发明名称 Image diagnostic method, image diagnostic apparatus, and image diagnostic program
摘要 The image diagnostic apparatus includes: acquisition means 11 for acquiring data on an image having a first resolution, which is generated by imaging a diagnostic object; conversion means 12 for converting the resolution of the image into a second resolution which is lower than the first resolution; extraction means 13 for analyzing the image of which resolution has been converted into the second resolution and extracting an ROI of the diagnostic object; determination means 14 for determining whether or not a characteristic region which indicates a predetermined image characteristic exists in the ROI extracted by the extraction means 13; and creation means 15 for creating, when the characteristic region exists, an output image by converting the resolution so that the characteristic region is confined to an image size to be output, and occupies the image by a predetermined ratio or more.
申请公布号 US9014443(B2) 申请公布日期 2015.04.21
申请号 US201013520490 申请日期 2010.12.21
申请人 NEC Corporation 发明人 Marugame Atsushi
分类号 G06K9/00;G06T3/40 主分类号 G06K9/00
代理机构 Sughrue Mion, PLLC 代理人 Sughrue Mion, PLLC
主权项 1. An image diagnostic apparatus, comprising: an acquisition unit configured to acquire data of an image having a first resolution, the data generated by imaging a pathological diagnostic object; a conversion unit configured to convert the resolution of the image into a second resolution which is lower than the first resolution; an extraction unit configured to analyze the image of which resolution has been converted into the second resolution and to extract a region of interest (ROI) of the pathological diagnostic object; a determination unit configured to determine whether or not a characteristic region indicating a predetermined image characteristic to be a reference of pathological diagnosis exists in the ROI extracted by the extraction unit, when the ROI is extracted by the extraction unit; and a creation unit configured to create, when the characteristic region exists, an output image by converting the resolution so that a ratio of an area of the characteristic region in the output image to an area of the output image is at least a predetermined ratio and less than 100%, wherein when the ROI is not extracted from the image of which resolution has been converted into the second resolution by the extraction unit, the conversion unit is further configured to convert the resolution of the image into a third resolution which is less than or equal to the first resolution and higher than the second resolution, andthe extraction unit is further configured to analyze the image converted into the third resolution, and to extract the ROI of the diagnostic object.
地址 Tokyo JP