Can AI Help in Screening Viral and COVID-19 Pneumonia?

IEEE Access, Journal Year: 2020, Volume and Issue: 8, P. 132665 - 132676

Published: Jan. 1, 2020

Coronavirus disease (COVID-19) is a pandemic disease, which has already caused thousands of causalities and infected several millions people worldwide. Any technological tool enabling rapid screening the COVID-19 infection with high accuracy can be crucially helpful to healthcare professionals. The main clinical currently in use for diagnosis Reverse transcription polymerase chain reaction (RT-PCR), expensive, less-sensitive requires specialized medical personnel. X-ray imaging an easily accessible that excellent alternative diagnosis. This research was taken investigate utility artificial intelligence (AI) accurate detection from chest images. aim this paper propose robust technique automatic pneumonia digital images applying pre-trained deep-learning algorithms while maximizing accuracy. A public database created by authors combining databases also collecting recently published articles. contains mixture 423 COVID-19, 1485 viral pneumonia, 1579 normal Transfer learning used help image augmentation train validate deep Convolutional Neural Networks (CNNs). networks were trained classify two different schemes: i) pneumonia; ii) normal, without augmentation. classification accuracy, precision, sensitivity, specificity both schemes 99.7%, 99.7% 99.55% 97.9%, 97.95%, 98.8%, respectively.

Language: Английский

---

Recent Advances in Electrochemical Biosensors: Applications, Challenges, and Future Scope

Biosensors, Journal Year: 2021, Volume and Issue: 11(9), P. 336 - 336

Published: Sept. 14, 2021

The electrochemical biosensors are a class of which convert biological information such as analyte concentration that is recognition element (biochemical receptor) into current or voltage. Electrochemical depict propitious diagnostic technology can detect biomarkers in body fluids sweat, blood, feces, urine. Combinations suitable immobilization techniques with effective transducers give rise to an efficient biosensor. They have been employed the food industry, medical sciences, defense, studying plant biology, etc. While sensing complex structures and entities, large data obtained, it becomes difficult manually interpret all data. Machine learning helps interpreting In case biosensors, presence impurity affects performance sensor machine removing signals obtained from contaminants obtain high sensitivity. this review, we discuss different types along their applications benefits learning. This followed by discussion on challenges, missing gaps knowledge, solutions field biosensors. review aims serve valuable resource for scientists engineers entering interdisciplinary Furthermore, provides insight type applications, importance (ML) biosensing, challenges future outlook.

Language: Английский

---

Efficient Pneumonia Detection in Chest Xray Images Using Deep Transfer Learning

Diagnostics, Journal Year: 2020, Volume and Issue: 10(6), P. 417 - 417

Published: June 19, 2020

Pneumonia causes the death of around 700,000 children every year and affects 7% global population. Chest X-rays are primarily used for diagnosis this disease. However, even a trained radiologist, it is challenging task to examine chest X-rays. There need improve accuracy. In work, an efficient model detection pneumonia on digital X-ray images proposed, which could aid radiologists in their decision making process. A novel approach based weighted classifier introduced, combines predictions from state-of-the-art deep learning models such as ResNet18, Xception, InceptionV3, DenseNet121, MobileNetV3 optimal way. This supervised network predicts result quality dataset used. Transfer fine-tune obtain higher training validation Partial data augmentation techniques employed increase balanced The proposed able outperform all individual models. Finally, evaluated, not only terms test accuracy, but also AUC score. final achieve accuracy 98.43% score 99.76 unseen Guangzhou Women Children's Medical Center dataset. Hence, can be quick

Language: Английский

---

Making use of nanoenergy from human – Nanogenerator and self-powered sensor enabled sustainable wireless IoT sensory systems

Nano Today, Journal Year: 2020, Volume and Issue: 36, P. 101016 - 101016

Published: Nov. 7, 2020

Language: Английский

---

Deep-Pneumonia Framework Using Deep Learning Models Based on Chest X-Ray Images

Diagnostics, Journal Year: 2020, Volume and Issue: 10(9), P. 649 - 649

Published: Aug. 28, 2020

Pneumonia is a contagious disease that causes ulcers of the lungs, and one main reasons for death among children elderly in world. Several deep learning models detecting pneumonia from chest X-ray images have been proposed. One extreme challenges has to find an appropriate efficient model meets all performance metrics. Proposing powerful classifying purpose this work. In paper, four different are developed by changing used method; two pre-trained models, ResNet152V2 MobileNetV2, Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM). The proposed implemented evaluated using Python compared with recent similar research. results demonstrate our framework improves accuracy, precision, F1-score, recall, Area Under Curve (AUC) 99.22%, 99.43%, 99.44%, 99.77%, respectively. As clearly illustrated results, outperforms other recently works. Moreover, models-MobileNetV2, CNN, LSTM-CNN-achieved more than 91% AUC, exceed introduced literature.

