Search

Objectives: This study aimed to develop an artificial intelligence (Al) model that can determine automatic tooth numbering, frenulum attachments, gingival overgrowth areas, and gingival inflammation signs on intraoral photographs and to evaluate the performance of this model. Method and materials: A total of 654 intraoral photographs were used in the study (n = 654). All photographs were reviewed by three periodontists, and all teeth, frenulum attachment, gingival overgrowth areas, and gingival inflammation signs on photographs were labeled using the segmentation method in a web-based labeling software. In addition, tooth numbering was carried out according to the FDI system. An Al model was developed with the help of YOLOv5x architecture with labels of 16,795 teeth, 2,493 frenulum attachments, 1,211 gingival overgrowth areas, and 2,956 gingival inflammation signs. The confusion matrix system and ROC (receiver operator characteristic) analysis were used to statistically evaluate the success of the developed model. Results: The sensitivity, precision, F1 score, and AUC (area under the curve) for tooth numbering were 0.990, 0.784, 0.875, and 0.989; for frenulum attachment these were 0.894, 0.775,0.830, and 0.827; for gingival overgrowth area these were 0.757, 0.675, 0.714, and 0.774; and for gingival inflammation sign 0.737, 0.823, 0.777, and 0.802, respectively. Conclusion: The results of the present study show that Al systems can be successfully used to interpret intraoral photographs. These systems have the potential to accelerate the digital transformation in the clinical and academic functioning of dentistry with the automatic determination of anatomical structures and dental conditions from intraoral photographs. [ABSTRACT FROM AUTHOR]

Brain metastases present a formidable challenge in cancer management due to the infiltration of malignant cells from distant sites into the brain. Precise segmentation of brain metastases (BM) in medical imaging is vital for treatment planning and assessment. Leveraging deep learning techniques has shown promise in automating BM identification, facilitating faster and more accurate detection. This paper aims to develop an innovative novel deep learning model tailored for BM segmentation, addressing current approach limitations. Utilizing a comprehensive dataset of annotated magnetic resonance imaging (MRI) from Stanford University, the proposed model will undergo thorough evaluation using standard performance metrics. Comparative analysis with existing segmentation methods will highlight the superior performance and efficacy of our model. The anticipated outcome of this research is a highly accurate and efficient deep learning model for brain metastasis segmentation. Such a model holds potential to enhance treatment planning, monitoring, and ultimately improve patient care and clinical outcomes in managing brain metastases. [ABSTRACT FROM AUTHOR]

Explainable Artificial Intelligence (XAI) is an emerging machine learning field that has been successful in medical image analysis. Interpretable approaches are able to "unbox" the black-box decisions made by AI systems, aiding medical doctors to justify their diagnostics better. In this paper, we analyze the performance of three different XAI strategies for medical image analysis in ophthalmology. We consider a multimodal deep learning model that combines optical coherence tomography (OCT) and infrared reflectance (IR) imaging for the diagnosis of age-related macular degeneration (AMD). The classification model is able to achieve an accuracy of 0.94, performing better than other unimodal alternatives. We analyze the XAI methods in terms of their ability to identify retinal damage and ease of interpretation, concluding that grad-CAM and guided grad-CAM can be combined to have both a coarse visual justification and a fine-grained analysis of the retinal layers. We provide important insights and recommendations for practitioners on how to design automated and explainable screening tests based on the combination of two image sources. [ABSTRACT FROM AUTHOR]

Summary: Deep neural networks for medical image segmentation often face the problem of insufficient clean labeled data. Although non‐expert annotations are more readily accessible, these low‐quality annotations lead to significant performance degradation of existing neural network methods. In this paper, we focus on robust learning of medical image segmentation with noisy annotations and propose a novel noise‐tolerant framework based on dual‐strategy sample selection, which selects the informative samples to provide effective supervision information. First, we propose the first round of sample selection by designing a novel joint loss, which includes conventional supervised loss and regularization loss. To further select information‐rich samples, we propose confidence‐based pseudo‐label sample selection from a novel perspective as the complement. The dual strategies are used in a collaborative manner and the network is optimized with mined informative samples. We conducted extensive experiments on datasets with both simulated noisy labels and real‐world noisy labels. For instance, on a simulated dataset with 25% noise ratio, our method achieves segmentation Dice value with 90.56% ±$$ \pm $$ 0.03%. Furthermore, increasing the noise ratio to 95%, our method still maintains a high Dice value of 73.85% ±$$ \pm $$ 0.28% compared to other baselines. Extensive results have demonstrated that our method can weaken the effects of noisy labels on medical image segmentation. [ABSTRACT FROM AUTHOR]

The rapid advancement of technology has brought significant changes to various fields, including medical imaging (MI). This discussion paper explores the integration of computing technologies (e.g. Python and MATLAB), digital image processing (e.g. image enhancement, segmentation and three‐dimensional reconstruction) and artificial intelligence (AI) into the undergraduate MI curriculum. By examining current educational practices, gaps and limitations that hinder the development of future‐ready MI professionals are identified. A comprehensive curriculum framework is proposed, incorporating essential computational skills, advanced image processing techniques and state‐of‐the‐art AI tools, such as large language models like ChatGPT. The proposed curriculum framework aims to improve the quality of MI education significantly and better equip students for future professional practice and challenges while enhancing diagnostic accuracy, improving workflow efficiency and preparing students for the evolving demands of the MI field. [ABSTRACT FROM AUTHOR]

The article discusses the use of deep learning in image segmentation for cancer management, emphasizing the importance of medical imaging in cancer diagnosis and treatment. It highlights the role of quantitative imaging in characterizing tumors and monitoring treatment response. The text explores the application of deep learning models like U-Nets and CNNs in medical imaging, specifically for image segmentation tasks. Two studies are referenced, showcasing the effectiveness of AI models in automating image segmentation processes and improving accuracy in cancer management. [Extracted from the article]

Multi-modal image analysis using deep learning (DL) lays the foundation for neoadjuvant treatment (NAT) response monitoring. However, existing methods prioritize extracting multi-modal features to enhance predictive performance, with limited consideration on real-world clinical applicability, particularly in longitudinal NAT scenarios with multi-modal data. Here, we propose the Multi-modal Response Prediction (MRP) system, designed to mimic real-world physician assessments of NAT responses in breast cancer. To enhance feasibility, MRP integrates cross-modal knowledge mining and temporal information embedding strategy to handle missing modalities and remain less affected by different NAT settings. We validated MRP through multi-center studies and multinational reader studies. MRP exhibited comparable robustness to breast radiologists, outperforming humans in predicting pathological complete response in the Pre-NAT phase (ΔAUROC 14% and 10% on in-house and external datasets, respectively). Furthermore, we assessed MRP's clinical utility impact on treatment decision-making. MRP may have profound implications for enrolment into NAT trials and determining surgery extensiveness. Deep learning for medical image analysis is a promising new avenue to predict treatment response, however the clinical application of these methods has been so far limited. Here, the authors propose a model to predict chemotherapy response in breast cancer in real world clinical settings. [ABSTRACT FROM AUTHOR]

