{"id":892,"date":"2026-01-20T00:30:19","date_gmt":"2026-01-19T15:30:19","guid":{"rendered":"https:\/\/itexplore.org\/jp\/columns\/ai-trends-ecg-facial-expression-timescale-learning\/"},"modified":"2026-01-20T00:30:19","modified_gmt":"2026-01-19T15:30:19","slug":"ai-trends-ecg-facial-expression-timescale-learning","status":"publish","type":"post","link":"https:\/\/itexplore.org\/jp\/columns\/ai-trends-ecg-facial-expression-timescale-learning\/","title":{"rendered":"AI\u6280\u8853\u306e\u6700\u65b0\u52d5\u5411\uff1a\u5fc3\u96fb\u56f3\u3001\u8868\u60c5\u8a8d\u8b58\u3001\u6642\u9593\u30b9\u30b1\u30fc\u30eb\u5b66\u7fd2"},"content":{"rendered":"<p>\u672c\u65e5\u306e\u6ce8\u76eeAI\u30fb\u30c6\u30c3\u30af\u30cb\u30e5\u30fc\u30b9\u3092\u3001\u5c02\u9580\u7684\u306a\u5206\u6790\u3068\u5171\u306b\u304a\u5c4a\u3051\u3057\u307e\u3059\u3002<\/p>\n<div class=\"wp-block-vk-blocks-alert vk_alert alert alert-warning has-alert-icon\">\n<div class=\"vk_alert_icon\">\n<div class=\"vk_alert_icon_icon\"><i class=\"fa-solid fa-triangle-exclamation\" aria-hidden=\"true\"><\/i><\/div>\n<div class=\"vk_alert_icon_text\"><span>Warning<\/span><\/div>\n<\/div>\n<div class=\"vk_alert_content\">\n<p>\u3053\u306e\u8a18\u4e8b\u306fAI\u306b\u3088\u3063\u3066\u81ea\u52d5\u751f\u6210\u30fb\u5206\u6790\u3055\u308c\u305f\u3082\u306e\u3067\u3059\u3002AI\u306e\u6027\u8cea\u4e0a\u3001\u4e8b\u5b9f\u8aa4\u8a8d\u304c\u542b\u307e\u308c\u308b\u53ef\u80fd\u6027\u304c\u3042\u308b\u305f\u3081\u3001\u91cd\u8981\u306a\u5224\u65ad\u3092\u4e0b\u3059\u969b\u306f\u5fc5\u305a\u30ea\u30f3\u30af\u5148\u306e\u4e00\u6b21\u30bd\u30fc\u30b9\u3092\u3054\u78ba\u8a8d\u304f\u3060\u3055\u3044\u3002<\/p>\n<\/div>\n<\/div>\n<div class=\"wp-block-group\" style=\"margin-top:40px;margin-bottom:40px\">\n<h2 class=\"wp-block-heading\">\u5fc3\u81d3\u3068\u8133\u306e\u3064\u306a\u304c\u308a\u3092\u89e3\u660e\uff1a\u8a8d\u77e5\u30d1\u30d5\u30a9\u30fc\u30de\u30f3\u30b9\u306b\u304a\u3051\u308bECG\u306e\u5206\u6790<\/h2>\n<ul>\n<li><strong>\u539f\u984c:<\/strong> Unveiling the Heart-Brain Connection: An Analysis of ECG in Cognitive Performance<\/li>\n<\/ul>\n<h3 class=\"wp-block-heading\">\u5c02\u9580\u30a2\u30ca\u30ea\u30b9\u30c8\u306e\u5206\u6790<\/h3>\n<div class=\"ai-summary-content\">\n<p>\u672c\u7814\u7a76\u3067\u306f\u3001\u30a6\u30a7\u30a2\u30e9\u30d6\u30eb\u30c7\u30d0\u30a4\u30b9\u3067\u5e83\u304f\u5229\u7528\u53ef\u80fd\u306a<strong>ECG<\/strong>\u4fe1\u53f7\u304c\u8a8d\u77e5\u8ca0\u8377\u3092\u3069\u306e\u7a0b\u5ea6 reliably \u306b\u53cd\u6620\u3067\u304d\u308b\u304b\u3092\u8abf\u67fb\u3057\u3066\u3044\u307e\u3059\u3002<strong>XGBoost<\/strong>\u30d5\u30ec\u30fc\u30e0\u30ef\u30fc\u30af\u3092\u7528\u3044\u305f\u30af\u30ed\u30b9\u30e2\u30fc\u30c0\u30eb\u30a2\u30d7\u30ed\u30fc\u30c1\u306b\u3088\u308a\u3001<strong>ECG<\/strong>\u7279\u5fb4\u91cf\u3092<strong>EEG<\/strong>\u4ee3\u8868\u7684\u306a\u8a8d\u77e5\u7a7a\u9593\u306b\u6295\u5f71\u3057\u3001<strong>ECG<\/strong>\u306e\u307f\u3067\u30ef\u30fc\u30af\u30ed\u30fc\u30c9\u63a8\u8ad6\u3092\u53ef\u80fd\u306b\u3059\u308b\u3053\u3068\u3092\u76ee\u6307\u3057\u3066\u3044\u307e\u3059\u3002<\/p>\n<p>\u7d50\u679c\u3068\u3057\u3066\u3001<strong>ECG<\/strong>\u7531\u6765\u306e\u6295\u5f71\u306f\u8a8d\u77e5\u72b6\u614b\u306e\u5909\u5316\u3092\u52b9\u679c\u7684\u306b\u6349\u3048\u3001\u6b63\u78ba\u306a\u5206\u985e\u3092\u30b5\u30dd\u30fc\u30c8\u3059\u308b\u3053\u3068\u304c\u793a\u3055\u308c\u307e\u3057\u305f\u3002\u3053\u306e\u767a\u898b\u306f\u3001\u65e5\u5e38\u7684\u306a\u8a8d\u77e5\u30e2\u30cb\u30bf\u30ea\u30f3\u30b0\u306e\u305f\u3081\u306e\u89e3\u91c8\u53ef\u80fd\u3067\u30ea\u30a2\u30eb\u30bf\u30a4\u30e0\u306a\u30a6\u30a7\u30a2\u30e9\u30d6\u30eb\u30bd\u30ea\u30e5\u30fc\u30b7\u30e7\u30f3\u3068\u3057\u3066\u306e<strong>ECG<\/strong>\u306e\u53ef\u80fd\u6027\u3092\u88cf\u4ed8\u3051\u3066\u3044\u307e\u3059\u3002<\/p>\n<p>\ud83d\udc49 <strong><a href=\"https:\/\/arxiv.org\/abs\/2601.01424\" target=\"_blank\" rel=\"noopener\">arXiv \u3067\u8a18\u4e8b\u5168\u6587\u3092\u8aad\u3080<\/a><\/strong><\/p>\n<\/div>\n<ul>\n<li><strong>\u8981\u70b9:<\/strong> ECG signals can be used to infer cognitive load, offering a wearable alternative to EEG for continuous monitoring.<\/li>\n<li><strong>\u8457\u8005:<\/strong> Akshay Sasi, Malavika Pradeep, Nusaibah Farrukh, Rahul Venugopal, Elizabeth Sherly<\/li>\n<\/ul>\n<blockquote class=\"wp-block-quote\"><p><span>English Summary:<\/span><\/p>\n<p>This research investigates whether <strong>ECG<\/strong> signals, readily available through wearable devices, can reliably reflect cognitive load. Utilizing a cross-modal <strong>XGBoost<\/strong> framework, the study aims to project <strong>ECG<\/strong> features onto <strong>EEG<\/strong>-representative cognitive spaces, enabling workload inference using only <strong>ECG<\/strong>.<\/p>\n<p>The results demonstrate that <strong>ECG<\/strong>-derived projections effectively capture variations in cognitive states and provide strong support for accurate classification. These findings underscore the potential of <strong>ECG<\/strong> as an interpretable, real-time, and wearable solution for everyday cognitive monitoring.