Deep learning models on medical image analysis and processing

Health care institutions produce extensive heterogeneous data in different structures and sources daily; depending on this situation, with traditional methods, the foresight of interpreting and managing data in this structure can be reduced. It emerges as a powerful tool for managing, interpreting, and analyzing such data with machine learning and deep learning methods. The correct diagnosis of the disease depends on the correct analysis of the image data and the acquisition and interpretation of the appropriate visual data in the prediction. Radiological imaging devices have improved significantly in recent years. NAC chemotherapy, a treatment method for breast cancer cases discussed in the study, aims to predict patients' response to treatment and the disease's development process in pathological and radiological areas. With the developing X-Ray technology and MRI scans, high-resolution radiological images can be obtained. However, the automation and benefits of image interpretation have only just begun to be obtained. Classification performances of CNN and VGG-based proposed models for tumor status after NAC treatment has been evaluated in detail through MRI images, which are frequently used in the healthcare industry. The number of convolutional layers, data set quality, and the main criteria affecting the model's success during training have been evaluated. Since it can provide strong feature representation power, a higher detection rate has obtained with the YOLO model proposed in the study in object detection methods based on CNN models from deep neural networks. A user interface provided for radiological image analysis, pathological test results and interpretation of radiological images with deep learning methods provide clinicians with a solution in determining the correct diagnosis and treatment method with the prognosis follow-up of the patient.