Learning In Medical Imaging Research Articles

AI-Assisted Detection and Localization of Spinal Metastatic Lesions.

The integration of machine learning and radiomics in medical imaging has significantly advanced diagnostic and prognostic capabilities in healthcare. This study focuses on developing and validating an artificial intelligence (AI) model using U-Net architectures for the accurate detection and segmentation of spinal metastases from computed tomography (CT) images, addressing both osteolytic and osteoblastic lesions. Our methodology employs multiple variations of the U-Net architecture and utilizes two distinct datasets: one consisting of 115 polytrauma patients for vertebra segmentation and another comprising 38 patients with documented spinal metastases for lesion detection. The model demonstrated strong performance in vertebra segmentation, achieving Dice Similarity Coefficient (DSC) values between 0.87 and 0.96. For metastasis segmentation, the model achieved a DSC of 0.71 and an F-beta score of 0.68 for lytic lesions but struggled with sclerotic lesions, obtaining a DSC of 0.61 and an F-beta score of 0.57, reflecting challenges in detecting dense, subtle bone alterations. Despite these limitations, the model successfully identified isolated metastatic lesions beyond the spine, such as in the sternum, indicating potential for broader skeletal metastasis detection. The study concludes that AI-based models can augment radiologists' capabilities by providing reliable second-opinion tools, though further refinements and diverse training data are needed for optimal performance, particularly for sclerotic lesion segmentation. The annotated CT dataset produced and shared in this research serves as a valuable resource for future advancements.

Diagnostic accuracy of deep learning-based algorithms in laryngoscopy: a systematic review and meta-analysis.

Laryngoscopy is routinely used for suspicious vocal cord lesions with limited performance. Accumulated studies have demonstrated the bright prospect of deep learning in processing medical imaging. In this study, we perform a systematic review and meta-analysis to investigate diagnostic utility of deep learning in laryngoscopy. The study was performed according to the Primary Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines. We comprehensively retrieved articles from the PubMed, Scopus, Embase, and Web of Science up to July 14, 2024. Eligible studies with application of deep learning algorithm in laryngoscopy were assessed and enrolled by two independent investigators. The pooled sensitivity, specificity, positive likelihood ratio, negative likelihood ratio and diagnostic odds ratio with 95% confidence intervals (CIs) were calculated using a random effects model. We retained 9 eligible studies adding up to 106,175 endoscopic images for the meta-analysis. The pooled sensitivity and specificity to diagnose laryngeal cancer were 0.95(95% CI: 0.85-0.98) and 0.96 (95% CI: 0.91-0.98). The area under the curve of deep learning was 0.99 (95%CI: 0.97-0.99). Deep learning demonstrated excellent diagnostic efficacy in assessing laryngeal cancer with laryngoscope images in current studies, which manifests its potential of aiding endoscopist for laryngeal cancer diagnosis and clinical decision making.

An introduction to Self-Aware Deep Learning for medical imaging and diagnosis

An introduction to Self-Aware Deep Learning for medical imaging and diagnosis

Insights and Considerations in Development and Performance Evaluation of Generative Adversarial Networks (GANs): What Radiologists Need to Know.

The rapid development of deep learning in medical imaging has significantly enhanced the capabilities of artificial intelligence while simultaneously introducing challenges, including the need for vast amounts of training data and the labor-intensive tasks of labeling and segmentation. Generative adversarial networks (GANs) have emerged as a solution, offering synthetic image generation for data augmentation and streamlining medical image processing tasks through models such as cGAN, CycleGAN, and StyleGAN. These innovations not only improve the efficiency of image augmentation, reconstruction, and segmentation, but also pave the way for unsupervised anomaly detection, markedly reducing the reliance on labeled datasets. Our investigation into GANs in medical imaging addresses their varied architectures, the considerations for selecting appropriate GAN models, and the nuances of model training and performance evaluation. This paper aims to provide radiologists who are new to GAN technology with a thorough understanding, guiding them through the practical application and evaluation of GANs in brain imaging with two illustrative examples using CycleGAN and pixel2style2pixel (pSp)-combined StyleGAN. It offers a comprehensive exploration of the transformative potential of GANs in medical imaging research. Ultimately, this paper strives to equip radiologists with the knowledge to effectively utilize GANs, encouraging further research and application within the field.

