{"title":"Multi-level feature extraction model for high dimensional medical image features","authors":"M. Saad, M. Mohsin, H. Hamid, Z. Muda","doi":"10.1109/AiDAS47888.2019.8970698","DOIUrl":null,"url":null,"abstract":"Recent technology evolution has emerged many applications that consumed data in extremely highly dimensional. For medical images, outsourcing the computation of image feature extraction to the cloud has become common method in order to alleviate the heavy computation workload for local devices. However, unlike other images, the medical images content cannot be easily manipulated because they exist in visual presentation that cannot be explored with textual data in order to capture the visual structure of the image. Hence, appropriate features are required to classify these images. Feature extraction for medical images based on image shape, color and texture using machine learning can improve the performance to categorize image features into homogeneous group. Feature extraction automatically learn and recognize complex patterns and make intelligent decisions based on features attributes. Therefore, this proposed a multi-level feature extraction model for high dimensional medical image features. By applying the multi-level model, features from medical images are extracted from general image features into specific features category. Later, a specified features categories are assigned to the image so that the image presentation can become more meaningful and assist the performance of image classification. We expect the findings derived from our method provides new approaches for extracting medical image features from big data source. It also improve the relevance and quality of image classification, thus enhance performance of medical imaging in the radiology service.","PeriodicalId":227508,"journal":{"name":"2019 1st International Conference on Artificial Intelligence and Data Sciences (AiDAS)","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2019 1st International Conference on Artificial Intelligence and Data Sciences (AiDAS)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/AiDAS47888.2019.8970698","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1
Abstract
Recent technology evolution has emerged many applications that consumed data in extremely highly dimensional. For medical images, outsourcing the computation of image feature extraction to the cloud has become common method in order to alleviate the heavy computation workload for local devices. However, unlike other images, the medical images content cannot be easily manipulated because they exist in visual presentation that cannot be explored with textual data in order to capture the visual structure of the image. Hence, appropriate features are required to classify these images. Feature extraction for medical images based on image shape, color and texture using machine learning can improve the performance to categorize image features into homogeneous group. Feature extraction automatically learn and recognize complex patterns and make intelligent decisions based on features attributes. Therefore, this proposed a multi-level feature extraction model for high dimensional medical image features. By applying the multi-level model, features from medical images are extracted from general image features into specific features category. Later, a specified features categories are assigned to the image so that the image presentation can become more meaningful and assist the performance of image classification. We expect the findings derived from our method provides new approaches for extracting medical image features from big data source. It also improve the relevance and quality of image classification, thus enhance performance of medical imaging in the radiology service.