Language: Английский

---

Classification of COVID-19 from Chest X-ray images using Deep Convolutional Neural Network

Published: Dec. 11, 2020

The COVID-19 pandemic continues to have a devastating effect on the health and well-being of global population. A vital step in combat towards is successful screening contaminated patients, with one key approaches being radiological imaging using chest radiography. This study aimed automatically detect pneumonia patients digital x-ray images while maximizing accuracy detection deep convolutional neural networks (DCNN). dataset consists 864 COVID-19, 1345 viral 1341 normal xray images. In this study, DCNN based model Inception V3 transfer learning been proposed for coronavirus infected X-ray radiographs gives classification more than 98% (training 97% validation 93%). results demonstrate that proved be effective, showed robust performance easily deployable approach detection.

Language: Английский

---

Deep Learning in the Detection and Diagnosis of COVID-19 Using Radiology Modalities: A Systematic Review

Journal of Healthcare Engineering, Journal Year: 2021, Volume and Issue: 2021, P. 1 - 10

Published: March 15, 2021

Introduction. The early detection and diagnosis of COVID-19 the accurate separation non-COVID-19 cases at lowest cost in stages disease are among main challenges current pandemic. Concerning novelty disease, diagnostic methods based on radiological images suffer from shortcomings despite their many applications centers. Accordingly, medical computer researchers tend to use machine-learning models analyze radiology images. Material Methods. present systematic review was conducted by searching three databases PubMed, Scopus, Web Science November 1, 2019, July 20, 2020, a search strategy. A total 168 articles were extracted and, applying inclusion exclusion criteria, 37 selected as research population. Result. This study provides an overview state all for through modalities processing deep learning. According findings, learning-based have extraordinary capacity offer efficient system COVID-19, which would lead significant increase sensitivity specificity values. Conclusion. application learning field radiologic image reduces false-positive negative errors this offers unique opportunity provide fast, cheap, safe services patients.

Language: Английский

---

Soft Electronics for Health Monitoring Assisted by Machine Learning

Nano-Micro Letters, Journal Year: 2023, Volume and Issue: 15(1)

Published: March 15, 2023

Abstract Due to the development of novel materials, past two decades have witnessed rapid advances soft electronics. The electronics huge potential in physical sign monitoring and health care. One important advantages is forming good interface with skin, which can increase user scale improve signal quality. Therefore, it easy build specific dataset, performance machine learning algorithm. At same time, assistance algorithm, become more intelligent realize real-time analysis diagnosis. machining algorithms complement each other very well. It indubitable that will bring us a healthier world near future. this review, we give careful introduction about new material, physiological detected by devices, devices assisted Some materials be discussed such as two-dimensional carbon nanotube, nanowire, nanomesh, hydrogel. Then, sensors according types (pulse, respiration, human motion, intraocular pressure, phonation, etc.). After that, various reviewed, including some classical powerful neural network algorithms. Especially, device introduced carefully. Finally, outlook, challenge, conclusion system powered algorithm discussed.

Language: Английский

---

Transfer Learning with Deep Convolutional Neural Network (CNN) for Pneumonia Detection Using Chest X-ray

Applied Sciences, Journal Year: 2020, Volume and Issue: 10(9), P. 3233 - 3233

Published: May 6, 2020

Pneumonia is a life-threatening disease, which occurs in the lungs caused by either bacterial or viral infection. It can be life-endangering if not acted upon right time and thus an early diagnosis of pneumonia vital. The aim this paper to automatically detect using digital x-ray images. provides detailed report on advances made making accurate detection then presents methodology adopted authors. Four different pre-trained deep Convolutional Neural Network (CNN)- AlexNet, ResNet18, DenseNet201, SqueezeNet were used for transfer learning. 5247 Bacterial, normal chest x-rays images underwent preprocessing techniques modified trained learning based classification task. In work, authors have reported three schemes classifications: vs pneumonia, normal, pneumonia. accuracy images, 98%, 95%, 93.3% respectively. This highest any scheme than accuracies literature. Therefore, proposed study useful faster-diagnosing radiologist help fast airport screening patients.

Language: Английский

---

Classification of COVID-19 from Chest X-ray images using Deep Convolutional Neural Networks

medRxiv (Cold Spring Harbor Laboratory), Journal Year: 2020, Volume and Issue: unknown

Published: May 6, 2020

Abstract The COVID-19 pandemic continues to have a devastating effect on the health and well-being of global population. A vital step in combat towards is successful screening contaminated patients, with one key approaches being radiological imaging using chest radiography. This study aimed automatically detect COVID‐ 19 pneumonia patients digital x‐ ray images while maximizing accuracy detection deep convolutional neural networks (DCNN). dataset consists 864 19, 1345 viral 1341 normal images. In this study, DCNN based model Inception V3 transfer learning been proposed for coronavirus infected X-ray radiographs gives classification more than 98% (training 97% validation 93%). results demonstrate that proved be effective, showed robust performance easily deployable approach detection.

Language: Английский

---