Diagnosis of Pulmonary Nodules on CT Images Using YOLOv4

Shital Bhatt; Himanshu Soni; Tanmay Pawar; Heena Kher

doi:10.3991/ijoe.v18i05.29529

Authors

Shital D. Bhatt Gujarat Technological University, Ahmedabad, Gujarat, India
Dr. Himanshu B. Soni Gujarat Technological University, Ahmedabad, Gujarat, India
Dr. Tanmay D. Pawar Department of Electronics, Birla Vishvakarma Mahavidyalaya, V.V. Nagar, Anand, India
Dr. Heena R. Kher Department of Electronics & Communication, A.D. Patel Institute of Technology, New V.V. Nagar, Anand, India

DOI:

https://doi.org/10.3991/ijoe.v18i05.29529

Keywords:

Fast Library for Approximate Nearest Neighbors (FLANN), Pulmonary Nodules, Scale-Invariant Feature Transform (SIFT), You Only Look Once (YOLO)

Abstract

In this paper, the Scale-Invariant Feature Transform (SIFT) and Fast Library for Approximate Nearest Neighbors (FLANN) based algorithm is used to detect the abnormalities in the National Lung Screening Trial (NLST) CT scans as the exact clinical nodule locations are not provided in the dataset. These identified nodules on NLST CT Scans are then annotated using LabelImg tool. This process consumes time and so furthermore, the automatic nodule detection, You Only Look Once version 4 (YOLOv4) object detection model is implemented. The YOLOv4 object detection model is provided with total of 4187 labelled images in form of training (70%), validation (20%), and test (10%) datasets. Our YOLOv4 model achieves precision of 95%, sensitivity of 81% and mean Average Precision (mAP) of 89.1%.

Author Biographies

Shital D. Bhatt, Gujarat Technological University, Ahmedabad, Gujarat, India

Shital Bhatt received the B.E. degree in Electronics & Communication Engineering from Atmiya Institute of Technology, Saurashtra University, Gujarat, India in 2007, the M.E. degree in Communication Engineering from Chandubhai S Patel Institute of Technology, Changa, Gujarat, India in 2010. She is currently working toward the Ph.D. degree at Gujarat Technological University, Gujarat, India. She is working as an Assistant Professor at Madhuben & Bhanubhai Patel Institute of Technology since 2010. She has published various papers in journals and conferences at international and national levels. Her area of interest includes Wireless Communication, Image Processing, Machine Learning and Deep Learning.

Dr. Himanshu B. Soni, Gujarat Technological University, Ahmedabad, Gujarat, India

Himanshu Soni received the B.E. degree in Electronics & Communication Engineering from B.V.M. Engineering College, Sardar Patel University, Gujarat, India in 1997, the M.E. degree in Communication Engineering from Birla Institute of Technology and Science (BITS), Pilani, India in 2004. He obtained Ph. D from Indian Institute of Technology, Bombay, India in 2009. Currently he is working as a Professor in Electronics & Communication Department. and as a Principal of G H Patel college of Engineering & Technology, Gujarat, India. He has published more than 30 research papers in various international journals and conferences. He has been awarded Research Endowment Prize award: for his research paper published in EUSIPCO-Poland in year 2010 by Hari Ohm Ashram. He has attended various training programmes in area of communication engineering from India and abroad. His research interests include statistical signal processing, adaptive signal processing, wireless and mobile communication and UWB communication, 5G and IoT.

Dr. Tanmay D. Pawar, Department of Electronics, Birla Vishvakarma Mahavidyalaya, V.V. Nagar, Anand, India

Tanmay Pawar received the B.E. degree Electronics & Communication Engineering from Dharmsinh Desai Institute of Technology, Gujarat, India in 1996. He obtained the M.Tech, and Ph.D. from Indian Institute of Technology, Bombay, India in 2002 and 2008, respectively. He is a Professor and Head of Electronics Engineering Department at Birla Vishvakarma Mahavidyalaya (BVM), Vallabh Vidyanagar, Gujarat, India. He is doing his research activities at BVM and the Gujarat Technological University (GTU), Gandhinagar, Gujarat, India. His research interests include Signal Processing, Image Processing and Biomedical Signal Processing with an emphasis on machine learning.

Dr. Heena R. Kher, Department of Electronics & Communication, A.D. Patel Institute of Technology, New V.V. Nagar, Anand, India

Heena Kher received the B.E. degree in Instrumentation and Control from Sarvajanik College of Engineering and Technology, Surat, Gujarat, India in 2001, M.E. in Microprocessors Systems and Applications from M. S. University, Baroda, Gujarat, India in 2006 and Ph.D. from Sardar Patel University, Vallabh Vidyanagar, Gujarat, India in 2014. She is working as an Assistant Professor in A. D. Patel Institute of Technology, New Vallabh Vidyanagar, Gujarat, India. She has 18 years of teaching experience. She has published around 30 research papers in journals and conferences at international and national levels. She is a reviewer of many reputed journals. Her areas of interest are Digital Image Processing, Machine Learning, Deep Learning, Biomedical Signal Processing, Data Analytics, Optimization Techniques.