Bidirectional brain image translation using transfer learning from generic pre-trained models

Brain imaging plays a crucial role in the diagnosis and treatment of various neurological disorders, providing valuable insights into the structure and function of the brain. Techniques such as magnetic resonance imaging (MRI) and computed tomography (CT) enable non-invasive visualization of the brain, aiding in the understanding of brain anatomy, abnormalities, and functional connectivity. However, cost and radiation dose may limit the acquisition of specific image modalities, so medical image synthesis can be used to generate required medical images without actual addition. CycleGAN and other GANs are valuable tools for generating synthetic images across various fields. In the medical domain, where obtaining labeled medical images is labor-intensive and expensive, addressing data scarcity is a major challenge. Recent studies propose using transfer learning to overcome this issue. This involves adapting pre-trained CycleGAN models, initially trained on non-medical data, to generate realistic medical images. In this work, transfer learning was applied to the task of MR-CT image translation and vice versa using 18 pre-trained non-medical models, and the models were fine-tuned to have the best result. The models’ performance was evaluated using four widely used image quality metrics: Peak-signal-to-noise-ratio, Structural Similarity Index, Universal Quality Index, and Visual Information Fidelity. Quantitative evaluation and qualitative perceptual analysis by radiologists demonstrate the potential of transfer learning in medical imaging and the effectiveness of the generic pre-trained model. The results provide compelling evidence of the model’s exceptional performance, which can be attributed to the high quality and similarity of the training images to actual human brain images. These results underscore the significance of carefully selecting appropriate and representative training images to optimize performance in brain image analysis tasks.

Next-Gen Medical Imaging: U-Net Evolution and the Rise of Transformers.

The advancement of medical imaging has profoundly impacted our understanding of the human body and various diseases. It has led to the continuous refinement of related technologies over many years. Despite these advancements, several challenges persist in the development of medical imaging, including data shortages characterized by low contrast, high noise levels, and limited image resolution. The U-Net architecture has significantly evolved to address these challenges, becoming a staple in medical imaging due to its effective performance and numerous updated versions. However, the emergence of Transformer-based models marks a new era in deep learning for medical imaging. These models and their variants promise substantial progress, necessitating a comparative analysis to comprehend recent advancements. This review begins by exploring the fundamental U-Net architecture and its variants, then examines the limitations encountered during its evolution. It then introduces the Transformer-based self-attention mechanism and investigates how modern models incorporate positional information. The review emphasizes the revolutionary potential of Transformer-based techniques, discusses their limitations, and outlines potential avenues for future research.

A survey on deep learning in medical ultrasound imaging

Ultrasound imaging has a history of several decades. With its non-invasive, low-cost advantages, this technology has been widely used in medicine and there have been many significant breakthroughs in ultrasound imaging. Even so, there are still some drawbacks. Therefore, some novel image reconstruction and image analysis algorithms have been proposed to solve these problems. Although these new solutions have some effects, many of them introduce some other side effects, such as high computational complexity in beamforming. At the same time, the usage requirements of medical ultrasound equipment are relatively high, and it is not very user-friendly for inexperienced beginners. As artificial intelligence technology advances, some researchers have initiated efforts to deploy deep learning to address challenges in ultrasound imaging, such as reducing computational complexity in adaptive beamforming and aiding novices in image acquisition. In this survey, we are about to explore the application of deep learning in medical ultrasound imaging, spanning from image reconstruction to clinical diagnosis.

Artificial Intelligence in Medical Imaging: Applications of Deep Learning for Disease Detection and Diagnosis

The integration of artificial intelligence (AI) and deep learning techniques into medical imaging has revolutionized disease detection and diagnosis. This paper provides a comprehensive overview of the applications of deep learning in medical imaging and its impact on healthcare. The paper begins with an introduction to the fundamentals of deep learning, emphasizing convolutional neural networks (CNNs) and their relevance in analyzing medical images. It then explores various applications of deep learning in medical imaging, including automated disease detection and classification, image segmentation for precise anatomical localization, quantitative analysis for predictive modeling, personalized medicine, and workflow optimization. Case studies and examples from different medical specialties, such as oncology, cardiology, and neurology, are presented to illustrate the practical implementation and effectiveness of AI-driven approaches.

Evaluation of Neonatal Cerebral Circulation Under Hypoxic Ischemic Risk Factors Based on Quantitative Analysis of Cerebral Veins with Magnetic Resonance Susceptibility Weighted Imaging.

