会员体验
专利管家(专利管理)
工作空间(专利管理)
风险监控(情报监控)
数据分析(专利分析)
侵权分析(诉讼无效)
联系我们
交流群
官方交流:
QQ群: 891211   
微信请扫码    >>>
现在联系顾问~
热词
    • 6. 发明授权
    • Module for detecting poorly differentiated cancers and pathological image diagnosis support apparatus, program and recording medium including the same
    • 用于检测分化不良的癌症和病理图像诊断支持装置,程序和记录介质的模块
    • US08199998B2
    • 2012-06-12
    • US12265292
    • 2008-11-05
    • Kenji OkajimaYoshiko YamashitaAkira Saito
    • Kenji OkajimaYoshiko YamashitaAkira Saito
    • G06K9/00
    • G06K9/00127
    • The detection accuracy of poorly differentiated cancers in adenocarcinoma is improved by restricting false detection. Cell nucleus detection means 1 receives a digitized pathological image as an input and extracts the region of a cell nucleus therefrom. Gland duct detection means 2 detects a gland duct structure in the image. Poorly differentiated cancer detection means 4 detects poorly differentiated cancers only in the region other than the gland duct region. False detection rejection means 7 compares the detection density of poorly differentiated cancer in the vicinity of a detection point with a threshold that is predetermined depending on gland duct density in the vicinity of the detection point, at each detection point detected by poorly differentiated cancer detection means 4 and rejects the detection point as a false detection if the detection density of a poorly differentiated cancer is smaller than the threshold.
    • 通过限制假检测可以提高腺癌分化不良的癌症的检测准确性。 细胞核检测装置1接收数字化的病理图像作为输入,并从中提取细胞核的区域。 腺体管检测装置2检测图像中的腺体管结构。 分化不良的癌症检测手段4仅在腺体管区域以外的区域检测分化不良的癌症。 假检测抑制装置7将检测点附近的分化不良的癌症的检测密度与根据检测点附近的腺体密度预先确定的阈值,在低分辨率癌症检测手段检测的各检测点进行比较 4,如果分化不良的癌症的检测密度小于阈值,则将检测点拒绝为假检测。
    • 10. 发明申请
    • Pattern feature selection method, classification method, judgment method, program, and device
    • 模式特征选择方法,分类方法,判断方法,程序和装置
    • US20050169516A1
    • 2005-08-04
    • US10505903
    • 2003-02-27
    • Kenji OkajimaHitoshi ImaokaMasanobu Miyasita
    • Kenji OkajimaHitoshi ImaokaMasanobu Miyasita
    • G06F17/30G06K9/62G06N3/00G06T7/00
    • G06K9/623
    • Feature decision means (303) decides a set of features appropriate for pattern identification from a plenty of feature candidates generated by feature candidate generation means (302) by using learning patterns stored in learning, pattern storage means (301). The feature decision means (303) successively decides features according to a reference of information maximization under the condition that the decided feature is known while adding an effective noise to the learning pattern and performs information amount calculation approximately and at a high speed while merging the learning patterns into a set of N elements when required. As a result, it is possible to automatically create a feature set appropriate for pattern identification of a high performance without requiring enormous learning. Moreover, by using a transition table (305) containing transitions between sets, it is possible to perform pattern judgment with a high efficiency.
    • 特征决定装置(303)通过使用存储在学习模式存储装置(301)中的学习模式,从特征候选生成装置(302)产生的大量特征候选中,确定适合于模式识别的特征的一组。 特征决定装置(303)在所确定的特征已知的条件下,根据信息最大化的参考依次确定特征,同时向学习模式添加有效噪声,并且在合并学习期间大约和高速地执行信息量计算 当需要时,模式成为一组N个元素。 因此,可以自动创建适合于高性能的图案识别的特征集,而不需要巨大的学习。 此外,通过使用包含组之间的转换的转换表(305),可以以高效率执行模式判断。