Robust Learning-Based Parsing and Annotation of Medical Radiographs

Abstract

In this paper, we propose a learning-based algorithm for automatic medical image annotation based on robust aggregation of learned local appearance cues, achieving high accuracy and robustness against severe diseases, imaging artifacts, occlusion, or missing data. The algorithm starts with a number of landmark detectors to collect local appearance cues throughout the image, which are subsequently verified by a group of learned sparse spatial configuration models. In most cases, a decision could already be made at this stage by simply aggregating the verified detections. For the remaining cases, an additional global appearance filtering step is employed to provide complementary information to make the final decision. This approach is evaluated on a large-scale chest radiograph view identification task, demonstrating a very high accuracy ( > 99.9%) for a posteroanterior/anteroposterior (PA-AP) and lateral view position identification task, compared with the recently reported large-scale result of only 98.2% (Luo, , 2006). Our approach also achieved the best accuracies for a three-class and a multiclass radiograph annotation task, when compared with other state of the art algorithms. Our algorithm was used to enhance advanced image visualization workflows by enabling content-sensitive hanging-protocols and auto-invocation of a computer aided detection algorithm for identified PA-AP chest images. Finally, we show that the same methodology could be utilized for several image parsing applications including anatomy/organ region of interest prediction and optimized image visualization.

Image-Centric Integrated Data Model of Medical Information by Diseases: Two Case Studies for AMI and Ischemic Stroke

“…Some previous works presented analysis or processing methods [5][6][7][8] and transfer or exchange systems [9,10] for medical images. However, since they have focused on a specific type of HISs or clinical materials, they did not consider the diversity and semantic relations of medical data in multi-systems.…”

“…Nowadays most of medical institutions like hospitals naturally use HIS. Recently, many studies [1][2][3][4][5][6][7][8][9][10] have tried to efficiently store, transfer, and manage clinical materials. However, HIS consists of several systems such as Order Communication System (OCS), Electronic Medical Record (EMR), and Picture Archiving and Communication System (PACS) and they generate clinical materials of different types.…”

Analysis of Semantic Relations Between Multimodal Medical Images Based on Coronary Anatomy for Acute Myocardial Infarction

“…Nowadays, most of the works related to PACS are focused on techniques that analyze and process the medical image itself [6,7] or technology that can quickly transfer or exchange the medical image data stored in PACS [8,9]. However, it is important to effectively provide core information about the medical images.…”

“…A number of software tools for processing medical images have been developed based on PACS. In addition, it has been studied on analyzing or processing medical images at a low level [6,7], transmitting or exchanging the file in PACS [8,9]. However, most of the studies on PACS have only focused on processing and transmitting the medical images.…”

Parsing radiographs by integrating landmark set detection and multi-object active appearance models

“…Landmark detection has been used to previously to parse radiographs 1 however false negatives are not inferred nor are the detections refined with an active appearance model. We show that these steps substantially improve accuracy.…”