Differential diagnosis of Interstitial Lung Diseases using Deep Learning networks

ABSTRACT

An architecture for automatic lung tissue classification method based on the Deep Learning techniques is designed in this paper. Recent works on Deep Learning techniques achieved impressive results in the field of medical image classification. So, we designed a Convolution Neural Network (CNN) for the classification of five categories of Interstitial Lung Diseases (ILD) patterns in High-Resolution Computed Tomography (HRCT) images. The CNN consists of 3 Convolution layers, Leaky ReLU activation followed by Maximum pooling layer and dense layer. The last Fully Connected (FC) layer has 5 outputs equivalent to the classes considered such as Normal, Ground Glass (GG), Emphysema, Micro Nodules, and Fibrosis. The proposed CNN is trained and evaluated on the publicly available ILD database provided by the University Hospitals of Geneva (HUG). Experimental results are compared with the state-of-art, which shows an outstanding performance of the proposed CNN model giving 94.67% accuracy and 94.65% F_avg.

KEYWORDS:

• Interstitial Lung Disease (ILD)
• Convolution Neural Network (CNN)
• lung tissue classification
• High-Resolution Computed Tomography (HRCT)
• Deep Learning
• medical image classification‌
• University Hospitals of Geneva (HUG)‌
• accuracy‌

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes on contributors

Mrs V. N. Sukanya Doddavarapu is a Ph.D. student of Jawaharlal Nehru Technological University, Kakinada, and working as Assistant Professor of Electronics and Communication Department in St.Ann’s College of Engineering and Technology, Chirala. She received her Master’s degree in VLSI & ES in the year 2012 from Jawaharlal Nehru Technological University, Kakinada, and Bachelor's degree in Electronics and Communication Engineering in the year 2008 from Acharya Nagarjuna University, A.P., India. Her research interests focus on medical image processing, with specific emphasis on the classification of various lung tissue patterns from CT scan images.

Dr Giri Babu Kande is a Professor and Principal i/c of Electronics and Communication Department in Vasireddy Venkatadri Institute of Engineering and Technology, Guntur. He has teaching experience of about 21 years. His research interests include digital image processing, Machine Learning, and Computer Vision. He received his Ph.D. degree in Digital Image Processing from Jawaharlal Nehru Technological University, Hyderabad in the year 2010. About 2 scholars were awarded Ph.D. under his tutelage. He is a member of various professional chapters and published many research papers in various SCI journals, national and international conferences.

Dr B. Prabhakara Rao was born in the year 1955 at Buchireddypalem, Nellore Dist., A.P., India. He obtained B.Tech. & M.Tech. from S.V. University, Tirupathi with Specializations in Electronics and Communications Engineering (ECE), Electronic Instrumentation, and Communications Systems in the years 1979 and 1981 respectively. Dr Rao received Doctorate from Indian Institute of Science, Bangalore in the area of Sonar Signal processing in the year 1995. Dr Rao joined as Assistant Professor in the JNT University in the Year 1982 and became Professor in ECE Dept. in the year 2003. He worked as Director (Institute of Science & Technology), Rector, and Vice-Chancellor i/c at JNT University, Kakinada. His current areas of research interest are Optical, Wireless, Microwave and Digital Communications & Image processing. About 35 scholars were awarded Ph.D. under his tutelage. Rao published many research papers in various SCI journals and National and International conferences and is a fellow of many Professional Organizations. Presently he is Vice president of IETE, New Delhi.