Introduction: Colorectal cancer (CRC) is one of the main causes of deaths worldwide. Early detection and diagnosis of its precursor lesion, the polyp, is key to reduce its mortality and to improve procedure efficiency. During the last two decades, several computational methods have been proposed to assist clinicians in detection, segmentation and classification tasks but the lack of a common public validation framework makes it difficult to determine which of them is ready to be deployed in the exploration room. Methods: This study presents a complete validation framework and we compare several methodologies for each of the polyp characterization tasks. Results: Results show that the majority of the approaches are able to provide good performance for the detection and segmentation task, but that there is room for improvement regarding polyp classification. Discussion: While studied show promising results in the assistance of polyp detection and segmentation tasks, further research should be done in classification task to obtain reliable results to assist the clinicians during the procedure. The presented framework provides a standarized method for evaluating and comparing different approaches, which could facilitate the identification of clinically prepared assisting methods. [ABSTRACT FROM AUTHOR]

Objectives: 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. Methods: 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. Results: 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. Conclusions: 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. [ABSTRACT FROM AUTHOR]

Simple Summary: Patients with skin lesions suspicious for skin cancer or atypical melanocytic nevi of uncertain malignant potential often present to dermatologists, who may have variable dermoscopy triage clinical experience. We evaluated the clinical utility of a digital dermoscopy image-based artificial intelligence algorithm (DDI-AI device) on the diagnosis and management of skin cancers. Thirty-six United States board-certified dermatologists evaluated 50 cases with clinical images and DDI of the same skin lesions (25 malignant and 25 benign), first without and then with knowledge of the DDI-AI device output. We identified that when using the DDI-AI device, management sensitivity, management specificity, diagnostic sensitivity, and diagnostic specificity all increased over baseline use of clinical images and dermoscopy images. Hence, the use of the DDI-AI device may quickly, safely, and effectively improve dermoscopy performance, skin cancer diagnosis, and management when used by dermatologists, independent of training and experience. Background: Patients with skin lesions suspicious for skin cancer or atypical melanocytic nevi of uncertain malignant potential often present to dermatologists, who may have variable dermoscopy triage clinical experience. Objective: To evaluate the clinical utility of a digital dermoscopy image-based artificial intelligence algorithm (DDI-AI device) on the diagnosis and management of skin cancers by dermatologists. Methods: Thirty-six United States board-certified dermatologists evaluated 50 clinical images and 50 digital dermoscopy images of the same skin lesions (25 malignant and 25 benign), first without and then with knowledge of the DDI-AI device output. Participants indicated whether they thought the lesion was likely benign (unremarkable) or malignant (suspicious). Results: The management sensitivity of dermatologists using the DDI-AI device was 91.1%, compared to 84.3% with DDI, and 70.0% with clinical images. The management specificity was 71.0%, compared to 68.4% and 64.9%, respectively. The diagnostic sensitivity of dermatologists using the DDI-AI device was 86.1%, compared to 78.8% with DDI, and 63.4% with clinical images. Diagnostic specificity using the DDI-AI device increased to 80.7%, compared to 75.9% and 73.6%, respectively. Conclusion: The use of the DDI-AI device may quickly, safely, and effectively improve dermoscopy performance, skin cancer diagnosis, and management when used by dermatologists, independent of training and experience. [ABSTRACT FROM AUTHOR]

Background: Tuberculomas are prevalent in developing countries and demonstrate variable signals on MRI resulting in the overlap of the conventional imaging phenotype with other entities including glioma and brain metastasis. An accurate MRI diagnosis is important for the early institution of anti-tubercular therapy, decreased patient morbidity, mortality, and prevents unnecessary neurosurgical excision. This study aims to assess the potential of radiomics features of regular contrast images including T1W, T2W, T2W FLAIR, T1W post contrast images, and ADC maps, to differentiate between tuberculomas, high-grade-gliomas and metastasis, the commonest intra parenchymal mass lesions encountered in the clinical practice. Methods: This retrospective study includes 185 subjects. Images were resampled, co-registered, skull-stripped, and zscore-normalized. Automated lesion segmentation was performed followed by radiomics feature extraction, train-test split, and features reduction. All machine learning algorithms that natively support multiclass classification were trained and assessed on features extracted from individual modalities as well as combined modalities. Model explainability of the best performing model was calculated using the summary plot obtained by SHAP values. Results: Extra tree classifier trained on the features from ADC maps was the best classifier for the discrimination of tuberculoma from high-grade-glioma and metastasis with AUC-score of 0.96, accuracy-score of 0.923, Brier-score of 0.23. Conclusion: This study demonstrates that radiomics features are effective in discriminating between tuberculoma, metastasis, and high-grade-glioma with notable accuracy and AUC scores. Features extracted from the ADC maps surfaced as the most robust predictors of the target variable. [ABSTRACT FROM AUTHOR]

Aim: Deep learning (DL) algorithms can be used for automated analysis of medical imaging. The aim of this study was to assess the accuracy of an innovative, fully automated DL algorithm for analysis of sagittal balance in adult spinal deformity (ASD). Material and methods: Sagittal balance (sacral slope, pelvic tilt, pelvic incidence, lumbar lordosis and sagittal vertical axis) was evaluated in 141 preoperative and postoperative radiographs of patients with ASD. The DL, landmark-based measurements, were compared with the ground truth values from validated manual measurements. Results: The DL algorithm showed an excellent consistency with the ground truth measurements. The intra-class correlation coefficient between the DL and ground truth measurements was 0.71–0.99 for preoperative and 0.72–0.96 for postoperative measurements. The DL detection rate was 91.5% and 84% for preoperative and postoperative images, respectively. Conclusion: This is the first study evaluating a complete automated DL algorithm for analysis of sagittal balance with high accuracy for all evaluated parameters. The excellent accuracy in the challenging pathology of ASD with long construct instrumentation demonstrates the eligibility and possibility for implementation in clinical routine. [ABSTRACT FROM AUTHOR]

Purpose: Generating polar map (PM) from [68Ga]Ga-DOTA-FAPI-04 PET images is challenging and inaccurate using existing automatic methods that rely on the myocardial anatomical integrity in PET images. This study aims to enhance the accuracy of PM generated from [68Ga]Ga-DOTA-FAPI-04 PET images and explore the potential value of PM in detecting reactive fibrosis after myocardial infarction and assessing its relationship with cardiac function. Methods: We proposed a deep-learning-based method that fuses multi-modality images to compensate for the cardiac structural information lost in [68Ga]Ga-DOTA-FAPI-04 PET images and accurately generated PMs. We collected 133 pairs of [68Ga]Ga-DOTA-FAPI-04 PET/MR images from 87 ST-segment elevated myocardial infarction patients for training and evaluation purposes. Twenty-six patients were selected for longitudinal analysis, further examining the clinical value of PM-related imaging parameters. Results: The quantitative comparison demonstrated that our method was comparable with the manual method and surpassed the commercially available software-PMOD in terms of accuracy in generating PMs for [68Ga]Ga-DOTA-FAPI-04 PET images. Clinical analysis revealed the effectiveness of [68Ga]Ga-DOTA-FAPI-04 PET PM in detecting reactive myocardial fibrosis. Significant correlations were demonstrated between the difference of baseline PM FAPI% and PM LGE%, and the change in cardiac function parameters (all p < 0.001), including LVESV% (r = 0.697), LVEDV% (r = 0.621) and LVEF% (r = -0.607). Conclusion: The [68Ga]Ga-DOTA-FAPI-04 PET PMs generated by our method are comparable to manually generated and sufficient for clinical use. The PMs generated by our method have potential value in detecting reactive fibrosis after myocardial infarction and were associated with cardiac function, suggesting the possibility of enhancing clinical diagnostic practices. Trial registration: ClinicalTrials.gov (NCT04723953). Registered 26 January 2021. [ABSTRACT FROM AUTHOR]