To observe the regulation of cerebral circulation in vivo based on image segmentation algorithms for deep learning in medical imaging to automatically detect and quantify the neonatal deep medullary veins (DMVs) on susceptibility weighted imaging (SWI) images. To evaluate early cerebral circulation self-rescue for neonates undergoing risk of cerebral hypoxia-ischaemia in vivo. SWI images and clinical data of 317 neonates with or without risk of cerebral hypoxia-ischaemia were analyzed. Quantitative parameters showing the number, width, and curvature of DMVs were obtained using an image segmentation algorithm. The number of DMVs was greater in males than in females (p < 0.01), and in term than in preterm infants (p = 0.001). The width of DMVs was greater in term than in preterm infants (p < 0.01), in low-risk than in high-risk group (p < 0.01), and in neonates without intracranial extracerebral haemorrhage (ICECH) than with ICECH (p < 0.05). The curvature of DMVs was greater in term than in preterm infants (P < 0.05). The width of both bilateral thalamic veins and anterior caudate nucleus veins were positively correlated with the number of DMVs; the width of bilateral thalamic veins was positively correlated with the width of DMVs. The DMVs quantification based on image segmentation algorithm may provide more detailed and stable quantitative information in neonate. SWI vein quantification may be an observable indicator for in vivo assessment of cerebral circulation self-regulation in neonatal hypoxic-ischemic brain injury.

Saliency-driven explainable deep learning in medical imaging: bridging visual explainability and statistical quantitative analysis

Deep learning shows great promise for medical image analysis but often lacks explainability, hindering its adoption in healthcare. Attribution techniques that explain model reasoning can potentially increase trust in deep learning among clinical stakeholders. In the literature, much of the research on attribution in medical imaging focuses on visual inspection rather than statistical quantitative analysis.In this paper, we proposed an image-based saliency framework to enhance the explainability of deep learning models in medical image analysis. We use adaptive path-based gradient integration, gradient-free techniques, and class activation mapping along with its derivatives to attribute predictions from brain tumor MRI and COVID-19 chest X-ray datasets made by recent deep convolutional neural network models.The proposed framework integrates qualitative and statistical quantitative assessments, employing Accuracy Information Curves (AICs) and Softmax Information Curves (SICs) to measure the effectiveness of saliency methods in retaining critical image information and their correlation with model predictions. Visual inspections indicate that methods such as ScoreCAM, XRAI, GradCAM, and GradCAM++ consistently produce focused and clinically interpretable attribution maps. These methods highlighted possible biomarkers, exposed model biases, and offered insights into the links between input features and predictions, demonstrating their ability to elucidate model reasoning on these datasets. Empirical evaluations reveal that ScoreCAM and XRAI are particularly effective in retaining relevant image regions, as reflected in their higher AUC values. However, SICs highlight variability, with instances of random saliency masks outperforming established methods, emphasizing the need for combining visual and empirical metrics for a comprehensive evaluation.The results underscore the importance of selecting appropriate saliency methods for specific medical imaging tasks and suggest that combining qualitative and quantitative approaches can enhance the transparency, trustworthiness, and clinical adoption of deep learning models in healthcare. This study advances model explainability to increase trust in deep learning among healthcare stakeholders by revealing the rationale behind predictions. Future research should refine empirical metrics for stability and reliability, include more diverse imaging modalities, and focus on improving model explainability to support clinical decision-making.

Auto-machine learning for opportunistic thyroid nodule detection in lung cancer screening chest CT.

