Investigation of viewing and answering approaches of questionnaire in detecting abnormalities in intra-oral dental radiographs

2014 
One of medical images problems are low image quality due to various factors such as acquisition process, radiographer skill, low contrast images and viewing condition. Thus the implementation of contrast enhancement algorithms (CEAs) to visually improve the image quality is an acceptable practice in diagnosing medical images. Since medical image diagnosis involves various diseases, the improvement of image quality become complex and the involvement of medical experts in evaluating the images become vital. However, to get a medical officer to participate in image quality evaluation for research purposes is challenging. One of the factors is their busy work scheduling. Thus this work conducts experiments to investigate the most effective approach relating to questionnaire answering and viewing approaches by dentists in evaluating the images. Medical images used are original and enhanced of intra-oral dental radiographs. The investigation is on the effect of how questionnaire answering and viewing condition approach give effect on dentists’ evaluation in term of detecting the abnormalities. The parameters of investigation are; supervised twin-view approach, unsupervised random approach and supervised random approach. The supervised approach is where the dentist answers the questionnaire with an assistant at the predefined time of two hours. The unsupervised approach is when the dentist answer the questionnaire at his/her own predefined time in three month duration. The viewing condition approach compare between twin-view versus random approach. Twin view method is about the arrangement of the images side by side between original and enhanced images. Random approach is where the images are arranged randomly. Results show that random approach of supervised method able to champion the investigation. © 2014 The Authors. Published by Global Illuminators Publishing. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/) Peer-review under responsibility of the Scientific & Review committee of ICMRP-2014.
    • Correction
    • Cite
    • Save
    • Machine Reading By IdeaReader
    17
    References
    1
    Citations
    NaN
    KQI
    []