LSTM - Recurrent Neural Network Model to Forecast the Risk Factors in Recurrent Cervical Carcinoma

Cervical carcinoma is a healthcare issue globally, and despite advancements in prevention, diagnosis and treatment, recurrence remains as a challenge. Understanding the risk factors associated with cervical cancer recurrence is crucial for optimizing patient care and improving outcomes. Multiple factors contribute to the possibility of cervical carcinoma metastasis. The important prognostic factor is the stage of the disease at diagnosis, with advanced stages are associated with a higher likelihood of recurrence. Other clinical chronic attributes like size of a cyst, inclusion of lymph nodes and what’s the kind of histology is also influence recurrence risk. Human papillomavirus (HPV) infection, particularly with HPV that poses a danger genotypes, danger element for the emergence of cervical cancer. HPV persistence following initial treatment is strongly associated with increased recurrence rates. The risk factors of recurrence cervical cancer is found by Long Short Term Memory which is advancement in recurrent neural network classifier. Gated Recurrent Unit (GRU) and Deep Neural Network (DNN) classifiers are compared with LSTM. In which LSTM Model attains the highest accuracy. In conclusion, cervical cancer recurrence is influenced by an intricate connection amongst standard of living, molecular and clinical variables. These risk factors is important for forecasting individuals at higher risk of recurrence and tailoring personalized treatment strategies.