e13639 Background: Automated Machine Learning (Auto-ML) in medical imaging is a process that allows non-experts to utilize machine learning techniques, opening the door for non-coder physician-driven exploitation of the technology. Auto-ML was applied for opportunistic detection of thyroid nodules in the context of low-dose lung cancer screening chest CT, facilitated by an innovative platform integration. By leveraging scans originally intended for lung cancer screening, suspicious appearing asymptomatic thyroid nodules can also be screened for where technically feasible. Methods: CT scans from the National Lung Screening Trial (NLST) dataset were utilized. This dataset contains low-dose chest CT examinations initially used for lung cancer screening. Annotation was carried out within a no-code web-based platform, Gesund.ai. A board-certified diagnostic radiologist annotated 100 CT examinations. A case was labeled "suspicious" if it presented a thyroid nodule of 1 cm or greater in diameter, while cases with no such nodules were deemed "not suspicious." The Auto-ML feature was then engaged within the software platform for parameter adjustments to refine the model without the need for manual coding. This phase was characterized by an iterative exploration of model performance, facilitated by the platform’s robust validation tools. Results: Upon training and validation, the Auto-ML model, cultivated from 100 annotated cases and assessed against a separate set of 130 cases, demonstrated a discernment accuracy of 0.51 in identifying "suspicious" thyroid nodules with a sensitivity and precision of 0.51 and 0.62 respectively. The AUC for this class stands at 0.69, indicating a moderate ability to distinguish between the presence and absence of thyroid nodules. This metric, while foundational, offers insight into the model's potential efficacy in real-world diagnostic scenarios, reflecting the initial capabilities of the platform's Auto-ML integration in enhancing diagnostic processes. Conclusions: The findings from this study highlight the potential of Auto-ML in revolutionizing opportunistic screening for thyroid nodules via lung cancer CT. The initial accuracy rate underscores the necessity for further refinement and validation of the model, but also demonstrates the promising potential of Auto-ML. Institutional Auto-ML would include extraction of appropriate studies through the no-code software platform integrated directly within the PACS framework. This process would ensure data immobility and foster a secure environment for dataset creation. Annotation would be carried out by multiple physician imagers within the same facility, enhancing the data’s accuracy and comprehensiveness specific for the institution’s patient population. This exploration highlights the strategic role of Auto-ML in advancing cancer patient care through innovative technological integration.

Editorial Special Issue on Explainable and Generalizable Deep Learning for Medical Imaging

Editorial Special Issue on Explainable and Generalizable Deep Learning for Medical Imaging

Swin MoCo: Improving parotid gland MRI segmentation using contrastive learning.

Segmentation of the parotid glands and tumors by MR images is essential for treating parotid gland tumors. However, segmentation of the parotid glands is particularly challenging due to their variable shape and low contrast with surroundingstructures. The lack of large and well-annotated datasets limits the development of deep learning in medical images. As an unsupervised learning method, contrastive learning has seen rapid development in recent years. It can better use unlabeled images and is hopeful to improve parotid glandsegmentation. We propose Swin MoCo, a momentum contrastive learning network with Swin Transformer as its backbone. The ImageNet supervised model is used as the initial weights of Swin MoCo, thus improving the training effects on small medical imagedatasets. Swin MoCo trained with transfer learning improves parotid gland segmentation to 89.78% DSC, 85.18% mIoU, 3.60 HD, and 90.08% mAcc. On the Synapse multi-organ computed tomography (CT) dataset, using Swin MoCo as the pre-trained model of Swin-Unet yields 79.66% DSC and 12.73 HD, which outperforms the best result of Swin-Unet on the Synapsedataset. The above improvements require only 4h of training on a single NVIDIA Tesla V100, which is computationally cheap. Swin MoCo provides new approaches to improve the performance of tasks on small datasets. The code is publicly available at https://github.com/Zian-Xu/Swin-MoCo.

Deep learning on medical imaging images

For a considerable amount of time, medical image processing has been an important topic of research in the field of medicine. The advent of deep learning technology has resulted in the development of revolutionary improvements in this particular sector. Because of its remarkable effectiveness in training enormous amounts of data and performing difficult tasks, deep learning has attracted a lot of attention. It has found extensive application in the analysis, diagnosis, and treatment of medical pictures. This work employs the literature review approach to examine and analyze the existing research conducted by academics on deep learning and medical imaging. It also provides a summary of the utilization of deep learning in medical imaging images and highlights the current development trend in this field.

Deep Learning in Medical Imaging: A Case Study on Lung Tissue Classification

INTRODUCTION: In the field of medical imaging, accurate categorization of lung tissue is essential for timely diagnosis and management of lung-related conditions, including cancer. Deep Learning (DL) methodologies have revolutionized this domain, promising improved precision and effectiveness in diagnosing ailments based on image analysis. This research delves into the application of DL models for classifying lung tissue, particularly focusing on histopathological imagery.&#x0D; OBJECTIVES: The primary objective of this study is to explore the deployment of DL models for the classification of lung tissue, emphasizing histopathological images. The research aims to assess the performance of various DL models in accurately distinguishing between different classes of lung tissue, including benign tissue, lung adenocarcinoma, and lung squamous cell carcinoma.&#x0D; METHODS: A dataset comprising 9,000 histopathological images of lung tissue was utilized, sourced from HIPAA compliant and validated sources. The dataset underwent augmentation to ensure diversity and robustness. The images were categorized into three distinct classes and balanced before being split into training, validation, and testing sets. Six DL models - DenseNet201, EfficientNetB7, EfficientNetB5, Vgg19, Vgg16, and Alexnet - were trained and evaluated on this dataset. Performance assessment was conducted based on precision, recall, F1-score for each class, and overall accuracy.&#x0D; RESULTS: The results revealed varying performance levels among the DL models, with EfficientNetB5 achieving perfect scores across all metrics. This highlights the capability of DL in improving the accuracy of lung tissue classification, which holds promise for enhancing diagnosis and treatment outcomes in lung-related conditions.&#x0D; CONCLUSION: This research significantly contributes to understanding the effective utilization of DL models in medical imaging, particularly for lung tissue classification. It emphasizes the critical role of a diverse and balanced dataset in developing robust and accurate models. The insights gained from this study lay the groundwork for further exploration into refining DL methodologies for medical imaging applications, with a focus on improving diagnostic accuracy and ultimately, patient outcomes.