Background: Deep learning (DL) techniques have been extensively applied in medical image classification. The unique characteristics of medical imaging data present challenges, including small labeled datasets, severely imbalanced class distribution, and significant variations in imaging quality. Recently, generative adversarial network (GAN)‐based classification methods have gained attention for their ability to enhance classification accuracy by incorporating realistic GAN‐generated images as data augmentation. However, the performance of these GAN‐based methods often relies on high‐quality generated images, while large amounts of training data are required to train GAN models to achieve optimal performance. Purpose: In this study, we propose an adversarial learning‐based classification framework to achieve better classification performance. Innovatively, GAN models are employed as supplementary regularization terms to support classification, aiming to address the challenges described above. Methods: The proposed classification framework, GAN‐DL, consists of a feature extraction network (F‐Net), a classifier, and two adversarial networks, specifically a reconstruction network (R‐Net) and a discriminator network (D‐Net). The F‐Net extracts features from input images, and the classifier uses these features for classification tasks. R‐Net and D‐Net have been designed following the GAN architecture. R‐Net employs the extracted feature to reconstruct the original images, while D‐Net is tasked with the discrimination between the reconstructed image and the original images. An iterative adversarial learning strategy is designed to guide model training by incorporating multiple network‐specific loss functions. These loss functions, serving as supplementary regularization, are automatically derived during the reconstruction process and require no additional data annotation. Results: To verify the model's effectiveness, we performed experiments on two datasets, including a COVID‐19 dataset with 13 958 chest x‐ray images and an oropharyngeal squamous cell carcinoma (OPSCC) dataset with 3255 positron emission tomography images. Thirteen classic DL‐based classification methods were implemented on the same datasets for comparison. Performance metrics included precision, sensitivity, specificity, and F1$F_1$‐score. In addition, we conducted ablation studies to assess the effects of various factors on model performance, including the network depth of F‐Net, training image size, training dataset size, and loss function design. Our method achieved superior performance than all comparative methods. On the COVID‐19 dataset, our method achieved 95.4%±0.6%$95.4\%\pm 0.6\%$, 95.3%±0.9%$95.3\%\pm 0.9\%$, 97.7%±0.4%$97.7\%\pm 0.4\%$, and 95.3%±0.9%$95.3\%\pm 0.9\%$ in terms of precision, sensitivity, specificity, and F1$F_1$‐score, respectively. It achieved 96.2%±0.7%$96.2\%\pm 0.7\%$ across all these metrics on the OPSCC dataset. The study to investigate the effects of two adversarial networks highlights the crucial role of D‐Net in improving model performance. Ablation studies further provide an in‐depth understanding of our methodology. Conclusion: Our adversarial‐based classification framework leverages GAN‐based adversarial networks and an iterative adversarial learning strategy to harness supplementary regularization during training. This design significantly enhances classification accuracy and mitigates overfitting issues in medical image datasets. Moreover, its modular design not only demonstrates flexibility but also indicates its potential applicability to various clinical contexts and medical imaging applications. [ABSTRACT FROM AUTHOR]

Eye tracking has become increasingly applied in medical education research for studying the cognitive processes that occur during the performance of a task, such as image interpretation and surgical skills development. However, analysis and interpretation of the large amount of data obtained by eye tracking can be confusing. In this article, our intention is to clarify the analysis and interpretation of the data obtained from eye tracking. Understanding the relationship between eye tracking metrics (such as gaze, pupil and blink rate) and cognitive processes (such as visual attention, perception, memory and cognitive workload) is essential. The importance of calibration and how the limitations of eye tracking can be overcome is also highlighted. [ABSTRACT FROM AUTHOR]

Purpose: Fractal analysis is a mathematical tool which allows the evaluation of complex microstructural features within materials that cannot be expressed in traditional geometric terms. The purpose of this study is to quantify the differences in polymethylmethacrylate intravertebral cement spatial distribution patterns following vertebroplasty using fractal analysis through the examination of osteoporotic and malignant compression fractures. Methods: Frontal and lateral post-vertebroplasty radiographs were evaluated from 29 patients with osteoporotic and malignant compression fractures who underwent vertebroplasty. The individually treated vertebra were divided into osteoporotic (n = 35) and malignant groups (n = 41). Images underwent segmentation, thresholding, and binarization prior to fractal analysis. Fractal dimension and lacunarity values were derived from the region of interest in treated vertebrae using the "box-counting" and "gliding-box" techniques respectively using ImageJ. The mean values of both parameters were compared between the 2 groups. Results: The mean fractal dimension was significantly higher in the malignant vertebral compression fracture group (1.53 ± 0.08) compared to the osteoporotic group (1.34 ± 0.17; P <.001). Similarly, mean lacunarity values were significantly higher in the malignant fracture group (0.50 ± 0.09) compared to the osteoporotic group (0.37 ± 0.10; P <.001). Conclusions: Fractal dimension and lacunarity values of cement spatial distribution patterns obtained from the post-vertebroplasty radiographs can differentiate between benign osteoporotic and malignant vertebral compression fractures. This novel technique may be useful for evaluating cement spatial distribution patterns in spine augmentation procedures, although further research is warranted in this area. [ABSTRACT FROM AUTHOR]

Purpose: To evaluate factors impacting the Segment Anything Model (SAM) and variant MedSAM performance for segmenting liver observations on contrast-enhanced (CE) magnetic resonance imaging (MRI) in high-risk patients with probable hepatocellular carcinoma (HCC) (LR-4) and definite HCC (LR-5). Methods: A retrospective cohort of liver observations (LR-4/LR-5) on CE-MRI from 97 patients at high-risk for HCC was derived (2013-2018). Using bounding-boxes as prompts under 5-fold cross-validation, segmentation performance was evaluated at the model and liver observation-levels for: (1) model types: SAM versus MedSAM, (2) image sizes: 256 × 256 versus 512 × 512, (3) image channel composition: CE sequences at 3 phases of enhancement independently and combined, (4) liver observation size: >10 mm versus >20 mm, (5) certainty of diagnosis: LR-4 versus LR-5, and (6) contrast-agent type: hepatobiliary versus extracellular. Segmentation performance, quantified using Dice coefficient, were compared using univariate (Wilcoxon signed-rank and t -test) and multivariable analyses (multiple correspondence analysis and subsequent linear modelling). Results: MedSAM trained on 512 × 512 combined CE sequences performed best with mean Dice coefficient 0.68 (95% confidence interval 0.66, 0.69). Overall, all factors except contrast-agent type affected performance, with larger image size resulting in the highest performance improvement (512 × 512: 0.57, 256 × 256: 0.26, P <.001) at the model-level. Contrast-agents affected performance for patients with LR-4 observations using MedSAM-based models (P <.03). Larger observation size, image size, and higher certainty of diagnosis were associated with better segmentation on multivariable analysis. Conclusion: A variety of factors were found to impact SAM/MedSAM performance for segmenting liver observations in patients with probable and definite HCC on CE-MRI. Future models may be optimized by accounting for these factors. [ABSTRACT FROM AUTHOR]