<\/p>\n<\/blockquote>\n<\/div>\n<div class=\"wp-block-group\" style=\"margin-top:40px;margin-bottom:40px\">\n<h2 class=\"wp-block-heading\">\u5cf6\u76ae\u8cea\u306e\u982d\u84cb\u5185\u6d3b\u52d5\u306f\u3001\u5358\u4e00\u30b3\u30f3\u30bf\u30af\u30c8\u30ec\u30d9\u30eb\u3067\u306e\u591a\u69d8\u3067\u6df7\u5728\u3059\u308b\u6642\u9593\u30d1\u30bf\u30fc\u30f3\u3092\u4ecb\u3057\u3066\u8907\u6570\u306e\u9854\u8868\u60c5\u3092\u8b58\u5225\u3059\u308b<\/h2>\n<ul>\n<li><strong>\u539f\u984c:<\/strong> Insular intracranial activity identifies multiple facial expressions via diverse, intermixed temporal patterns at the single-contact level<\/li>\n<\/ul>\n<h3 class=\"wp-block-heading\">\u5c02\u9580\u30a2\u30ca\u30ea\u30b9\u30c8\u306e\u5206\u6790<\/h3>\n<div class=\"ai-summary-content\">\n<p>\u672c\u7814\u7a76\u3067\u306f\u3001\u4eba\u9593\u306e\u88ab\u9a13\u8005\u304c\u9854\u306e\u8868\u60c5\u8a8d\u8b58\u30bf\u30b9\u30af\u3092\u5b9f\u884c\u3059\u308b\u969b\u306e<strong>\u5cf6\u76ae\u8cea<\/strong>\u306e<strong>\u982d\u84cb\u5185\u6d3b\u52d5<\/strong>\u3092<strong>\u8133\u6ce2<\/strong>(<strong>EEG<\/strong>)\u30c7\u30fc\u30bf\u3092\u7528\u3044\u3066\u8a18\u9332\u3057\u307e\u3057\u305f\u3002<strong>\u5cf6\u76ae\u8cea<\/strong>\u306e\u6d3b\u52d5\u306f\u3001\u03b8\u304b\u3089\u9ad8\u03b3\u5468\u6ce2\u6570\u5e2f\u57df\u306e\u30a4\u30d9\u30f3\u30c8\u95a2\u9023\u96fb\u4f4d(<strong>ERP<\/strong>)\u3068\u30a4\u30d9\u30f3\u30c8\u95a2\u9023\u30b9\u30da\u30af\u30c8\u30eb\u6442\u52d5(<strong>ERSP<\/strong>)\u306e\u4e21\u65b9\u306e\u5f62\u72b6\u3068\u30b9\u30b1\u30fc\u30eb\u3092\u6349\u3048\u308b\u3053\u3068\u3067\u3001\u8868\u60c5\u30ab\u30c6\u30b4\u30ea\u306e\u7279\u7570\u6027\u3092\u8a55\u4fa1\u3057\u307e\u3057\u305f\u3002<\/p>\n<p><strong>\u5cf6\u76ae\u8cea<\/strong>\u306e\u6d3b\u52d5\u306f\u3001\u5cf6\u76ae\u8cea\u5168\u4f53\u306b\u6df7\u5728\u3059\u308b\u591a\u69d8\u306a<strong>ERP<\/strong>\u5fdc\u7b54\u306b\u3088\u3063\u3066\u5a92\u4ecb\u3055\u308c\u3001\u8abf\u67fb\u3055\u308c\u305f\u3059\u3079\u3066\u306e\u8868\u60c5\u3092\u6210\u529f\u88cf\u306b\u8b58\u5225\u3057\u307e\u3057\u305f\u3002\u5bfe\u7167\u7684\u306b\u3001<strong>\u7d21\u9318\u72b6\u9854\u9818\u57df<\/strong>\u306f\u8868\u60c5\u3068\u30b3\u30f3\u30bf\u30af\u30c8\u9593\u3067\u53ce\u675f\u3057\u305f<strong>ERP<\/strong>\u5fdc\u7b54\u3092\u793a\u3057\u307e\u3057\u305f\u3002\u3053\u308c\u3089\u306e\u767a\u898b\u306f\u3001\u9854\u306e\u611f\u60c5\u77e5\u899a\u306b\u304a\u3051\u308b<strong>\u5cf6\u76ae\u8cea<\/strong>\u306e\u795e\u7d4c\u30e1\u30ab\u30cb\u30ba\u30e0\u3092\u89e3\u660e\u3057\u3001\u305d\u306e\u591a\u69d8\u306a\u5fdc\u7b54\u30d7\u30ed\u30d5\u30a1\u30a4\u30eb\u3092\u6d3b\u7528\u3057\u3066\u591a\u7528\u9014\u306a\u8a8d\u77e5\u304a\u3088\u3073\u611f\u60c5\u6a5f\u80fd\u306e\u91cd\u8981\u306a\u30cf\u30d6\u3068\u3057\u3066\u6a5f\u80fd\u3059\u308b\u53ef\u80fd\u6027\u3092\u793a\u5506\u3057\u3066\u3044\u307e\u3059\u3002<\/p>\n<p>\ud83d\udc49 <strong><a href=\"https:\/\/arxiv.org\/abs\/2601.01782\" target=\"_blank\" rel=\"noopener\">arXiv \u3067\u8a18\u4e8b\u5168\u6587\u3092\u8aad\u3080<\/a><\/strong><\/p>\n<\/div>\n<ul>\n<li><strong>\u8981\u70b9:<\/strong> Insular cortex activity, characterized by diverse temporal patterns at the single-contact level, can identify multiple facial expressions, highlighting its role in versatile cognitive and emotional functions.<\/li>\n<li><strong>\u8457\u8005:<\/strong> Yingyu Huang, Lisen Sui, Liying Zhan, Chaolun Wang, Zhihan Guo, Yanjuan Li, Xiang Wu<\/li>\n<\/ul>\n<blockquote class=\"wp-block-quote\"><p><span>English Summary:<\/span><\/p>\n<p>This study recorded <strong>intracranial activity<\/strong> from the <strong>insula<\/strong> in human subjects performing a facial emotion recognition task using <strong>EEG<\/strong> data. The specificity of insular activity to expression categories was assessed by capturing both the shape and scale of event-related potentials (<strong>ERPs<\/strong>) and event-related spectral perturbations (<strong>ERSPs<\/strong>) across theta to high-gamma frequency ranges.<\/p>\n<p>Insular activity successfully identified all investigated expressions, mediated by diverse <strong>ERP<\/strong> responses intermixed across the insula. In contrast, the <strong>fusiform face area<\/strong> exhibited convergent <strong>ERP<\/strong> responses across expressions and contacts. These findings elucidate the insula's neural mechanisms for facial emotion perception and suggest its potential role as a key hub for versatile cognitive and emotional functions by leveraging its heterogeneous response profiles.<\/p>\n<\/blockquote>\n<\/div>\n<div class=\"wp-block-group\" style=\"margin-top:40px;margin-bottom:40px\">\n<h2 class=\"wp-block-heading\">\u751f\u7269\u5b66\u7684\u306b\u5236\u7d04\u3055\u308c\u305f\u30b9\u30b1\u30fc\u30eb\u4e0d\u5909\u6df1\u5c64\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u306b\u304a\u3051\u308b\u968e\u5c64\u7684\u6642\u9593\u53d7\u5bb9\u91ce\u3068\u30bc\u30ed\u30b7\u30e7\u30c3\u30c8\u6642\u9593\u30b9\u30b1\u30fc\u30eb\u6c4e\u5316<\/h2>\n<ul>\n<li><strong>\u539f\u984c:<\/strong> Hierarchical temporal receptive windows and zero-shot timescale generalization in biologically constrained scale-invariant deep networks<\/li>\n<\/ul>\n<h3 class=\"wp-block-heading\">\u5c02\u9580\u30a2\u30ca\u30ea\u30b9\u30c8\u306e\u5206\u6790<\/h3>\n<div