A hybrid explainable ensemble transformer encoder for pneumonia identification from chest X-ray images

Journal of Advanced Research, Journal Year: 2022, Volume and Issue: 48, P. 191 - 211

Published: Sept. 7, 2022

Pneumonia is a microorganism infection that causes chronic inflammation of the human lung cells. Chest X-ray imaging most well-known screening approach used for detecting pneumonia in early stages. While chest-Xray images are mostly blurry with low illumination, strong feature extraction required promising identification performance. A new hybrid explainable deep learning framework proposed accurate disease using chest images. The workflow developed by fusing capabilities both ensemble convolutional networks and Transformer Encoder mechanism. backbone to extract features from raw input two different scenarios: (i.e., DenseNet201, VGG16, GoogleNet) B InceptionResNetV2, Xception). Whereas, built based on self-attention mechanism multilayer perceptron (MLP) identification. visual saliency maps derived emphasize crucial predicted regions end-to-end training process models over all scenarios performed binary multi-class classification scenarios. model recorded 99.21% performance terms overall accuracy F1-score task, while it achieved 98.19% 97.29% multi-classification task. For scenario, 97.22% 97.14% F1-score, 96.44% F1-score. multiclass 97.2% 95.8% 96.4% 94.9% could provide encouraging comparing individual, models, or even latest AI literature. code available here: https://github.com/chiagoziemchima/Pneumonia_Identificaton.

Language: Английский

---

A systematic review of hyperparameter optimization techniques in Convolutional Neural Networks

Decision Analytics Journal, Journal Year: 2024, Volume and Issue: 11, P. 100470 - 100470

Published: April 24, 2024

Convolutional Neural Network (CNN) is a prevalent topic in deep learning (DL) research for their architectural advantages. CNN relies heavily on hyperparameter configurations, and manually tuning these hyperparameters can be time-consuming researchers, therefore we need efficient optimization techniques. In this systematic review, explore range of well used algorithms, including metaheuristic, statistical, sequential, numerical approaches, to fine-tune hyperparameters. Our offers an exhaustive categorization (HPO) algorithms investigates the fundamental concepts CNN, explaining role variants. Furthermore, literature review HPO employing above mentioned undertaken. A comparative analysis conducted based strategies, error evaluation accuracy results across various datasets assess efficacy methods. addition addressing current challenges HPO, our illuminates unresolved issues field. By providing insightful evaluations merits demerits objective assist researchers determining suitable method particular problem dataset. highlighting future directions synthesizing diversified knowledge, survey contributes significantly ongoing development optimization.

Language: Английский

---

PneuNet: deep learning for COVID-19 pneumonia diagnosis on chest X-ray image analysis using Vision Transformer

Medical & Biological Engineering & Computing, Journal Year: 2023, Volume and Issue: 61(6), P. 1395 - 1408

Published: Jan. 31, 2023

Language: Английский

---

A Deep Learning Fusion Approach to Diagnosis the Polycystic Ovary Syndrome (PCOS)

Applied Computational Intelligence and Soft Computing, Journal Year: 2023, Volume and Issue: 2023, P. 1 - 15

Published: Feb. 14, 2023

One of the leading causes female infertility is PCOS, which a hormonal disorder affecting women childbearing age. The common symptoms PCOS include increased acne, irregular period, increase in body hair, and overweight. Early diagnosis essential to manage reduce associated health risks. Nonetheless, based on Rotterdam criteria, including high level androgen hormones, ovulation failure, polycystic ovaries ultrasound image (PCOM). At present, doctors radiologists manually perform PCOM detection using ovary by counting number follicles determining their volume ovaries, one challenging diagnostic criteria. Moreover, such physicians require more tests checks for biochemical/clinical signs addition patient’s order decide diagnosis. Furthermore, clinicians do not utilize single test or specific method examine patients. This paper introduces data set that includes with clinical related patient has been classified as non-PCOS. Next, we proposed deep learning model can diagnose image, achieved 84.81% accuracy Inception model. Then, fusion if they have not. best developed 82.46% extracting features MobileNet architecture combine features.

Language: Английский

---

A Computer-Aided Diagnostic System to Identify Diabetic Retinopathy, Utilizing a Modified Compact Convolutional Transformer and Low-Resolution Images to Reduce Computation Time

Biomedicines, Journal Year: 2023, Volume and Issue: 11(6), P. 1566 - 1566

Published: May 28, 2023

Diabetic retinopathy (DR) is the foremost cause of blindness in people with diabetes worldwide, and early diagnosis essential for effective treatment. Unfortunately, present DR screening method requires skill ophthalmologists time-consuming. In this study, we an automated system severity classification employing fine-tuned Compact Convolutional Transformer (CCT) model to overcome these issues. We assembled five datasets generate a more extensive dataset containing 53,185 raw images. Various image pre-processing techniques 12 types augmentation procedures were applied improve quality create massive dataset. A new DR-CCTNet proposed. It modification original CCT address training time concerns work large amount data. Our proposed delivers excellent accuracy even low-pixel images still has strong performance fewer images, indicating that robust. compare our model's transfer learning models such as VGG19, VGG16, MobileNetV2, ResNet50. The test ResNet50, MobileNetV2 were, respectively, 72.88%, 76.67%, 73.22%, 71.98%. classify outperformed all 90.17% accuracy. This approach provides novel efficient detection DR, which may lower burden on expedite treatment patients.

