A Novel Long Short-Term Memory (LSTM) Deep Learning IoT Method for Lung Cancer Prediction and Detection
R. Ramani1*, Padmaja Nimmagadda2, Shruti Bhargava choubey3, S. Rajasekar4, Omega John Unogwu5,6, Abdel-Hameed Al-Mistarehi7, Mostafa Abotaleb8
1Department of Computer Science and Engineering, P.S.R Engineering College, Sivakasi, Tamil Nadu-626140, India.
5Professor of Electronics and Communication Engineering, Mohan Babu University, (Erstwhile Sree Vidyanikethan Engineering College), Tirupati, Andhra Pradesh-517102, India,
3Dean- Innovation, Department of Electronics and Communication Engineering, Sreenidhi Institute of Science & Technology, Hyderabad-501301, Telangana, India.
4 Department of Computer Science and Engineering, Bannari Amman Institute of technology, Sathyamangalam, Tamil Nadu 638401.
5Space Geodesy and Systems Division, Centre for Geodesy and Geodynamics, National Space Research and Development Agency, Nigeria
6Department of Computer Science and Engineering, Universidad Azteca, Chalco, Mexico
7School of Medicine, Johns Hopkins University, Baltimore, Maryland, USA
8Department of System Programming, South Ural State University, Chelyabinsk, Russia
Email:rramani.ananth@gmail.com; padmaja.n@vidyanikethan.edu; Shrutibhargava@sreenidhi.edu.in; ssrajasekar80@gmail.com; unogwuomega@gmail.com; aalmist1@jh.edu; abotalebmostafa@bk.ru
Abstract
Lung cancer is the primary cause of cancer-related mortality in this generation, and it is expected to stay in foreseeable future. When the early indications of lung cancer are identified, a successful treatment can be initiated. A prototype environment friendly approach for treating lung cancer might be developed using the most recent developments in computational intelligence. Time and money will be saved since fewer resources will be wasted and manual tasks will take less effort to complete. An LSTM (Long Short-Term Memory)-based learning model was used to predict the lung cancer and improve the dataset procedure. With applications across medical image-based and textural data modalities, deep learning is one of the areas of medical imaging that is growing the fastest. Physicians may more easily and reliably identify and classify lung nodules with help of Deep Learning (DL)-based medical imaging technologies. This system covers the most recent advancements in deep learning-based imaging approaches for the early identification of lung cancer. The LSTM classifier sensitivity, specificity, and accuracy of our suggested system are best achieved by the Python software, with values of 80%, 85%, and 95%, respectively. Additionally, IoT (internet of things) to monitoring the lung cancer through cloud system through Adafruit Io. The lung cancer level is updating to NodeMCU controller.
Keywords: deep learning, LSTM; lung cancer; sensitivity; specificity and accuracy