Automatic Detection of Temporomandibular Joint Effusion with Deep Learning Algorithm DOI Creative Commons
Yeon‐Hee Lee,

Seonggwang Jeon,

Jong-Hyun Won

et al.

Research Square (Research Square), Journal Year: 2023, Volume and Issue: unknown

Published: Sept. 6, 2023

Abstract This study investigated the usefulness of deep learning-based automatic detection temporomandibular joint (TMJ) effusion using magnetic resonance imaging (MRI) in patients with disorder (TMD) and whether diagnostic accuracy model improved when patients’ clinical information was provided addition to MRI images. The sagittal MR images 2,948 TMJs were collected from 1,017 women 457 men (mean age 37.19 ± 18.64 years). TMJ performances three convolutional neural networks (scratch, fine-tuning, freeze schemes) compared those human experts based on areas under curve (AUCs) diagnosis accuracies. fine-tuning proton density (PD) showed acceptable prediction performance (AUC = 0.7895), from-scratch (0.6193) (0.6149) models lower (p < 0.05). had excellent specificity (87.25% vs. 58.17%). However, superior sensitivity (80.00% 57.43%) (all p 0.001). In Grad-CAM visualizations, scheme focused more than other structures TMJ, sparsity higher that (82.40% 49.83%, visualizations agreed learned through important features area, particularly around articular disc. Two PD T2-weighted did not improve alone Diverse AUCs observed across each group divided according (0.7083–0.8375) sex (male:0.7576, female:0.7083). ensemble all data used (74.21% 67.71%, A network (DNN) developed process multimodal data, including patient data. Analysis four groups DNN 41–60 best 0.8258). There no significant difference between > optimal for judging may be prevent true negative cases aid performance. Assistive automated methods have potential increase clinicians’ accuracy.

Language: Английский

Applications of Artificial Intelligence in Acute Thoracic Imaging DOI Creative Commons
Hayley Briody, Kate Hanneman, Michael N. Patlas

et al.

Canadian Association of Radiologists Journal, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 19, 2025

The applications of artificial intelligence (AI) in radiology are rapidly advancing with AI algorithms being used a wide range disease pathologies and clinical settings. Acute thoracic including rib fractures, pneumothoraces, acute PE associated significant morbidity mortality their identification is crucial for prompt treatment. models which increase diagnostic accuracy, improve radiologist efficiency reduce time to diagnosis abnormalities the thorax have potential significantly patient outcomes. purpose this review summarize current imaging, highlighting strengths, limitations, future research opportunities.

Language: Английский

Citations

1
Efficacy of a deep learning-based software for chest X-ray analysis in an emergency department DOI

Sathiyamurthy Selvam,

Olivier Peyrony,

Arben Elezi

et al.

Diagnostic and Interventional Imaging, Journal Year: 2025, Volume and Issue: unknown

Published: April 1, 2025

Language: Английский

Citations

1
Artificial intelligence in radiology DOI Creative Commons
Guillermo Elizondo‐Riojas, Adrián A. Negreros-Osuna,

J. Mario Bernal-Ramirez

et al.

Journal of the Mexican Federation of Radiology and Imaging, Journal Year: 2024, Volume and Issue: 3(2)

Published: July 9, 2024

Artificial intelligence (AI) is revolutionizing clinical medicine, particularly radiology, by enhancing diagnostic accuracy and streamlining operational efficiency.Radiology benefits from AI's prowess in image pattern recognition, which not only augments radiologists' capabilities but also optimizes tasks such as scheduling radiation monitoring.AI's applications span interventional enabling the interpretation of complex imaging data through advanced technologies convolutional neural networks radiomics.These tools help detect subtle disease indicators often missed human eye.AI improves radiology department management automating routine prioritizing urgent cases to ensure timely medical interventions.Educational programs must evolve prepare next generation radiologists for a future where AI ubiquitous their professional landscape.However, integrating into brings challenges, including ethical legal concerns about patient privacy, security, potential bias algorithms.Ethical be addressed developing robust guidelines that keep pace with technological advancements.Addressing these issues requires rigorous validation across various settings demographics.Undoubtedly, will empower radiologists, enhance accuracy, contribute precision personalized medicine.

Language: Английский

Citations

5
Deep Learning for Pneumothorax Detection on Chest Radiograph: A Diagnostic Test Accuracy Systematic Review and Meta Analysis DOI Creative Commons
Benjamin D. Katzman, Mostafa Alabousi,

Nabil Islam

et al.

Canadian Association of Radiologists Journal, Journal Year: 2024, Volume and Issue: 75(3), P. 525 - 533

Published: Jan. 8, 2024

Pneumothorax is a common acute presentation in healthcare settings. A chest radiograph (CXR) often necessary to make the diagnosis, and minimizing time between diagnosis critical deliver optimal treatment. Deep learning (DL) algorithms have been developed rapidly identify pathologic findings on various imaging modalities.

