An Analysis on the Integration of Machine Learning and Advanced Imaging Technologies for Predicting the Liver Cancer

This research analyzed the accuracy of machine learning models in classifying liver cancer based on CT and MRI scans. A dataset consisting of 2334 images of benign and malignant liver diseases is used. The authors perform a complete preprocessing pipeline that includes normalization, noise reduction...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2024 4th International Conference on Pervasive Computing and Social Networking (ICPCSN) s. 1082 - 1086
Hlavní autoři: Kelagadi, Hemantaraj M, Kumar K, Amit, D, Anandan, Vishnu Raja, P., Senthilkumar, G., L, Natrayan
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 03.05.2024
Témata:
Accuracy
Analytical models
Artificial Neural Networks
Computed tomography
e-healthcare
IoT
Liver cancer
Liver diseases
machine learning
Magnetic resonance imaging
patient monitoring
predictive healthcare
sensors
Visualization
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:This research analyzed the accuracy of machine learning models in classifying liver cancer based on CT and MRI scans. A dataset consisting of 2334 images of benign and malignant liver diseases is used. The authors perform a complete preprocessing pipeline that includes normalization, noise reduction, contrasting, and artifact removal. Image feature extraction uses traditional techniques like summing or deep learning algorithms. CNN architecture is central for liver condition classification. Specific convolutional networks included VGG16, ResNet50, and MobileNet. All of the models manifested considerable accuracy. VGG16 performed the best mechanically with an accuracy of 89.2 percent. The confusion matrices help visualize the models' abilities for correctly diagnosing liver disease states, even though there were some mistakes. Overall, the research emphasizes the importance and potential of advanced imaging technology combined with machine learning methods in early detection and diagnosis of liver cancer. The results have profound implications for patient care patients because they hint at new methods and machinery capable of earlier detection of cancers. Thus, technologies like those used in this experiment highlight the great promise shown by higher-tech methods. These results hold broad implications for thus improving the treatment of liver diseases by greater medical knowledge and technology. These results will require more rigorous research to validate them and continue developing medical imaging technology for liver cancers.
DOI:10.1109/ICPCSN62568.2024.00180