3-DIMENSIONAL AUTOMATIC SEGMENTATION OF PITUARITY TUMOR USING DEEP LEARNING

Abstract

The development of a benign pituitary tumor progresses very slowly. Due to this slow development, it may take time to diagnose the patient. The Tumor that will form in the Pituitary Gland, which is effective in the secretion of many hormones and located behind the optic nerves, may cover 2/3 of the Pituitary Gland. In people for whom hormonal balance is essential, due to Pituitary Tumor, Cushing's syndrome diseases can be seen as a result of irregular menstruation, visual disturbances, headache, imbalance in breast milk production, and excess ACTH production. Excess ACTH can lead to excessive weight gain, the appearance of fragile bone structure, skin scars, and emotional changes. The Pituitary Tumor is located in the deepest part of the brain, and it is tough to perform a surgical operation there. Semantic segmentation using deep learning techniques can be successful. With our study, automatic segmentation of the Tumor with an IoU score of up to 98% was possible. This success is relatively high, and promises hope for the CAD system to be created for Pulmonary tumors. The 3D-Unet technique developed recently, can perform automatic segmentation in 3 dimensions. This study aims to automatically segment a Pituitary Tumor, which requires a complex operation, with the 3D-Unet model.