Medical image super-resolution for smart healthcare applications: A comprehensive survey

Information Fusion, Journal Year: 2023, Volume and Issue: 103, P. 102075 - 102075

Published: Oct. 18, 2023

The digital transformation in healthcare, propelled by the integration of deep learning models and Internet Things (IoT), is creating unprecedented opportunities for improving patient care. However, utilization low-resolution images, often generated IoT devices, introduces biases models, thereby affecting overall clinical decision-making process. While super-resolution techniques have been extensively employed to transform images into high-resolution counterparts, challenge achieving highly accurate image restoration remains unresolved. This especially critical medical imaging domain, where even minor inaccuracies can lead significant model training and, consequently, impact outcomes. Although existing surveys explored various methods their applications across different fields, a comprehensive review emphasizing accuracy its subsequent influence on notably lacking. survey seeks bridge this gap offering systematic current state-of-the-art highlighting limitations surveys, underscoring open questions that merit further research. Specifically, we delve intricacies restoration, identify research gaps unmet challenges optimal emphasize crucial role developing more precise resilient enhance quality performance healthcare applications. Ultimately, fosters deeper comprehension prevailing unresolved field, thus setting stage future efforts focused refining subsequently, boosting efficacy healthcare.

Language: Английский

---

A Deep Learning Framework for the Prediction and Diagnosis of Ovarian Cancer in Pre- and Post-Menopausal Women

Diagnostics, Journal Year: 2023, Volume and Issue: 13(10), P. 1703 - 1703

Published: May 11, 2023

Ovarian cancer ranks as the fifth leading cause of cancer-related mortality in women. Late-stage diagnosis (stages III and IV) is a major challenge due to often vague inconsistent initial symptoms. Current diagnostic methods, such biomarkers, biopsy, imaging tests, face limitations, including subjectivity, inter-observer variability, extended testing times. This study proposes novel convolutional neural network (CNN) algorithm for predicting diagnosing ovarian cancer, addressing these limitations. In this paper, CNN was trained on histopathological image dataset, divided into training validation subsets augmented before training. The model achieved remarkable accuracy 94%, with 95.12% cancerous cases correctly identified 93.02% healthy cells accurately classified. significance lies overcoming challenges associated human expert examination, higher misclassification rates, analysis presents more accurate, efficient, reliable approach cancer. Future research should explore recent advances field enhance effectiveness proposed method further.

Language: Английский

---

Role of Artificial Intelligence in Medical Image Analysis: A Review of Current Trends and Future Directions

Journal of Medical and Biological Engineering, Journal Year: 2024, Volume and Issue: 44(2), P. 231 - 243

Published: April 1, 2024

Language: Английский

---

Physics-Informed Computer Vision: A Review and Perspectives

ACM Computing Surveys, Journal Year: 2024, Volume and Issue: 57(1), P. 1 - 38

Published: Aug. 20, 2024

The incorporation of physical information in machine learning frameworks is opening and transforming many application domains. Here the process augmented through induction fundamental knowledge governing laws. In this work, we explore their utility for computer vision tasks interpreting understanding visual data. We present a systematic literature review more than 250 papers on formulation approaches to guided by begin decomposing popular pipeline into taxonomy stages investigate incorporate equations each stage. Existing are analyzed terms modeling processes, including modifying input data (observation bias), network architectures (inductive training losses (learning bias). offers unified view physics-informed capability, highlighting where has been conducted gaps opportunities are. Finally, highlight open problems challenges inform future research. While still its early days, study promise develop better models that can improve plausibility, accuracy, efficiency, generalization increasingly realistic applications.

