專利授權區國立清華大學國際產學營運總中心 Operations Center for Industry Collaboration

搜尋專利授權區

關鍵字

» 新增關鍵字

選單

專利授權區

專利授權區
專利名稱(中)	寄生蟲卵判讀方法及寄生蟲卵判讀系統
專利名稱(英)	INTERPRETATION METHOD FOR PARASITE EGG AND INTERPRETATION SYSTEM FOR PARASITE EGG
專利家族	中華民國：I907243
專利權人	國立清華大學 50.00% ,高雄市立小港醫院 50.00%
發明人	蔡志強,洪志興,梁景超,陳奕儒,譚書旻,金愷崙,朱家杰,楊梵孛,蔡惠君,陳怡蘋,李秋慧,謝雅年,蔡金玲,吳芷綺
技術領域	生化醫藥,資訊工程

專利摘要(中)
本揭示內容提供一種寄生蟲卵判讀方法及寄生蟲卵判讀系統。所述方法包含拍攝一寄生蟲檢體以產生一電子影像；對電子影像進行切割以產生複數寄生蟲卵影像，並取得寄生蟲卵影像之每一者之一蟲卵資訊；透過一深度學習模型從寄生蟲卵影像之每一者擷取至少一生物特徵，並將生物特徵分類為複數特徵類別之至少一者；及根據蟲卵資訊及此些特徵類別對具有相同特徵類別之此些寄生蟲卵影像進行交叉比對，藉以產生此些寄生蟲卵影像之每一者的一寄生蟲卵判讀結果。藉此，可縮短人工閱片的判讀時間，進而提高檢出率與準確率。

專利摘要(中)

本揭示內容提供一種寄生蟲卵判讀方法及寄生蟲卵判讀系統。所述方法包含拍攝一寄生蟲檢體以產生一電子影像；對電子影像進行切割以產生複數寄生蟲卵影像，並取得寄生蟲卵影像之每一者之一蟲卵資訊；透過一深度學習模型從寄生蟲卵影像之每一者擷取至少一生物特徵，並將生物特徵分類為複數特徵類別之至少一者；及根據蟲卵資訊及此些特徵類別對具有相同特徵類別之此些寄生蟲卵影像進行交叉比對，藉以產生此些寄生蟲卵影像之每一者的一寄生蟲卵判讀結果。藉此，可縮短人工閱片的判讀時間，進而提高檢出率與準確率。

專利摘要(英)
An interpretation method for a parasite egg and an interpretation system for a parasite egg are proposed. The interpretation method includes photographing a parasite specimen to generate an electronic image; cutting the electronic image to generate a plurality of parasite egg images, obtaining egg information on each of the parasite egg images; extracting at least one biological feature from each of the parasite egg images by a deep learning model, and classifying the at least one biological feature into at least one of a plurality of feature categories; and cross-comparing the parasite egg images with the same characteristic category according to the egg information and the characteristic categories to generate a parasite egg interpretation result for each of the parasite egg images. Therefore, the interpretation time of manual reading can be shortened, thereby improving the detection rate and accuracy.

專利摘要(英)

An interpretation method for a parasite egg and an interpretation system for a parasite egg are proposed. The interpretation method includes photographing a parasite specimen to generate an electronic image; cutting the electronic image to generate a plurality of parasite egg images, obtaining egg information on each of the parasite egg images; extracting at least one biological feature from each of the parasite egg images by a deep learning model, and classifying the at least one biological feature into at least one of a plurality of feature categories; and cross-comparing the parasite egg images with the same characteristic category according to the egg information and the characteristic categories to generate a parasite egg interpretation result for each of the parasite egg images. Therefore, the interpretation time of manual reading can be shortened, thereby improving the detection rate and accuracy.

聯絡資訊
承辦人姓名	李曉琪
承辦人電話	03-5715131 #31061
承辦人Email	hsiaochi@mx.nthu.edu.tw