class=\"ai-summary-content\">\n<p>\u672c\u7814\u7a76\u3067\u306f\u3001\u4eba\u9593\u306e\u8a8d\u77e5\u304c\u5165\u308c\u5b50\u306b\u306a\u3063\u305f\u6642\u9593\u30b9\u30b1\u30fc\u30eb\u306b\u308f\u305f\u3063\u3066\u60c5\u5831\u3092\u7d71\u5408\u3059\u308b\u3068\u3044\u3046\u8003\u3048\u306b\u57fa\u3065\u304d\u3001\u751f\u7269\u5b66\u7684\u306b\u5236\u7d04\u3055\u308c\u305f\u6df1\u5c64\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u3092\u8a13\u7df4\u3057\u307e\u3057\u305f\u3002\u30b9\u30b1\u30fc\u30eb\u4e0d\u5909\u306e\u6d77\u99ac\u6642\u9593\u7d30\u80de\u306b\u57fa\u3065\u3044\u305f<strong>SITHCon<\/strong>\u30e2\u30c7\u30eb\u3067\u306f\u3001\u5c64\u5185\u306b<strong>\u968e\u5c64\u7684\u306a\u6642\u9593\u53d7\u5bb9\u91ce(TRWs)<\/strong>\u304c\u81ea\u7136\u306b\u73fe\u308c\u308b\u3053\u3068\u304c\u767a\u898b\u3055\u308c\u307e\u3057\u305f\u3002<\/p>\n<p>\u3053\u306e\u767a\u898b\u3092\u751f\u7269\u5b66\u7684\u306b\u59a5\u5f53\u306a\u518d\u5e30\u578b\u30a2\u30fc\u30ad\u30c6\u30af\u30c1\u30e3\u3067\u3042\u308b<strong>SITH-RNN<\/strong>\u306b\u84b8\u7559\u3057\u305f\u7d50\u679c\u3001<strong>SITH-RNN<\/strong>\u306f\u3001\u3088\u308a\u5c11\u306a\u3044\u30d1\u30e9\u30e1\u30fc\u30bf\u3067\u3088\u308a\u901f\u304f\u5b66\u7fd2\u3057\u3001\u5206\u5e03\u5916\u306e\u6642\u9593\u30b9\u30b1\u30fc\u30eb\u306b\u5bfe\u3057\u3066<strong>\u30bc\u30ed\u30b7\u30e7\u30c3\u30c8\u6c4e\u5316<\/strong>\u3092\u793a\u3059\u3053\u3068\u304c\u660e\u3089\u304b\u306b\u306a\u308a\u307e\u3057\u305f\u3002\u3053\u308c\u3089\u306e\u7d50\u679c\u306f\u3001\u8133\u304c\u30b9\u30b1\u30fc\u30eb\u4e0d\u5909\u306e\u9010\u6b21\u7684\u306a\u4e8b\u524d\u77e5\u8b58\u3092\u5229\u7528\u3057\u3066\u300c\u3044\u3064\u300d\u300c\u4f55\u304c\u300d\u8d77\u3053\u3063\u305f\u304b\u3092\u7b26\u53f7\u5316\u3057\u3066\u3044\u308b\u53ef\u80fd\u6027\u3092\u793a\u5506\u3057\u3066\u304a\u308a\u3001\u305d\u306e\u3088\u3046\u306a\u4e8b\u524d\u77e5\u8b58\u3092\u6301\u3064\u518d\u5e30\u578b\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u304c\u4eba\u9593\u306e\u8a8d\u77e5\u3092\u8a18\u8ff0\u3059\u308b\u306e\u306b\u7279\u306b\u9069\u3057\u3066\u3044\u308b\u3053\u3068\u3092\u793a\u3057\u3066\u3044\u307e\u3059\u3002<\/p>\n<p>\ud83d\udc49 <strong><a href=\"https:\/\/arxiv.org\/abs\/2601.02618\" target=\"_blank\" rel=\"noopener\">arXiv \u3067\u8a18\u4e8b\u5168\u6587\u3092\u8aad\u3080<\/a><\/strong><\/p>\n<\/div>\n<ul>\n<li><strong>\u8981\u70b9:<\/strong> Scale-invariant recurrent neural networks with hierarchical temporal receptive windows exhibit faster learning and zero-shot generalization to different timescales, suggesting a biologically plausible model for cognitive processing.<\/li>\n<li><strong>\u8457\u8005:<\/strong> Aakash Sarkar, Marc W. Howard<\/li>\n<\/ul>\n<blockquote class=\"wp-block-quote\"><p><span>English Summary:<\/span><\/p>\n<p>This study trained biologically constrained deep networks based on the principle that human cognition integrates information across nested timescales. In the scale-invariant hippocampal time cell-based model, <strong>SITHCon<\/strong>, a hierarchy of <strong>temporal receptive windows (TRWs)<\/strong> emerged naturally across layers.