息监测及应用、医学仪器设计、医学法规与应用等相关的教学、科研工作。",{"type":18,"tag":26,"props":160,"children":161},{},[162],{"type":24,"value":163},"论文通讯作者，张旭教授，在生命信息监测、信号处理等领域具有较高的学术造诣；在内外发表论文十余篇；相关领域的专利十余项。并主编本领域的标准3部。并获得广东省专利优秀奖",{"type":18,"tag":26,"props":165,"children":166},{},[167],{"type":18,"tag":109,"props":168,"children":169},{},[170],{"type":24,"value":171},"论文简介",{"type":18,"tag":26,"props":173,"children":174},{},[175],{"type":18,"tag":109,"props":176,"children":177},{},[178],{"type":24,"value":179},"0****1",{"type":18,"tag":26,"props":181,"children":182},{},[183],{"type":18,"tag":109,"props":184,"children":185},{},[186],{"type":24,"value":187},"研究目标",{"type":18,"tag":26,"props":189,"children":190},{},[191],{"type":24,"value":192},"本研究的目标是构建一个高效、准确且适用于家用设备的睡眠呼吸障碍筛查模型。通过使用MobileNetV2作为基础网络，结合MindSpore框架的优化能力，实现了在设备资源受限条件下的高性能预测。",{"type":18,"tag":26,"props":194,"children":195},{},[196],{"type":18,"tag":109,"props":197,"children":198},{},[199],{"type":24,"value":200},"02",{"type":18,"tag":26,"props":202,"children":203},{},[204],{"type":18,"tag":109,"props":205,"children":206},{},[207],{"type":24,"value":208},"方法概要",{"type":18,"tag":26,"props":210,"children":211},{},[212],{"type":24,"value":213},"1、数据处理",{"type":18,"tag":26,"props":215,"children":216},{},[217],{"type":24,"value":218},"**数据来源：**三大公共数据库提供的睡眠呼吸障碍数据。",{"type":18,"tag":26,"props":220,"children":221},{},[222],{"type":24,"value":223},"**数据预处理：**通过特征提取和归一化处理，提高模型的输入质量。",{"type":18,"tag":26,"props":225,"children":226},{},[227],{"type":24,"value":228},"**标签定义：**以AHI（呼吸暂停低通气指数）作为主要衡量指标。",{"type":18,"tag":26,"props":230,"children":231},{},[232],{"type":18,"tag":101,"props":233,"children":235},{"alt":7,"src":234},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2025/01/10/32fe2c814e814369b4e13434af8b5923.png",[],{"type":18,"tag":26,"props":237,"children":238},{},[239],{"type":24,"value":240},"图2 数据预处理流程（部分）",{"type":18,"tag":26,"props":242,"children":243},{},[244],{"type":24,"value":245},"2、模型设计",{"type":18,"tag":26,"props":247,"children":248},{},[249],{"type":24,"value":250},"基于MobileNetV2的主干网络，集成了一层全连接分类层。",{"type":18,"tag":26,"props":252,"children":253},{},[254],{"type":24,"value":255},"对MobileNetV2中的关键模块进行了轻量化调整，以适配家用设备的数据特性。",{"type":18,"tag":26,"props":257,"children":258},{},[259],{"type":24,"value":260},"**深度可分离卷积：**分解标准卷积为深度卷积和点卷积，有效降低计算量。",{"type":18,"tag":26,"props":262,"children":263},{},[264],{"type":24,"value":265},"**倒残差结构：**通过线性瓶颈连接，减少冗余特征的传播。",{"type":18,"tag":26,"props":267,"children":268},{},[269],{"type":18,"tag":101,"props":270,"children":272},{"alt":7,"src":271},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2025/01/10/b4a08b6691e84ca8bab6dc32d3c3c64e.png",[],{"type":18,"tag":26,"props":274,"children":275},{},[276],{"type":24,"value":277},"图3 深度可分离卷积图解",{"type":18,"tag":26,"props":279,"children":280},{},[281],{"type":18,"tag":101,"props":282,"children":284},{"alt":7,"src":283},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2025/01/10/1de0f35408794df59ead84c19fbf89c8.png",[],{"type":18,"tag":26,"props":286,"children":287},{},[288],{"type":24,"value":289},"图4 