2015/08/31 by Devinder Kumar, Mohammad Javad Shafiee, Kumar, Devinder +9 · 1 citation
Computer Science · Engineering · Medicine · #AI in cancer detection #Advanced X-ray and CT Imaging #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Lung Cancer Diagnosis and Treatment #Radiomics and Machine Learning in Medical Imaging
paper · pdf · doi:10.48550/arxiv.1509.00117
openalex publication_date 2015/08/31 · openalex created_date 2022/10/01 · openalex updated_date 2026/07/28
Lung cancer is the leading cause for cancer related deaths. As such, there is\nan urgent need for a streamlined process that can allow radiologists to provide\ndiagnosis with greater efficiency and accuracy. A powerful tool to do this is\nradiomics: a high-dimension imaging feature set. In this study, we take the\nidea of radiomics one step further by introducing the concept of discovery\nradiomics for lung cancer prediction using CT imaging data. In this study, we\nrealize these custom radiomic sequencers as deep convolutional sequencers using\na deep convolutional neural network learning architecture. To illustrate the\nprognostic power and effectiveness of the radiomic sequences produced by the\ndiscovered sequencer, we perform cancer prediction between malignant and benign\nlesions from 97 patients using the pathologically-proven diagnostic data from\nthe LIDC-IDRI dataset. Using the clinically provided pathologically-proven data\nas ground truth, the proposed framework provided an average accuracy of 77.52%\nvia 10-fold cross-validation with a sensitivity of 79.06% and specificity of\n76.11%, surpassing the state-of-the art method.\n