Lung cancer, a leading cause of global mortality, demands a combat for its effective prevention, early diagnosis, and advanced treatment methods. Traditional diagnostic methods face limitations in accuracy and efficiency, necessitating innovative solutions. Large Language Models (LLMs) and Natural Language Processing (NLP) offer promising avenues for overcoming these challenges by providing comprehensive insights into medical data and personalizing treatment plans. This systematic review explores the transformative potential of LLMs and NLP in automating lung cancer diagnosis. It evaluates their applications, particularly in medical imaging and the interpretation of complex medical data, and assesses achievements and associated challenges. Emphasizing the critical role of Artificial Intelligence (AI) in medical imaging, the review highlights advancements in lung cancer screening and deep learning approaches. Furthermore, it underscores the importance of on‐going advancements in diagnostic methods and encourages further exploration in this field. [ABSTRACT FROM AUTHOR]

The article discusses the comparison between two radiomics research quality scoring tools, METRICS and RQS, highlighting their unique features and differences. METRICS, endorsed by leading imaging societies, aims to standardize the methodological quality assessment of radiomics research, while RQS has been widely adopted in the research community despite its limitations. The article presents hypothetical scenarios to compare the two tools, emphasizing the distinct focus of each on different aspects of the radiomic pipeline. Researchers are advised to consider the unique features of each tool based on the insights provided in the analysis. [Extracted from the article]

Breast cancer is widespread across the globe. It's the primary cause of death in cancer fatalities. According to the Palestinian Ministry of Health annual report, it ranked as the third reported death of all reported cancer deaths in the West Bank. Mammogram screening is the most common technique to diagnose breast abnormalities, but there is a challenge in the lack of skilled experts able to accurately interpret mammograms. Machine learning plays an important role in medical image processing particularly in early detection when the treatment is less expensive and available. In this paper we proposed different convolutional neural network (CNN) models to detect breast abnormalities with promising results. Six CNN models were used in this research on a unique (first-hand) dataset collected from the Palestinian Ministry of Health. The models are VGG16, VGG19, DenseNet121, ResNet50, Xception, and EfficientNetB7. Consequently, DenseNet121 outperformed other models with 0.83 and 0.85 for testing accuracy and area under curve (AUC) respectively. As a future work, the outperformed model can be combined with other patient data like genetic information, medical history, and lifestyle factors to evaluate the risk of developing specific diseases. This would increase the survival rate and enable proactive measures. [ABSTRACT FROM AUTHOR]

Periapical periodontitis may manifest as a radiographic lesion radiographically. Periapical lesions are amongst the most common dental pathologies that present as periapical radiolucencies on panoramic radiographs. The objective of this research is to assess the diagnostic accuracy of an artificial intelligence (AI) model based on U²-Net architecture in the detection of periapical lesions on dental panoramic radiographs and to determine whether they can be useful in aiding clinicians with diagnosis of periapical lesions and improving their clinical workflow. 400 panoramic radiographs that included at least one periapical radiolucency were selected retrospectively. 780 periapical radiolucencies in these anonymized radiographs were manually labeled by two independent examiners. These radiographs were later used to train the AI model based on U²-Net architecture trained using a deep supervision algorithm. An AI model based on the U²-Net architecture was implemented. The model achieved a dice score of 0.8 on the validation set and precision, recall, and F1-score of 0.82, 0.77, and 0.8 respectively on the test set. This study has shown that an AI model based on U²-Net architecture can accurately diagnose periapical lesions on panoramic radiographs. The research provides evidence that AI-based models have promising applications as adjunct tools for dentists in diagnosing periapical radiolucencies and procedure planning. Further studies with larger data sets would be required to improve the diagnostic accuracy of AI-based detection models. [ABSTRACT FROM AUTHOR]

With the rapid advancement in artificial intelligence (AI) and its application in the Internet of Things (IoT), intelligent technologies are being introduced in the medical field, giving rise to smart healthcare systems. The medical imaging data contains sensitive information, which can easily be stolen or tampered with, necessitating secure encryption schemes designed specifically to protect these images. This paper introduces an artificial intelligence-driven novel encryption scheme tailored for the secure transmission and storage of high-resolution medical images. The proposed scheme utilizes an artificial intelligence-based autoencoder to compress high-resolution medical images and to facilitate fast encryption and decryption. The proposed autoencoder retains important diagnostic information even after reducing the image dimensions. The low-resolution images then undergo a four-stage encryption process. The first two encryption stages involve permutation and the next two stages involve confusion. The first two stages ensure the disruption of the structure of the image, making it secure against statistical attacks. Whereas the two stages of confusion ensure the effective concealment of the pixel values making it difficult to decrypt without secret keys. This encrypted image is then safe for storage or transmission. The proposed scheme has been extensively evaluated against various attacks and statistical security parameters confirming its effectiveness in securing medical image data. [ABSTRACT FROM AUTHOR]

Medical image machines serve as a valuable tool to monitor and diagnose a variety of diseases. However, manual and centralized interpretation are both error-prone and time-consuming due to malicious attacks. Numerous diagnostic algorithms have been developed to improve precision and prevent poisoning attacks by integrating symptoms, test methods, and imaging data. But in today's digital technology world, it is necessary to have a global cloud-based diagnostic artificial intelligence model that is efficient in diagnosis and preventing poisoning attacks and might be used for multiple purposes. We propose the Healthcare Federated Ensemble Internet of Learning Cloud Doctor System (FDEIoL) model, which integrates different Internet of Things (IoT) devices to provide precise and accurate interpretation without poisoning attack problems, thereby facilitating IoT-enabled remote patient monitoring for smart healthcare systems. Furthermore, the FDEIoL system model uses a federated ensemble learning strategy to provide an automatic, up-to-date global prediction model based on input local models from the medical specialist. This assures biomedical security by safeguarding patient data and preserving the integrity of diagnostic processes. The FDEIoL system model utilizes local model feature selection to discriminate between malicious and non-malicious local models, and ensemble strategies use positive and negative samples to optimize the performance of the test dataset, enhancing its capability for remote patient monitoring. The FDEIoL system model achieved an exceptional accuracy rate of 99.24% on the Chest X-ray dataset and 99.0% on the MRI dataset of brain tumors compared to centralized models, demonstrating its ability for precision diagnosis in IoT-enabled healthcare systems. [ABSTRACT FROM AUTHOR]

Accurate medical image segmentation plays a vital role in clinical practice. Convolutional Neural Network and Transformer are mainstream architectures for this task. However, convolutional neural network lacks the ability of modeling global dependency while Transformer cannot extract local details. In this paper, we propose DATTNet, DualATTentionNetwork, an encoder-decoder deep learning model for medical image segmentation. DATTNet is exploited in hierarchical fashion with two novel components: (1) Dual Attention module is designed to model global dependency in spatial and channel dimensions. (2) Context Fusion Bridge is presented to remix the feature maps with multiple scales and construct their correlations. The experiments on ACDC, Synapse and Kvasir-SEG datasets are conducted to evaluate the performance of DATTNet. Our proposed model shows superior performance, effectiveness and robustness compared to SOTA methods, with mean Dice Similarity Coefficient scores of 92.2%, 84.5% and 89.1% on cardiac, abdominal organs and gastrointestinal poly segmentation tasks. The quantitative and qualitative results demonstrate that our proposed DATTNet attains favorable capability across different modalities (MRI, CT, and endoscopy) and can be generalized to various tasks. Therefore, it is envisaged as being potential for practicable clinical applications. The code has been released on https://github.com/MhZhang123/DATTNet/tree/main. [ABSTRACT FROM AUTHOR]

