Enhanced deep learning based on Fusion data to diagnosis malignancy Thyroid tumour
Bennadji Ziad, Labib Sadek Terrissa, Karima Benmohammed, Noureddine Zerhouni · Advances in intelligent systems research/Advances in Intelligent Systems Research · 2024
The predominant type of cancer within the endocrine system is Thyroid Cancer (TC), with the majority falling under the category of low-risk tumour.However, the over-diagnosis and over-treatment of such conditions serve as primary factors contributing to a patient's deteriorating state, heightening the risk of recurrence and potentially complicating future interventions.Consequently, these practices elevate mortality rates and hinder complete recovery.Our paper focuses on developing a robust neural network model that integrates ultrasound radiomics with clinical data to accurately diagnose malignant thyroid tumours, aiming to mitigate issues associated with misdiagnosis and over-diagnosis.Based on independent cohort testing, the model demonstrates outstanding performance metrics with values of 0.97, 0.99, 0.97, and 0.98 for accuracy, AUC, precision, and recall, respectively.