Language: Английский

Citations

4
AI Workflow Integration and Pneumothorax Detection—A Blueprint for Sustainable Implementation? DOI
Lukáš Müller, Daniel Santos

Academic Radiology, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 1, 2025

Language: Английский

Citations

0
Applications of artificial intelligence in thoracic imaging: a review DOI Creative Commons
Arjun Kalyanpur, Neetika Mathur

Academia Medicine, Journal Year: 2025, Volume and Issue: 2(1)

Published: Feb. 21, 2025

Artificial intelligence (AI) is transforming the field of radiology. Among various radiologic subspecialties, thoracic imaging has seen a significant rise in demand due to global increase heart, vascular, lung, and diseases such as lung cancer, pneumonia, pulmonary embolism, cardiovascular diseases. AI promises revolutionize diagnostics by enhancing detection, improving accuracy, reducing time required interpret images. It leverages deep learning algorithms, particularly convolutional neural networks, which are increasingly integrated into workflows assist radiologists diagnosing evaluating systems can help identify subtle findings that might otherwise be overlooked, thereby increasing efficiency diagnostic errors. Studies have shown several algorithms been trained detect acute chest conditions aortic dissection, rib fractures, nodules with high sensitivity specificity, offering substantial benefits emergency high-workload environments. This review article focuses on presenting syndrome or trauma settings. provides an overview applications imaging, focusing advancements screening, early disease triage prioritization, automated image analysis, workflow optimization. These points supported articles published subject, including our own publications. We further explore challenges regulatory barriers, interpretability, need for large, diverse datasets. Finally, we discuss future directions highlighting its potential enhance patient outcomes healthcare system efficiencies.

Language: Английский

Citations

0
Prediction and Detection of Pneumothorax on X-Rays Using Neural Network DOI

Joshitha Udhaykumar,

Pradeep Kumar Yadalam, Swarnambiga Ayyachamy

et al.

IGI Global eBooks, Journal Year: 2025, Volume and Issue: unknown, P. 173 - 194

Published: March 28, 2025

This study aims to use neural networks predict pneumothorax on X-rays, highlighting the transformative impact of these systems diagnosing and managing condition, thereby improving clinical decision-making. The uses a Kaggle dataset chest X-rays with annotated labels detect pneumothorax. Convolutional are used for image classification tasks, including detection. accuracy logistics regression was 59% 61%, respectively. A deep learning model detection has shown high sensitivity, specificity, accuracy, interpretability. Neural demonstrate potential in radiographs. Deep models surpass traditional logistic this task. Further research is required optimize performance evaluate utility.

Language: Английский

Citations

0
Developing an explainable diagnosis system utilizing deep learning model: a case study of spontaneous pneumothorax DOI Creative Commons

Frank Cheau-Feng Lin,

Chia-Jung Wei,

Zhe-Rui Bai

et al.

Physics in Medicine and Biology, Journal Year: 2024, Volume and Issue: 69(14), P. 145017 - 145017

Published: July 2, 2024

Abstract Objective. The trend in the medical field is towards intelligent detection-based diagnostic systems. However, these methods are often seen as ‘black boxes’ due to their lack of interpretability. This situation presents challenges identifying reasons for misdiagnoses and improving accuracy, which leads potential risks misdiagnosis delayed treatment. Therefore, how enhance interpretability models crucial patient outcomes reducing treatment delays. So far, only limited researches exist on deep learning-based prediction spontaneous pneumothorax, a pulmonary disease that affects lung ventilation venous return. Approach. study develops an integrated image analysis system using explainable learning model recognition visualization achieve interpretable automatic diagnosis process. Main results. achieves impressive 95.56% accuracy pneumothorax classification, emphasizes significance blood vessel penetration defect clinical judgment. Significance. would lead improve trustworthiness, reduce uncertainty, accurate various diseases, results better patients utilization resources. Future research can focus implementing new detect diagnose other diseases generalizability this system.

Language: Английский

Citations

3
Automatic detection and visualization of temporomandibular joint effusion with deep neural network DOI Creative Commons
Yeon‐Hee Lee,

Seonggwang Jeon,

Jong-Hyun Won

et al.

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Aug. 14, 2024

This study investigated the usefulness of deep learning-based automatic detection temporomandibular joint (TMJ) effusion using magnetic resonance imaging (MRI) in patients with disorder and whether diagnostic accuracy model improved when patients' clinical information was provided addition to MRI images. The sagittal MR images 2948 TMJs were collected from 1017 women 457 men (mean age 37.19 ± 18.64 years). TMJ performances three convolutional neural networks (scratch, fine-tuning, freeze schemes) compared those human experts based on areas under curve (AUCs) diagnosis accuracies. fine-tuning proton density (PD) showed acceptable prediction performance (AUC = 0.7895), from-scratch (0.6193) (0.6149) models lower (p < 0.05). had excellent specificity (87.25% vs. 58.17%). However, superior sensitivity (80.00% 57.43%) (all p 0.001). In gradient-weighted class activation mapping (Grad-CAM) visualizations, scheme focused more than other structures TMJ, sparsity higher that (82.40% 49.83%, Grad-CAM visualizations agreed learned through important features area, particularly around articular disc. Two PD T2-weighted did not improve alone Diverse AUCs observed across each group divided according (0.7083–0.8375) sex (male:0.7576, female:0.7083). ensemble all data used (74.21% 67.71%, A network (DNN) developed process multimodal data, including patient data. Analysis four groups DNN 41–60 best 0.8258). optimal for judging may be prevent true negative cases aid performance. Assistive automated methods have potential increase clinicians' accuracy.

Language: Английский

Citations

3
Diagnostic evaluation of blunt chest trauma by imaging-based application of artificial intelligence: A review DOI
Tingting Zhao, Xianghong Meng, Zhi Wang

et al.

The American Journal of Emergency Medicine, Journal Year: 2024, Volume and Issue: 85, P. 35 - 43

Published: Aug. 15, 2024

Language: Английский

Citations

3