In recent times, automated detection of diseases from pathological images leveraging Machine Learning (ML) models has become fairly common, where the ML models learn detecting the disease by identifying biomarkers from the images. However, such an approach requires the models to be trained on a vast amount of data, and healthcare organizations often tend to limit access due to privacy concerns. Consequently, collecting data for traditional centralized training becomes challenging. These privacy concerns can be handled by Federation Learning (FL), which builds an unbiased global model from local models trained with client data while maintaining the confidentiality of local data. Using FL, this study solves the problem of centralized data collection by detecting deformations in images of Red Blood Cells (RBC) in a decentralized way. To achieve this, RBC data is used to train multiple Deep Learning (DL) models, and among the various DL models, the most efficient one is considered to be used as the global model inside the FL framework. The FL framework works by copying the global model's weight to the client's local models and then training the local models in client-specific devices to average the weights of the local model back to the global model. In the averaging process, direct averaging is performed and alongside, weighted averaging is also done to weigh the individual local model's contribution according to their performance, keeping the FL framework immune to the effects of bad clients and attacks. In the process, the data of the client remains confidential during training, while the global model learns necessary information. The results of the experiments indicate that there is no significant difference in the performance of the FL method and the best-performing DL model, as the best-performing DL model reaches an accuracy of 96% and the FL environment reaches 94%-95%. This study shows that the FL technique, in comparison to the classic DL methodology, can accomplish confidentiality secured RBC deformation classification from RBC images without substantially diminishing the accuracy of the categorization. Finally, the overall validity of the classification results has been verified by employing GradCam driven Explainable AI techniques. [ABSTRACT FROM AUTHOR]

Background: Breast cancer (BC) is the most common malignancy in women and the second cause of cancer death. In recent years, there has been a strong development in artificial intelligence (AI) applications in medical imaging for several tasks. Our aim was to evaluate the potential of transfer learning with convolutional neural networks (CNNs) in discriminating suspicious breast lesions on ultrasound images. Methods: Transfer learning performances of five different CNNs (Inception V3, Xception, Densenet121, VGG 16, and ResNet50) were evaluated on a public and on an institutional dataset (526 and 392 images, respectively), customizing the top layers for the specific task. Institutional images were contoured by an expert radiologist and processed to feed the CNNs for training and testing. Postimaging biopsies were used as a reference standard for classification. The area under the receiver operating curve (AUROC) was used to assess diagnostic performance. Results: Networks performed very well on the public dataset (AUROC 0.938–0.996). The direct generalization to the institutional dataset resulted in lower performances (max AUROC 0.676); however, when tested on BI-RADS 3 and BI-RADS 5 only, results were improved (max AUROC 0.792). Good results were achieved on the institutional dataset (AUROC 0.759–0.818) and, when selecting a threshold of 2% for classification, a sensitivity of 0.983 was obtained for three of five CNNs, with the potential to spare biopsy in 15.3%–18.6% of patients. Conclusion: In conclusion, transfer learning with CNNs may achieve high sensitivity and might be used as a support tool in managing suspicious breast lesions on ultrasound images. Relevance statement: Transfer learning is a powerful technique to exploit the performances of well-trained CNNs for image classification. In a clinical scenario, it might be useful for the management of suspicious breast lesions on breast ultrasound, potentially sparing biopsy in a non-negligible number of patients. Key Points: Properly trained CNNs with transfer learning are highly effective in differentiating benign and malignant lesions on breast ultrasound. Setting clinical thresholds increased sensitivity. CNNs might be useful as support tools in managing suspicious lesions on breast ultrasound. [ABSTRACT FROM AUTHOR]

Background: Alzheimer's disease (AD) is a complex neurodegenerative disorder characterized by progressively worsening cognitive decline and memory loss. Excessive iron accumulation produces severe cognitive impairment. However, there are no uniform conclusions about changes in brain iron content in AD. This study aimed to investigate the iron content of the deep brain nuclei in AD, and its correlation with cognitive function. Methods: Thirty-one patients with mild to moderate AD, 17 patients with mild cognitive impairment (MCI), and 20 age-, sex-, and education-matched healthy controls (HC) were collected. The QSM was used to quantify the magnetic susceptibility values of the caudate nucleus, putamen, globus pallidus, substantia nigra, red nucleus, and dentate nucleus, and to analyze the differences that existed between the three groups. As well as the correlation between the magnetic susceptibility values and cognitive function was calculated. Results: The magnetic susceptibility values of bilateral globus pallidus, left putamen, and bilateral substantia nigra were significantly higher in AD patients than in HC, and the magnetic susceptibility values of the right globus pallidus were significantly higher in AD patients than in MCI (all p < 0.05). The magnetic susceptibility values of the left dentate nucleus in the AD group were negatively correlated with the writing function of the MMSE subitem (r = -0.42, p = 0.020), and the magnetic susceptibility values of the left caudate nucleus and right dentate nucleus were significantly and negatively correlated with the naming function and language function of the MoCA subitem, respectively (r = -0.43, p = 0.019; r = -0.36, p = 0.048). Conclusion: Magnetic susceptibility values based on QSM correlate with cognitive function are valuable in discriminating AD from MCI and AD from HC. [ABSTRACT FROM AUTHOR]

This study aims to explore an autoencoder-based method for generating brain MRI images of patients with Autism Spectrum Disorder (ASD) and non-ASD individuals, and to discriminate ASD based on the generated images. Initially, we introduce the research background of ASD and related work, as well as the application of deep learning in the field of medical imaging. Subsequently, we detail the architecture and training process of the proposed autoencoder model, and present the results of generating MRI images for ASD and non-ASD patients. Following this, we designed an ASD classifier based on the generated images and elucidated its structure and training methods. Finally, through analysis and discussion of experimental results, we validated the effectiveness of the proposed method and explored future research directions and potential clinical applications. This research offers new insights and methodologies for addressing challenges in ASD studies using deep learning technology, potentially contributing to the automated diagnosis and research of ASD. [ABSTRACT FROM AUTHOR]