<\/p>\n<p>Distilling these findings into a biologically plausible recurrent architecture, <strong>SITH-RNN<\/strong>, revealed that it learned faster with orders-of-magnitude fewer parameters and demonstrated <strong>zero-shot generalization<\/strong> to out-of-distribution timescales. These results suggest that the brain employs scale-invariant, sequential priors to encode 'what' happened 'when', making recurrent networks with such priors particularly well-suited to describe human cognition.<\/p>\n<\/blockquote>\n<\/div>\n","protected":false},"excerpt":{"rendered":"<p>\u5fc3\u96fb\u56f3(ECG)\u306b\u3088\u308b\u8a8d\u77e5\u8ca0\u8377\u63a8\u5b9a\u3001\u5cf6\u76ae\u8cea\u6d3b\u52d5\u306b\u3088\u308b\u8868\u60c5\u8a8d\u8b58\u3001\u751f\u7269\u5b66\u7684\u306b\u5236\u7d04\u3055\u308c\u305f\u6df1\u5c64\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u306e\u6642\u9593\u30b9\u30b1\u30fc\u30eb\u6c4e\u5316\u306b\u95a2\u3059\u308b\u6700\u65b0AI\u7814\u7a76\u3092\u8981\u7d04\u3002<\/p>\n","protected":false},"author":1,"featured_media":852,"comment_status":"","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"vkexunit_cta_each_option":"","footnotes":""},"categories":[3],"tags":[8,94,64,96,34,47,93,15,95,63],"class_list":{"0":"post-892","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","6":"hentry","7":"category-columns","8":"tag-ai","9":"tag-ecg","10":"tag-64","11":"tag-96","12":"tag-34","13":"tag-47","14":"tag-93","16":"tag-95","17":"tag-63"},"_links":{"self":[{"href":"https:\/\/itexplore.org\/jp\/wp-json\/wp\/v2\/posts\/892","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/itexplore.org\/jp\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/itexplore.org\/jp\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/itexplore.org\/jp\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/itexplore.org\/jp\/wp-json\/wp\/v2\/comments?post=892"}],"version-history":[{"count":0,"href":"https:\/\/itexplore.org\/jp\/wp-json\/wp\/v2\/posts\/892\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/itexplore.org\/jp\/wp-json\/wp\/v2\/media\/852"}],"wp:attachment":[{"href":"https:\/\/itexplore.org\/jp\/wp-json\/wp\/v2\/media?parent=892"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/itexplore.org\/jp\/wp-json\/wp\/v2\/categories?post=892"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/itexplore.org\/jp\/wp-json\/wp\/v2\/tags?post=892"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}