Checklist for Reproducibility of Deep Learning in Medical Imaging

The application of deep learning (DL) in medicine introduces transformative tools with the potential to enhance prognosis, diagnosis, and treatment planning. However, ensuring transparent documentation is essential for researchers to enhance reproducibility and refine techniques. Our study addresses the unique challenges presented by DL in medical imaging by developing a comprehensive checklist using the Delphi method to enhance reproducibility and reliability in this dynamic field. We compiled a preliminary checklist based on a comprehensive review of existing checklists and relevant literature. A panel of 11 experts in medical imaging and DL assessed these items using Likert scales, with two survey rounds to refine responses and gauge consensus. We also employed the content validity ratio with a cutoff of 0.59 to determine item face and content validity. Round 1 included a 27-item questionnaire, with 12 items demonstrating high consensus for face and content validity that were then left out of round 2. Round 2 involved refining the checklist, resulting in an additional 17 items. In the last round, 3 items were deemed non-essential or infeasible, while 2 newly suggested items received unanimous agreement for inclusion, resulting in a final 26-item DL model reporting checklist derived from the Delphi process. The 26-item checklist facilitates the reproducible reporting of DL tools and enables scientists to replicate the study’s results.

Machine Learning-Based Medical Imaging Detection and Diagnostic Assistance

With the development of medical technology, medical imaging technology plays an increasingly important role in diagnosing and monitoring diseases. At the same time, machine learning technology has shown strong capabilities in data processing and analysis, providing new possibilities for the detection and auxiliary diagnosis of medical images. This paper will deeply discuss the application of machine learning in medical images, especially the application of deep learning in medical images, and how to achieve effective medical image detection and auxiliary diagnosis through steps such as preprocessing, feature extraction, model training and optimization.

AI and machine learning in medical imaging: key points from development to translation.

Innovation in medical imaging artificial intelligence (AI)/machine learning (ML) demands extensive data collection, algorithmic advancements, and rigorous performance assessments encompassing aspects such as generalizability, uncertainty, bias, fairness, trustworthiness, and interpretability. Achieving widespread integration of AI/ML algorithms into diverse clinical tasks will demand a steadfast commitment to overcoming issues in model design, development, and performance assessment. The complexities of AI/ML clinical translation present substantial challenges, requiring engagement with relevant stakeholders, assessment of cost-effectiveness for user and patient benefit, timely dissemination of information relevant to robust functioning throughout the AI/ML lifecycle, consideration of regulatory compliance, and feedback loops for real-world performance evidence. This commentary addresses several hurdles for the development and adoption of AI/ML technologies in medical imaging. Comprehensive attention to these underlying and often subtle factors is critical not only for tackling the challenges but also for exploring novel opportunities for the advancement of AI in radiology.

Deep Learning Models for Classification of Lung Diseases

This thesis focuses on the importance of early detection in lung cancer through the use of medical imaging techniques and deep learning models. The current practice of examining nodules larger than 7 mm can delay detection and allow cancerous nodules to grow undetected. The project aims to detect nodules as small as 3 mm to improve the chances of early cancer identification. The use of constrained volume datasets and transfer learning techniques addresses the scarcity of medical data, and deep neural networks are employed for classification and segmentation tasks. Despite the limited dataset, the results demonstrate the effectiveness of the proposed models. Class activation maps and segmentation techniques enhance accuracy and provide insights into the most critical areas for diagnosis. This research contributes to the understanding of lung disease diagnosis and highlights the potential of deep learning in medical imaging.

A causal perspective on dataset bias in machine learning for medical imaging

A causal perspective on dataset bias in machine learning for medical imaging