• SUJITH M
• Mrs. S. KALAIVANI

– According to the most current Global Cancer Statistics 2018 reports, lung tumours are the fifth most common type of tumour in males and the ninth most common type in women. After collecting a sample from the lung tissue, imaging tests including computed tomography (CT), magnetic resonance imaging (MRI), and ultrasound can be used to identify the lung tumour. These process is Costly and time-consuming for these tests implementation. This study suggested that deep learning Convolutional Neural Networks (CNNs) RCNN model image processing could replace traditional approaches for early tumour diagnosis. The two Cascaded CNNs have been primarily utilised in the current investigations for the segmentation and evaluation of the lung's Area Of Interest (ROI). RCNN, an enhanced version of U-Net and RCNN Models that makes advantage of the service of Residential blocks, is used in this study. For the 3D-IRCADb01 dataset, which is based on CT slices of patients with lung tumours, we apply the above approach. The results showed that the F1 score performance was approximately 95%, and the True Value Accuracy was around 99%. This technique will help in the early and precise diagnosis of lung tumours in the field of biotechnology, perhaps saving many patients' lives.

"LUNGS SCREEN: DETECTION AND CLASSIFICATION OF LUNGS NODULES USING IMPROVED R-CNN", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 4, page no.a610-a618, April-2023, Available :http://www.jetir.org/papers/JETIR2304083.pdf

"LUNGS SCREEN: DETECTION AND CLASSIFICATION OF LUNGS NODULES USING IMPROVED R-CNN", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 4, page no. ppa610-a618, April-2023, Available at : http://www.jetir.org/papers/JETIR2304083.pdf