Browsing by Department "ABES Engineering College"
  • Institution Publication
    Smart Investigations into the Development of an Effective Computer-Assisted Diagnosis System for CT Scan Brain Depictions
    ( 2023-09-01)
    Attuluri S.
    ;
    Bhupati C.
    ;
    Ramya L.
    ;
    Tiwari A.
    ;
    Budaraju R.R.
    ;
    Cotrina-Aliaga J.C.
    Recent advances in medical image analysis on computers are anticipated to help radiologists and other healthcare workers with numerous diagnostic tasks including medical image interpretation. Accurate diagnosis and/or assessment of a condition in medical imaging relies on both the quality of the acquired images and the quality of the interpretation of those images. To accomplish this, large amounts of picture data and medical records must be combined. Computer-aided diagnosis (CAD) systems have been developed in response to a lack of accuracy to boost the radiologist's productivity and precision in their interpretations. Given the importance of computerised tomography (CT) imaging in this field, we have made an effort in our research to examine CT scan brain images by applying a number of feature extraction and selection techniques, as well as classification techniques, to diagnose different types of brain disorders. Brain CT scans are analysed here and classified as either normal, benign tumours, or malignant tumours. Finding the best characteristics to use in a classification system is called “feature selection,” and it requires sifting through a vast amount of extracted features to locate the most relevant ones. To evaluate the efficacy of the implemented classifiers, we used measures of accuracy, specificity, sensitivity, positive prediction value, and negative prediction value. Traditional classifiers are also examined alongside the suggested method's results. The proposed decision support systems outperform the standard classifiers in terms of accuracy. An FSVM is generated by applying the RBF kernel function to this dataset. This method is compared to the support vector machine (SVM) and the multi-layer perceptron neural network (MLPNN) in terms of accuracy, sensitivity, and specificity of the classifiers they produce. Accuracy (96.25%), sensitivity (96.67%), and specificity (95.83%) are all significantly higher when using the proposed method as compared to the control methods.