WhACC:具备人类专家级性能的胡须接触自动分类器
WhACC: Whisker Automatic Contact Classifier with Expert Human-Level Performance
AI总结:
针对啮齿动物胡须触觉事件人工标注耗时久的问题,本研究提出结合ResNet50V2与LightGBM的WhACC分类器,性能达人类专家水平,支持自定义重训练,可大幅降低标注人力成本。
摘要:
啮齿动物的 vibrissal(胡须)系统对推动神经科学研究至关重要,尤其在皮层可塑性、学习、决策、感觉编码以及感觉运动整合相关研究中作用突出。尽管该系统具备研究优势,但整理触觉事件的工作十分耗费人力,即便使用了 Janelia Whisker Tracker(詹妮莉亚胡须追踪器)这类自动化工具,每处理百万帧视频通常仍需3小时以上。\n为解决这一局限,我们提出了 Whisker Automatic Contact Classifier(WhACC,胡须接触自动分类器),这是一个Python工具包,可从头部固定的行为学啮齿动物高速视频中识别触觉时段,性能达到人类水平。WhACC 采用 ResNet50V2 进行特征提取,结合 LightGBM 完成分类任务。\n我们在超百万帧视频上,将其性能与三位人类专家的标注结果进行对比:在99.5%的视频帧上,二者的触觉分类两两一致性与人类之间的一致性水平相当。最后,我们提供了自定义重训练接口,支持基于小批量数据定制模型,该功能已在涵盖16组单单位电生理记录的400万帧数据上得到验证。加入该重训练步骤后,整理1亿帧数据集所需的人工时长从约333小时降至约6小时。
英文摘要:
The rodent vibrissal system is pivotal in advancing neuroscience research, particularly for studies of cortical plasticity, learning, decision-making, sensory encoding, and sensorimotor integration. Despite the advantages, curating touch events is labor intensive and often requires >3 hours per million video frames, even after leveraging automated tools like the Janelia Whisker Tracker. We address this limitation by introducing Whisker Automatic Contact Classifier (WhACC), a python package designed to identify touch periods from high-speed videos of head-fixed behaving rodents with human-level performance. WhACC leverages ResNet50V2 for feature extraction, combined with LightGBM for Classification. Performance is assessed against three expert human curators on over one million frames. Pairwise touch classification agreement on 99.5% of video frames, equal to between-human agreement. Finally, we offer a custom retraining interface to allow model customization on a small subset of data, which was validated on four million frames across 16 single-unit electrophysiology recordings. Including this retraining step, we reduce human hours required to curate a 100 million frame dataset from ~333 hours to ~6 hours.
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。