Objective: Intraoral photographic images are instrumental in the early screening and clinical diagnosis of oral diseases. In addition, people have been trying to apply artificial intelligence to these images. The purpose of this study is to investigate and evaluate a deep learning system designed to segment intraoral photographic images for the detection of dental caries, dental calculus, and gingivitis, and to assess the degree of dental calculus based on the overall features of the tooth surface and gingival margin. Material and methods: This cross-sectional study collected 3,365 oral endoscopic images, randomly distributed in training datasets (2,019 images), validation dataset (673 images), and test dataset (673 images). The training set and verification set images are manually labeled. An oral endoscopic image segmentation method based on Mamba (Oral-Mamba) and an intelligent evaluation model of dental calculus degree were proposed, achieving the segmentation of two types of oral diseases, namely gingivitis and dental caries, as well as the segmentation of dental calculus regions, and the intelligent evaluation of the degree of dental calculus. Results: Oral-Mamba demonstrated high accuracy in segmentation, with accuracy rates for gingivitis, dental caries, and dental calculus at 0.83, 0.83, and 0.81, respectively. In particular, these rates surpassed those of the U-Net model in IoU, accuracy, and recall metrics. Furthermore, Oral-Mamba runs 25% faster than U-Net.The accuracy of degree classification in the intelligent evaluation model of dental calculus degree is 85%. Conclusion: The proposed deep learning system is expected to be used for the detection of two types of oral diseases and dental calculus, and the degree judgment of photographic images from an intraoral camera. This system offers a practical method to assist in the oral screening of dental caries, dental calculus, and gingivitis, providing benefits such as intuitive use, time efficiency, cost-effectiveness, and ease of deployment. [ABSTRACT FROM AUTHOR]

Multimodal medical image fusion methods, which combine complementary information from many multi-modality medical images, are among the most important and practical approaches in numerous clinical applications. Various conventional image fusion techniques have been developed for multimodality image fusion. Complex procedures for weight map computing, fixed fusion strategy and lack of contextual understanding remain difficult in conventional and machine learning approaches, usually resulting in artefacts that degrade the image quality. This work proposes an efficient hybrid learning model for medical image fusion using pre-trained and non-pre-trained networks i.e. VGG-19 and SNN with stacking ensemble method. The model leveraging the unique capabilities of each architecture, can effectively preserve the detailed information with high visual quality, for numerous combinations of image modalities in image fusion challenges, notably improved contrast, increased resolution, and lower artefacts. Additionally, this ensemble model can be more robust in the fusion of various combinations of source images that are publicly available from Havard-Medical-Image-Fusion Datasets, GitHub. and Kaggle. Our proposed model performance is superior in terms of visual quality and performance metrics to that of the existing fusion methods in literature like PCA+DTCWT, NSCT, DWT, DTCWT+NSCT, GADCT, CNN and VGG-19. [ABSTRACT FROM AUTHOR]

Deep learning-based segmentation models often struggle to achieve optimal performance when encountering new, unseen semantic classes. Their effectiveness hinges on vast amounts of annotated data and high computational resources for training. However, a promising solution to mitigate these challenges is the adoption of few-shot segmentation (FSS) networks, which can train models with reduced annotated data. The inherent complexity of medical images limits the applicability of FSS in medical imaging, despite its potential. Recent advancements in self-supervised label-efficient FSS models have demonstrated remarkable efficacy in medical image segmentation tasks. This paper presents a novel FSS architecture that enhances segmentation accuracy by utilising fewer features than existing methodologies. Additionally, this paper proposes a novel self-supervised learning approach that utilises supervoxel and augmented superpixel images to further enhance segmentation accuracy. This paper assesses the efficacy of the proposed model on two different datasets: abdominal magnetic resonance imaging (MRI) and cardiac MRI. The proposed model achieves a mean dice score and mean intersection over union of 81.62% and 70.38% for abdominal images, and 79.38% and 65.23% for cardiac images. [ABSTRACT FROM AUTHOR]

Sarcopenia is a prevalent condition with significant clinical implications, and it is expected to escalate globally, demanding for effective diagnostic strategies, possibly at an early stage of the disease. Imaging techniques play a pivotal role in comprehensively evaluating sarcopenia, offering insights into both muscle quantity and quality. Among all the imaging techniques currently used for the diagnosis and follow up of sarcopenia, it is possible to distinguish two classes: Rx based techniques, using ionizing radiations, and non-invasive techniques, which are based on the use of safe and low risk diagnostic procedures. Dual-energy x-ray Absorptiometry and Computed Tomography, while widely utilized, entail radiation exposure concerns. Ultrasound imaging offers portability, real-time imaging, and absence of ionizing radiation, making it a promising tool Magnetic Resonance Imaging, particularly T1-weighted and Dixon sequences, provides cross-sectional and high-resolution images and fat-water separation capabilities, facilitating precise sarcopenia quantification. Bioelectrical Impedance Analysis (BIA), a non-invasive technique, estimates body composition, including muscle mass, albeit influenced by hydration status. Standardized protocols, such as those proposed by the Sarcopenia through Ultrasound (SARCUS) Working Group, are imperative for ensuring consistency across assessments. Future research should focus on refining these techniques and harnessing the potential of radiomics and artificial intelligence to enhance diagnostic accuracy and prognostic capabilities in sarcopenia. [ABSTRACT FROM AUTHOR]

Due to the potentially high cost and difficulty of obtaining labeled data in medical image classification, semi-supervised learning can effectively utilize limited labeled data and large amounts of unlabeled data to enhance the performance of classification models. However, most existing methods utilize pseudo-labeling techniques that filter out a large amount of data, which makes the unlabeled data underutilized. Moreover, the consistency regularization methods rely heavily on the perturbation function suffering from a lack of generalization ability. In this paper, we propose a semi-supervised medical image classification approach based on class prototype matching for soft pseudo-labels with consistency regularization. A class prototype matching module is used to predict soft pseudo-labels for unlabeled data, and the quality of the predicted pseudo-labels is improved by a meticulously designed cache queue with dynamic and unbiased updates. The linear mixture strategy is employed to handle both labeled and unlabeled data supplying the model with more complex data inputs, which boosts the ability of the model to learn intra- and inter-class features. Extensive experiments on ISIC2018 skin lesion dataset and the Chexpert chest disease dataset demonstrate that our approach has satisfactory performance and generalization ability on the medical image classification than other state-of-the-art semi-supervised methods. [ABSTRACT FROM AUTHOR]

Purpose: We present a probabilistic approach to medical image analysis that requires, and makes use of, explicit prior information provided by a medical expert. Depending on the choice of prior model the method can be used for image enhancement, analysis, and segmentation. Methods: The methodology is based on a probabilistic approach to medical image analysis, that allows integration of 1) arbitrarily complex prior information (for which realizations can be generated), 2) information about a convolution operator of the imaging system, and 3) information about the noise in the reconstructed image into a posterior probability density. The method was demonstrated on positron emission tomography (PET) images obtained from a phantom and a patient with lung cancer. The likelihood model (multivariate log-normal) and the convolution operator were derived from phantom data. Two examples of prior information were used to show the potential of the method. The extended Metropolis-Hastings algorithm, a Markov chain Monte Carlo method, was used to generate realizations of the posterior distribution of the tracer activity concentration. Results: A set of realizations from the posterior was used as the base of a quantitative PET image analysis. The mean and variance of activity concentrations were computed, as well as the probability of high tracer uptake and statistics on the size and activity concentration of high uptake regions. For both phantom and in vivo images, the estimated images of mean activity concentrations appeared to have reduced noise levels, and a sharper outline of high activity regions, as compared to the original PET. The estimated variance of activity concentrations was high at the edges of high activity regions. Conclusions: The methodology provides a probabilistic approach for medical image analysis that explicitly takes into account medical expert knowledge as prior information. The presented first results indicate the potential of the method to improve the detection of small lesions. The methodology allows for a probabilistic measure of the size and activity level of high uptake regions, with possible long-term perspectives for early detection of cancer, aswell as treatment, planning, and follow-up. [ABSTRACT FROM AUTHOR]

