Detection of Pancreatic Cancer Cells at Early Stage and Recommend the Treatment Process

Early identification and diagnosis of cancerous cells will focus on the symptomatic patients, to give the best way of successful treatment. When the detection of cancer is delayed, then there is a lower chance of survival, high problems with treatment and higher cost estimates. Considering Pancreas Cancer growth, a type of cancer that is difficult to identify because there are less or no symptoms found at the early stage. So, we introduce Artificial Intelligence (AI) technology to identify pancreatic cancer cells. This is the first insight that is provided to detect cancerous cells at the early stage. With the use of technologies, anything is possible, so we use the process of medical imaging analysis, data processing and Machine Learning (ML) for the recognition. Data is collected through MRI and CT scan reports, biomarkers tests, liquid biopsies, endoscopic techniques, early symptom monitoring, and genetic testing which is then categorized into train and test sets. Every training data set is analyzed to provide a standard image to extract the features with the use of OpenCV. Building an AI model with the use of the Convolutional Neural Network (CNN) algorithm with an efficient set of filters. Training the model with the training data set helps to improve the accuracy of predicting results. When it is ready, the testing data set is used to check the accuracy of predictions. Finally, by assessing the model based on the performance metrics from both training and testing datasets, accuracy is achieved. After achieving a certain level of accuracy, and leveraging ML algorithms such as Hybrid models, the system can provide treatment recommendations consistent with a separate training dataset. This training dataset comprises predicted treatments corresponding to the achieved accuracy. After, the system is prepared for deployment. The envisioned process can track the patient's status for future use.