Deep Learning-Based CT Imaging for the Diagnosis of Liver Tumor

Abstract

The objective of this research was to investigate the application value of deep learning-based computed tomography (CT) images in the diagnosis of liver tumors. Fifty-eight patients with liver tumors were selected, and their CT images were segmented using a convolutional neural network (CNN) algorithm. The segmentation results were quantitatively evaluated using the Dice similarity coefficient (DSC), precision, and recall. All the patients were examined and diagnosed by CT enhanced delayed scan technique, and the CT scan results were compared with the pathological findings. The results showed that the DSC, precision, and recall of the CNN algorithm reached 0.987, 0.967, and 0.954, respectively. The images segmented by the CNN were clearer. The diagnostic result of the examination on 56 cases by CT enhanced delay scanning was consistent with that of pathological diagnosis. According to the result of pathological diagnosis, there were 6 cases with hepatic cyst, 9 with hepatic hemangioma, 12 cases with liver metastasis, 10 cases with hepatoblastoma, 3 cases with focal nodular hyperplasia, and 18 cases with primary liver cancer. The result of CT enhanced delay scanning on 58 patients was consistent with that of pathological diagnosis, and the total diagnostic coincidence rate reached 96.55%. In conclusion, the CNN algorithm can perform accurate and efficient segmentation, with high resolution, providing a more scientific basis for the segmentation of liver tumors in CT images. CT enhanced scanning technology has a good effect on the diagnosis and differentiation of liver tumor patients, with high diagnostic coincidence rate. It has important value for the diagnosis of liver tumor and is worthy of clinical application.

An Image Classification and Retrieval Hybrid Model for Larger Healthcare Datasets using Deep Learning

“…The proposed method's classification accuracy is compared to existing strategies to determine its effectiveness in medical picture retrieval. The existing methods used for discussions is VGG16 (1) , KNN's with GLCM (2) , CNN (18) , DeepSVM (19) etc. The Table 4 shows the comparison of the proposed model and existing work.…”

“…Dataset Accuracy (%) VGG16 (1) Computed tomography (CT) 97.6 KNN's With GLCM (2) Chest CT images of COVID 98.9 Sparse Auto Encoder based DNN (8) OASIS 95.34 ICNN (9) Pap smear dataset 98.88 HoG-LTP method with CNN-Classifier (10) CE-MRI 98.8 GIPBT + DKD (11) Medical Image Set 98.4 CNN (18) liver tumors CT images 96.55 CNN-based deep learning techniques -ResNet-18 (21) GPD data set 96.21 DeepSVM (19) NCT-CRC-HE-100 K 98.75 Hybrid models (CNN, VGG16 and VGG19) (16) diabetic retinopathy (DR) 90.6 CNN-sequential model (20) Chest X-ray (1000 images) 98.437 Proposed Hybrid Model CNN-LSTM Medical Image Dataset 99. 4 The performance of ANN, CNN and Hybrid Model for the medical dataset considered in this work shown in Figure 6, the performance of proposed ensemble method provides more accuracy in image classification compared with ANN and CNN, because of more dense layers of CNN and additional layers of LSTM, also accurate feature extraction with GLCM.…”

“…4 The performance of ANN, CNN and Hybrid Model for the medical dataset considered in this work shown in Figure 6, the performance of proposed ensemble method provides more accuracy in image classification compared with ANN and CNN, because of more dense layers of CNN and additional layers of LSTM, also accurate feature extraction with GLCM. The existing works in (1,2,(9)(10)(11)16,(18)(19)(20)(21) shown in Table 4, used various classification and features extraction methods. Compared with the existing works ours proposed Hybrid model provides better classification accuracy of 99.4% and also the retrieval probability score also calculated for various classes of medical images as shown in Table 3.…”

“…Heng Zhang et al (18) in this work, CT images of patients with liver tumors were feature segmented using CNN. It proved to be precise and effective in segmenting images.…”

“…A variety of sets of images and datasets in existing techniques perform medical image retrieval such as, Computed tomography (CT) (1) , Chest CT images of COVID (2) , CE-MRI and OASIS dataset (8) , CE-MRI (10) , Diagnosis of Liver Tumour (18) , CRC-5000, CE-MRI, Diabetic Retinopathy (DR) (16) , NCT-CRC-HE-100 K (19) , X-ray images and Computerized Tomography (CT) scans, including pneumonia, tuberculosis, and lung cancer (20) etc., are used in the current approaches to perform medical image retrieval. The following are the primary contributions of this study extracting additional features in terms of parameters such as distance, angle, resolution, level, and symmetric values.…”