Language: Английский

---

Advanced Medical Image Segmentation Enhancement: A Particle-Swarm-Optimization-Based Histogram Equalization Approach

Applied Sciences, Journal Year: 2024, Volume and Issue: 14(2), P. 923 - 923

Published: Jan. 22, 2024

Accurate medical image segmentation is paramount for precise diagnosis and treatment in modern healthcare. This research presents a comprehensive study of the efficacy particle swarm optimization (PSO) combined with histogram equalization (HE) preprocessing segmentation, focusing on lung CT scan chest X-ray datasets. Best-cost values reveal PSO algorithm’s performance, HE demonstrating significant stabilization enhanced convergence, particularly complex images. Evaluation metrics, including accuracy, precision, recall, F1-score/Dice, specificity, Jaccard, show substantial improvements preprocessing, emphasizing its impact accuracy. Comparative analyses against alternative methods, such as Otsu, Watershed, K-means, confirm competitiveness PSO-HE approach, especially The also underscores positive influence clarity precision. These findings highlight promise approach advancing accuracy reliability pave way further method integration to enhance this critical healthcare application.

Language: Английский

---

The Advances in Computer Vision That Are Enabling More Autonomous Actions in Surgery: A Systematic Review of the Literature

Sensors, Journal Year: 2022, Volume and Issue: 22(13), P. 4918 - 4918

Published: June 29, 2022

This is a review focused on advances and current limitations of computer vision (CV) how CV can help us obtain to more autonomous actions in surgery. It follow-up article one that we previously published

Language: Английский

---

Integrated Generative Adversarial Networks and Deep Convolutional Neural Networks for Image Data Classification: A Case Study for COVID-19

Information, Journal Year: 2024, Volume and Issue: 15(1), P. 58 - 58

Published: Jan. 18, 2024

Convolutional Neural Networks (CNNs) have garnered significant utilisation within automated image classification systems. CNNs possess the ability to leverage spatial and temporal correlations inherent in a dataset. This study delves into use of cutting-edge deep learning for precise data classification, focusing on overcoming difficulties brought by COVID-19 pandemic. In order improve accuracy robustness introduces novel methodology that combines strength Deep (DCNNs) Generative Adversarial (GANs). proposed helps mitigate lack labelled coronavirus (COVID-19) images, which has been standard limitation related research, improves model’s distinguish between COVID-19-related patterns healthy lung images. The uses thorough case sizable dataset chest X-ray images covering cases, other respiratory conditions, conditions. integrated model outperforms conventional DCNN-based techniques terms after being trained this To address issues an unbalanced dataset, GAN will produce synthetic pictures extract features from every image. A understanding performance real-world scenarios is also provided study’s meticulous evaluation using variety metrics, including accuracy, precision, recall, F1-score.

Language: Английский

---

A hybrid framework for glaucoma detection through federated machine learning and deep learning models

BMC Medical Informatics and Decision Making, Journal Year: 2024, Volume and Issue: 24(1)

Published: May 2, 2024

Language: Английский

---

Optimized deep learning model for comprehensive medical image analysis across multiple modalities

Neurocomputing, Journal Year: 2024, Volume and Issue: 619, P. 129182 - 129182

Published: Dec. 12, 2024

Language: Английский

---

Deep learning-based image analysis in muscle histopathology using photo-realistic synthetic data

Communications Medicine, Journal Year: 2025, Volume and Issue: 5(1)

Published: March 6, 2025

Artificial intelligence (AI), specifically Deep learning (DL), has revolutionized biomedical image analysis, but its efficacy is limited by the need for representative, high-quality large datasets with manual annotations. While latest research on synthetic data using AI-based generative models shown promising results to tackle this problem, several challenges such as lack of interpretability and vast amounts real remain. This study aims introduce a new approach-SYNTA-for generation photo-realistic address associated state-of-the art DL-based analysis. The SYNTA method employs fully parametric approach create training tailored specific tasks. Its applicability tested in context muscle histopathology skeletal evaluated two real-world validate solve complex analysis tasks data. Here we show that enables expert-level segmentation unseen only By addressing representative data, achieves robust performance offering scalable, controllable interpretable alternative Generative Adversarial Networks (GANs) or Diffusion Models. demonstrates great potential accelerate improve ability generate reduces reliance extensive collection annotations, paving way advancements medical research.

Language: Английский

---