The exchange of medical images and patient data over the internet has attracted considerable attention in the past decade, driven by advancements in communication and health services. However, transferring confidential data through insecure channels, such as the internet, exposes it to potential manipulations and attacks. To ensure the authenticity of medical images while concealing patient data within them, this paper introduces a high-capacity and reversible fragile watermarking model in which an authentication watermark is initially generated from the cover image and merged with the patient's information, photo, and medical report to form the global watermark. This watermark is subsequently encrypted using the chaotic Chen system technique, enhancing the model's security and ensuring patient data confidentiality. The cover image then undergoes a Discrete Fourier Transform (DFT) and the encrypted watermark is inserted into the frequency coefficients using a new embedding technique. The experimental results demonstrate that the proposed method achieves great watermarked image quality, with a PSNR exceeding 113 dB and an SSIM close to 1, while maintaining a high embedding capacity of 3 BPP (Bits Per Pixel) and offering perfect reversibility. Furthermore, the proposed model demonstrates high sensitivity to attacks, successfully detecting tampering in all 18 tested attacks, and achieves nearly perfect watermark extraction accuracy, with a Bit Error Rate (BER) of 0.0004%. This high watermark extraction accuracy is crucial in our situation where patient data need to be retrieved from the watermarked images with almost no alteration. [ABSTRACT FROM AUTHOR]

Current machine learning methods for medical image analysis primarily focus on developing models tailored for their specific tasks, utilizing data within their target domain. These specialized models tend to be data-hungry and often exhibit limitations in generalizing to out-of-distribution samples. In this work, we show that employing models that incorporate multiple domains instead of specialized ones significantly alleviates the limitations observed in specialized models. We refer to this approach as multi-domain model and compare its performance to that of specialized models. For this, we introduce the incorporation of diverse medical image domains, including different imaging modalities like X-ray, MRI, CT, and ultrasound images, as well as various viewpoints such as axial, coronal, and sagittal views. Our findings underscore the superior generalization capabilities of multi-domain models, particularly in scenarios characterized by limited data availability and out-of-distribution, frequently encountered in healthcare applications. The integration of diverse data allows multi-domain models to utilize information across domains, enhancing the overall outcomes substantially. To illustrate, for organ recognition, multi-domain model can enhance accuracy by up to 8% compared to conventional specialized models. [ABSTRACT FROM AUTHOR]

Background: Recent advancements in generative artificial intelligence (AI) for image generation have presented significant opportunities for medical imaging, offering a promising way to generate realistic virtual medical images while ensuring patient privacy. The generation of a large number of virtual medical images through AI has the potential to augment training datasets for discriminative AI models, particularly in fields with limited data availability, such as neuroimaging. Current studies on generative AI in neuroimaging have mainly focused on disease discrimination; however, its potential for simulating complex phenomena in psychiatric disorders remains unknown. In this study, as examples of a simulation, we aimed to present a novel generative AI model that transforms magnetic resonance imaging (MRI) images of healthy individuals into images that resemble those of patients with schizophrenia (SZ) and explore its application. Methods: We used anonymized public datasets from the Center for Biomedical Research Excellence (SZ, 71 patients; healthy subjects [HSs], 71 patients) and the Autism Brain Imaging Data Exchange (autism spectrum disorder [ASD], 79 subjects; HSs, 105 subjects). We developed a model to transform MRI images of HSs into MRI images of SZ using cycle generative adversarial networks. The efficacy of the transformation was evaluated using voxel-based morphometry to assess the differences in brain region volumes and the accuracy of age prediction pre- and post-transformation. In addition, the model was examined for its applicability in simulating disease comorbidities and disease progression. Results: The model successfully transformed HS images into SZ images and identified brain volume changes consistent with existing case-control studies. We also applied this model to ASD MRI images, where simulations comparing SZ with and without ASD backgrounds highlighted the differences in brain structures due to comorbidities. Furthermore, simulating disease progression while preserving individual characteristics showcased the model's ability to reflect realistic disease trajectories. Discussion: The results suggest that our generative AI model can capture subtle changes in brain structures associated with SZ, providing a novel tool for visualizing brain changes in different diseases. The potential of this model extends beyond clinical diagnosis to advances in the simulation of disease mechanisms, which may ultimately contribute to the refinement of therapeutic strategies. [ABSTRACT FROM AUTHOR]

Intratumoral heterogeneity poses a significant challenge to the diagnosis and treatment of recurrent glioblastoma. This study addresses the need for non-invasive approaches to map heterogeneous landscape of histopathological alterations throughout the entire lesion for each patient. We developed BioNet, a biologically-informed neural network, to predict regional distributions of two primary tissue-specific gene modules: proliferating tumor (Pro) and reactive/inflammatory cells (Inf). BioNet significantly outperforms existing methods (p < 2e-26). In cross-validation, BioNet achieved AUCs of 0.80 (Pro) and 0.81 (Inf), with accuracies of 80% and 75%, respectively. In blind tests, BioNet achieved AUCs of 0.80 (Pro) and 0.76 (Inf), with accuracies of 81% and 74%. Competing methods had AUCs lower or around 0.6 and accuracies lower or around 70%. BioNet's voxel-level prediction maps reveal intratumoral heterogeneity, potentially improving biopsy targeting and treatment evaluation. This non-invasive approach facilitates regular monitoring and timely therapeutic adjustments, highlighting the role of ML in precision medicine. [ABSTRACT FROM AUTHOR]

Simple Summary: This study aimed to enhance the accuracy of detecting laryngeal carcinoma using a modified AI model based on U-Net. The model was designed to automatically identify lesions in endoscopic images. Researchers addressed issues such as mode collapse and gradient explosion to ensure stable performance, achieving 99% accuracy in detecting malignancies. The study found that malignant tumors were detected more reliably than benign ones. This technology could help reduce human error in diagnoses, allowing for earlier detection and treatment. Furthermore, it has the potential to be applied in other medical fields, benefiting overall healthcare. This study modifies the U-Net architecture for pixel-based segmentation to automatically classify lesions in laryngeal endoscopic images. The advanced U-Net incorporates five-level encoders and decoders, with an autoencoder layer to derive latent vectors representing the image characteristics. To enhance performance, a WGAN was implemented to address common issues such as mode collapse and gradient explosion found in traditional GANs. The dataset consisted of 8171 images labeled with polygons in seven colors. Evaluation metrics, including the F1 score and intersection over union, revealed that benign tumors were detected with lower accuracy compared to other lesions, while cancers achieved notably high accuracy. The model demonstrated an overall accuracy rate of 99%. This enhanced U-Net model shows strong potential in improving cancer detection, reducing diagnostic errors, and enhancing early diagnosis in medical applications. [ABSTRACT FROM AUTHOR]

