International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 11 Issue: 09 | Sep 2024
p-ISSN: 2395-0072
www.irjet.net
Deep Learning for Pulmonary Diseases Detection Using Chest X-Ray Aditya Ingole1, Yuvraj Patil1, Yashraj Wawkar1, Aboli Deole2 1Student, Dept. of Artificial Intelligence and Machine Learning, PES’s Modern College of Engineering,
Pune, Maharashtra, India
2Assistant Professor, Dept. of Artificial Intelligence and Machine Learning,
PES’s Modern College of Engineering, Pune-05, Maharashtra, India. -----------------------------------------------------------------------***------------------------------------------------------------------------Abstract - Pulmonary illnesses pose an enormous healthcare challenge globally, necessitating accurate and well-timed prognoses for effective remedies. Deep knowledge of Pulmonary ailment Detection using Chest X-rays provides a progressive strategy to decorate diagnostic accuracy and performance. Leveraging deep neural networks and a carefully curated dataset of chest X-ray photos, this assignment aims to automate the identification of pulmonary illnesses consisting of pneumonia, tuberculosis, emphysema and many more. The deep mastering version, educated and first-rate-tuned on this dataset, offers the potential to not only most effectively detect illnesses with high precision but additionally help healthcare specialists in early diagnosis, in the end enhancing patient results. The challenge's multifaceted technique consists of records preprocessing, model choice and training, interpretability, deployment in a scientific place, and non-stop collaboration with medical examiners to ensure both technological robustness and ethical compliance. As pulmonary disorder detection and healthcare technologies hold to adapt, this mission stands at the leading edge of innovation, presenting a promising method to increase the abilities of healthcare practitioners and deliver extra timely and accurate diagnoses.
Keywords: Pulmonary diseases, healthcare, diagnosis, deep learning, chest X-ray, diagnostic accuracy, efficiency, deep neural networks
1. INTRODUCTION Nearly 545 million individuals currently live with a chronic respiratory condition, representing 7·4% of the world's population, which provides additional evidence of the large health contribution of chronic respiratory diseases to premature morbidity and mortality. Consequently, there has been substantial growth in the field of automated medical image classification. This endeavour seeks to categorise medical images into specified groups. Lately, Deep Learning (DL) has emerged as a prevalent and extensively applied technique for creating medical image classification tasks. Further, DL models produced more effective performance than traditional techniques using chest X-ray images from patients suffering from pulmonary diseases. The DL architectures illustrated effective predictive ability. On chest X-ray images, multiple tasks were performed on DL models, including tuberculosis identification, tuberculosis segmentation, large-scale recognition, and Radio-graph classification. The automated classification of chest X-ray images using DL models is growing rapidly and choosing an appropriate region of interest (ROI) on chest X-ray images was used to treat. Furthermore, applying the DL modes helps to avoid problems that take a long time to solve in traditional approaches. However, these models require large volumes of welllabelled training samples. The DL architectures illustrated effective predictive ability. On chest X-ray images, multiple tasks can be performed on DL models, including tuberculosis identification, tuberculosis segmentation, atelectasis, pneumonia, silico-tuberculosis, fibrosis, Emphysema Radiograph classification and many more. Many researchers have done investigations to relate machine learning schemes for prediction of X-ray image diagnostic information. With the control of computers along with the huge volume of records being unrestricted to the public, this is high time to resolve this complication. This solution can put up decreasing medical costs with the enlargement of computer science for health and medical science projects. For the implementation, the NIH chest X-ray image dataset is collected from Kaggle repository, and it is fully an open-source platform. A new hybrid algorithm is introduced in this paper and this algorithm is successfully applied on the above-mentioned dataset to classify lung disease. The main contribution of this research is the development of this new hybrid deep learning algorithm suitable for predicting lung disease from X-ray images.
© 2024, IRJET
|
Impact Factor value: 8.315
|
ISO 9001:2008 Certified Journal
|
Page 100