模型训练流程",{"type":18,"tag":26,"props":291,"children":292},{},[293],{"type":24,"value":294},"3、模型训练",{"type":18,"tag":26,"props":296,"children":297},{},[298],{"type":24,"value":299},"**框架：**采用华为昇思MindSpore进行模型构建和训练。",{"type":18,"tag":26,"props":301,"children":302},{},[303],{"type":24,"value":304},"**优化器：**使用Adam优化器。",{"type":18,"tag":26,"props":306,"children":307},{},[308],{"type":24,"value":309},"**损失函数：**均方误差（MSE）。",{"type":18,"tag":26,"props":311,"children":312},{},[313],{"type":18,"tag":109,"props":314,"children":315},{},[316],{"type":24,"value":317},"实验结果",{"type":18,"tag":26,"props":319,"children":320},{},[321],{"type":24,"value":322},"为了验证我们模型的性能，我们在睡眠呼吸相关数据集上进行了实验。采用相关性系数和均方误差（MSE）作为主要评价指标。",{"type":18,"tag":26,"props":324,"children":325},{},[326],{"type":24,"value":327},"实验结果表明，该模型在预测AHI值时表现出较高的准确性，其预测曲线与实际曲线的相关性系数达到0.8831，显著优于其他轻量级模型。",{"type":18,"tag":26,"props":329,"children":330},{},[331],{"type":18,"tag":101,"props":332,"children":334},{"alt":7,"src":333},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2025/01/10/5681225537114a6694e307079666f8b9.png",[],{"type":18,"tag":26,"props":336,"children":337},{},[338],{"type":24,"value":339},"图5 四分类和二分类混淆矩阵",{"type":18,"tag":26,"props":341,"children":342},{},[343],{"type":18,"tag":101,"props":344,"children":346},{"alt":7,"src":345},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2025/01/10/77a75e972ce0457ab3cc4d316314252b.png",[],{"type":18,"tag":26,"props":348,"children":349},{},[350],{"type":24,"value":351},"图6 预测值和真实值的散点图&相关系数",{"type":18,"tag":26,"props":353,"children":354},{},[355],{"type":24,"value":356},"此外，在资源占用和推理速度方面，基于MindSpore的MobileNetV2模型也展现出明显优势。",{"type":18,"tag":26,"props":358,"children":359},{},[360],{"type":18,"tag":101,"props":361,"children":363},{"alt":7,"src":362},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2025/01/10/46e8a8bc51c648928df7e08d60a0c537.png",[],{"type":18,"tag":26,"props":365,"children":366},{},[367],{"type":18,"tag":109,"props":368,"children":369},{},[370],{"type":24,"value":371},"总结与展望",{"type":18,"tag":26,"props":373,"children":374},{},[375],{"type":24,"value":376},"本研究提出了一种基于MobileNetV2的家用睡眠呼吸障碍筛查模型，通过在MindSpore框架上的实现，该模型在保证高预测精度的同时大幅降低了计算成本。实验结果验证了该模型在实际应用场景中的潜力。我们的模型在公开数据集上实现了更好的性能，同时保持了更小的模型体积。",{"type":18,"tag":26,"props":378,"children":379},{},[380],{"type":24,"value":381},"MindSpore作为一个新兴的深度学习框架，显示出了其灵活性和易用性，特别是在支持多种硬件平台上。本研究利用了MindSpore的特性来实现和验证提出的模型，发现其提供了强大的工具链和丰富的API集合，有助于快速原型设计和模型迭代MindSpore有潜力成为深度学习框架中的佼佼者之一。",{"type":18,"tag":26,"props":383,"children":384},{},[385],{"type":24,"value":386},"MindSpore提供了丰富的功能和工具，可以帮助开发者们更快地实现自己的想法。社区的活跃也为解决问题和分享经验提供了一个良好的平台。希望更多的开发者能够加入进来，共同推动MindSpore的发展，并创造出更多有意义的应用。",{"title":7,"searchDepth":388,"depth":388,"links":389},4,[],"markdown","content:technology-blogs:zh:3575.md","content","technology-blogs/zh/3575.md","technology-blogs/zh/3575","md",1776506131416]