Content-based microscopic image retrieval system for multi-image queries.
Content-based microscopic image retrieval system for multi-image queries.
复制标题
DOI：
10.1109/titb.2012.2185829
复制
发表时间：
2012-07
期刊：
影响因子：
--
通讯作者：
Gurcan MN
中科院分区：
文献类型：
--
作者：
Akakin HC;Gurcan MN
In this paper, we describe the design and development of a multi-tiered CBIR system for microscopic images utilizing a reference database that contains images of more than one disease. Proposed CBIR system uses a multi-tiered approach to classify and retrieve microscopic images involving their specific subtypes which are mostly difficult to discriminate and classify. This system enables both multi-image query and slide-level image retrieval in order to protect the semantic consistency among the retrieved images. New weighting terms, inspired from information retrieval (IR) theory, are defined for multiple-image query and retrieval. Performance of the system was tested on a dataset including 1666 imaged high power fields (HPF) extracted from 57 Follicular Lymphoma (FL) tissue slides with three subtypes and 44 Neuroblastoma (NB) tissue slides with four subtypes, where each slide is semantically annotated according to their subtypes by expert pathologists. By using leave-one-slide out testing scheme, the multi-image query algorithm with the proposed weighting strategy achieves about 93% and 86% of average classification accuracy at the first rank retrieval, outperforming the image-level retrieval accuracy by about 38 and 26 percentage points, for FL and NB diseases, respectively.