Language: Английский

---

A comparative study of Detecting Covid 19 by Using Chest X-ray Images– A Deep Learning Approach

2022 IEEE World AI IoT Congress (AIIoT), Journal Year: 2023, Volume and Issue: unknown

Published: June 7, 2023

SARS-CoV-2's COVID-19 pandemic has quickly spread over the world, inflicting a sizable number of illnesses and fatalities. Stopping virus's depends on correctly rapidly identifying infected people. Although RT-PCR assays, for example, are thought to be most accurate way identify COVID-19, their cost availability may restricted in places with limited resources. In this study, we propose some deep-learning methods predicting detection using chest X-ray images. Chest imaging become an essential diagnostic tool management as it is non-invasive, widely available, cost-effective. However, interpretation X-rays can challenging, radiographic features pneumonia subtle overlap other respiratory diseases. performance different deep learning models, notably VGG16, VGG19, DenseNet121, Resnet50, was examined ability distinguish between coronavirus cases pneumonia. 4649 images patients (3526) (1123) were employed measures used assess each model. Confusion metrics also evaluate model's performance. The study's findings demonstrated that DenseNet121 performed better than competing accuracy rate 99.78%.

Language: Английский

---

Deep Learning for Pneumonia Detection in Chest X-ray Images: A Comprehensive Survey

Journal of Imaging, Journal Year: 2024, Volume and Issue: 10(8), P. 176 - 176

Published: July 23, 2024

This paper addresses the significant problem of identifying relevant background and contextual literature related to deep learning (DL) as an evolving technology in order provide a comprehensive analysis application DL specific pneumonia detection via chest X-ray (CXR) imaging, which is most common cost-effective imaging technique available worldwide for diagnosis. particular key period associated with COVID-19, 2020–2023, explain, analyze, systematically evaluate limitations approaches determine their relative levels effectiveness. The context applied both aid automated substitute existing expert radiography professionals, who often have limited availability, elaborated detail. rationale undertaken research provided, along justification resources adopted relevance. explanatory text subsequent analyses are intended sufficient detail being addressed, solutions, these, ranging from more general. Indeed, our evaluation agree generally held view that use transformers, specifically, vision transformers (ViTs), promising obtaining further effective results area using CXR images. However, ViTs require extensive address several limitations, specifically following: biased datasets, data code ease model can be explained, systematic methods accurate comparison, notion class imbalance possibility adversarial attacks, latter remains fundamental research.

Language: Английский

---

Medical image processing and COVID-19: A literature review and bibliometric analysis

Journal of Infection and Public Health, Journal Year: 2021, Volume and Issue: 15(1), P. 75 - 93

Published: Nov. 17, 2021

COVID-19 crisis has placed medical systems over the world under unprecedented and growing pressure. Medical imaging processing can help in diagnosis, treatment, early detection of diseases. It been considered as one modern technologies applied to fight against crisis. Although several artificial intelligence, machine learning, deep learning techniques have deployed image context disease, there is a lack research considering systematic literature review categorization published studies this field. A locates, assesses, interprets outcomes address predetermined goal present evidence-based practical theoretical insights. The main study methods With mind, available reliable databases were retrieved, studied, evaluated, synthesized. Based on in-depth literature, structured conceptual map that outlined three multi-layered folds: data gathering description, steps processing, evaluation metrics. themes elaborated each fold, allowing authors recommend upcoming paths for scholars. highlighted adopted classify images related diagnosis COVID-19. presented promising terms accuracy, cost, speed.

Language: Английский

---

COVID-19 classification using chest X-ray images: A framework of CNN-LSTM and improved max value moth flame optimization

Frontiers in Public Health, Journal Year: 2022, Volume and Issue: 10

Published: Aug. 30, 2022

Coronavirus disease 2019 (COVID-19) is a highly contagious that has claimed the lives of millions people worldwide in last 2 years. Because disease's rapid spread, it critical to diagnose at an early stage order reduce rate spread. The images lungs are used this infection. In years, many studies have been introduced help with diagnosis COVID-19 from chest X-Ray images. all researchers looking for quick method virus, deep learning-based computer controlled techniques more suitable as second opinion radiologists. article, we look issue multisource fusion and redundant features. We proposed CNN-LSTM improved max value features optimization framework classification address these issues. original acquired contrast increased using combination filtering algorithms architecture. dataset then augmented increase its size, which train two learning networks called Modified EfficientNet B0 CNN-LSTM. Both built scratch extract information layers. Following extraction features, serial based maximum technique combine best both models. However, few also noted; therefore, moth flame algorithm proposed. Through algorithm, selected finally classified through machine classifiers. experimental process was conducted on three publically available datasets achieved accuracy than existing techniques. Moreover, classifiers comparison cubic support vector gives better accuracy.

Language: Английский

---

Driver drowsiness detection and smart alerting using deep learning and IoT

Internet of Things, Journal Year: 2023, Volume and Issue: 22, P. 100705 - 100705

Published: Feb. 14, 2023

Language: Английский

---