Deep learning has significantly advanced the field of medical image segmentation. However, the complexity of network structures often leads to high computational demands, limiting their practical efficiency. To enhance the efficiency of image segmentation, this paper introduces an innovative, concise, and lightweight deep learning network. First, to reduce model complexity, we replaced the attention mechanism in the traditional vision transformer (ViT) structure with a shift operation, creating the ShiftViT architecture. This substitution significantly decreased computation and the number of parameters while preserving model performance. Second, to retain and enhance fine-grained features and facilitate more precise information transfer across different layers, we employed a full-scale progressive skip connection strategy. This approach effectively integrates multi-scale feature information, further enhancing model performance. Additionally, to further reduce network complexity, inspired by the independence of probabilities, we opted for depth-wise separable convolution over traditional convolution. This enhances the relative independence between layers. Together, these modifications achieved superior segmentation results on both the Synapse and Automated Cardiac Diagnostic Challenge (ACDC) datasets compared to mainstream models, with substantial advantages in terms of computational efficiency and parameter count. The proposed approach represents an effective solution for medical image applications with limited computational resources and holds great promise for clinical practice. [ABSTRACT FROM AUTHOR]

Early survival prediction is vital for the clinical management of cancer patients, as tumors can be better controlled with personalized treatment planning. Traditional survival prediction methods are based on radiomics feature engineering and/or clinical indicators (e.g., cancer staging). Recently, survival prediction models with advances in deep learning techniques have achieved state-of-the-art performance in end-to-end survival prediction by exploiting deep features derived from medical images. However, existing models are heavily reliant on the prognostic information within primary tumors and cannot effectively leverage out-of-tumor prognostic information characterizing local tumor metastasis and adjacent tissue invasion. Also, existing models are sub-optimal in leveraging multi-modality medical images as they rely on empirically designed fusion strategies to integrate multi-modality information, where the fusion strategies are pre-defined based on domain-specific human prior knowledge and inherently limited in adaptability. Here, we present an Adaptive Multi-modality Segmentation-to-Survival model (AdaMSS) for survival prediction from multi-modality medical images. The AdaMSS can self-adapt its fusion strategy based on training data and also can adapt its focus regions to capture the prognostic information outside the primary tumors. Extensive experiments with two large cancer datasets (1380 patients from nine medical centers) show that our AdaMSS surmounts the state-of-the-art survival prediction performance (C-index: 0.804 and 0.757), demonstrating the potential to facilitate personalized treatment planning. [ABSTRACT FROM AUTHOR]

Background: Three‐dimensional gradient‐echo (3D‐GRE) sequences provide isotropic or nearly isotropic 3D images, leading to better visualization of smaller structures, compared to two‐dimensional (2D) sequences. The aim of this study was to prospectively compare 2D and 3D‐GRE sequences in terms of key imaging metrics, including signal‐to‐noise ratio (SNR), contrast‐to‐noise ratio (CNR), glenohumeral joint space, image quality, artifacts, and acquisition time in shoulder joint images, using 1.5‐T MRI scanner. Methods: Thirty‐five normal volunteers with no history of shoulder disorders prospectively underwent a shoulder MRI examination with conventional 2D sequences, including T1‐ and T2‐weighted fast spin echo (T1/T2w FSE) as well as proton density‐weighted FSE with fat saturation (PD‐FS) followed by 3D‐GRE sequences including VIBE, TRUEFISP, DESS, and MEDIC techniques. Two independent reviewers assessed all images of the shoulder joints. Pearson correlation coefficient and intra‐RR were used for reliability test. Results: Among 3D‐GRE sequences, TRUEFISP showed significantly the best CNR between cartilage‐bone (31.37 ± 2.57, p < 0.05) and cartilage‐muscle (13.51 ± 1.14, p < 0.05). TRUEFISP also showed the highest SNR for cartilage (41.65 ± 2.19, p < 0.01) and muscle (26.71 ± 0.79, p < 0.05). Furthermore, 3D‐GRE sequences showed significantly higher image quality, compared to 2D sequences (p < 0.001). Moreover, the acquisition time of the 3D‐GRE sequences was considerably shorter than the total acquisition time of PD‐FS sequences in three orientations (p < 0.01). Conclusions: 3D‐GRE sequences provide superior image quality and efficiency for evaluating articular joints, particularly in shoulder imaging. The TRUEFISP technique offers the best contrast and signal quality, making it a valuable tool in clinical practice. [ABSTRACT FROM AUTHOR]

ABSTRACT In the context of Cloud and Fog computing settings, recent developments in deep learning techniques show great potential for changing several fields, including healthcare. In this study, we make a contribution to this changing field by proposing an enhanced deep learning‐based strategy for classifying chest X‐ray images, using pre‐trained models such as RetinaNet, EfficientNet and Faster‐R‐CNN, which we use through transfer learning. Our strategy outperforms single models and traditional techniques by leveraging critical data gleaned from multiple models, demonstrating the ability of deep learning to improve diagnostic precision. Our approach presents a novel dual‐check system in the context of worries about security, privacy and trust in Cloud and Fog‐based Smart Systems. In this case, a decision support system uses chest X‐ray images to make an initial diagnosis that is then confirmed by a medical practitioner. This cooperative strategy not only reduces diagnostic errors that come from machine and human sources but also emphasises how crucial it is to incorporate AI‐driven solutions into safe and reliable healthcare ecosystems. Our approach raises the bar for the quality of patient care and healthcare outcomes by overcoming the drawbacks of traditional diagnostic methods that depend on the subjective opinions of physicians. Our work brings out how deep learning might transform clinical diagnostics by distinguishing inflammatory regions in chest X‐ray images. Research is needed to fully grasp the transformative potential of deep learning in medical image processing, especially as the healthcare industry continues to embrace AI‐driven solutions. Further research endeavours have to dig into tactics like broadening the scope of datasets, executing data augmentation methodologies and incorporating bespoken features to augment the elasticity and effectiveness of AI‐driven diagnostic systems. [ABSTRACT FROM AUTHOR]

Accurately identifying and analyzing nuclei is pivotal for both the diagnosis and examination of cancer. However, the complexity of this task arises due to the presence of overlapping and cluttered nuclei with blurred boundaries, variations in nuclei sizes and shapes, and an imbalance in the available datasets. Although current methods utilize region proposal techniques and feature encoding frameworks, but they often fail to precisely identify occluded nuclei instances. We propose a model named AFINITI, which is both simple, efficient, achieves high accuracy, recognizes instance boundaries cluttered and overlapping nuclei, and addresses class imbalance issues. Our approach utilizes nuclei pixel positional information and a novel loss function to yield accurate class information for each nuclei. Our network features a lightweight, attention-aware feature fusion architecture with separate instance probability, shape radial estimator, and classification heads. We use a compound classification loss function to assign a weighted loss to each class according to its occurrence frequency, thereby addressing the class imbalance issues. The AFINITI model outperforms current leading networks across eight major publicly available nuclei segmentation datasets achieving up to an 8% increase in Dice Similarity Coefficient (DSc) and a 17% increase in Panoptic Quality (PQ) compared to existing techniques demonstrating its effectiveness and potential for clinical applications. The source code and the weights of the trained model have been released to the public and can be accessed at: https://github.com/Vision-At-SEECS/AF-Net. [ABSTRACT FROM AUTHOR]

Background: The cost of labeling to collect training data sets using deep learning is especially high in medical applications compared to other fields. Furthermore, due to variances in images depending on the computed tomography (CT) devices, a deep learning based segmentation model trained with a certain device often does not work with images from a different device. Methods: In this study, we propose an efficient learning strategy for deep learning models in medical image segmentation. We aim to overcome the difficulties of segmentation in CT images by training a VNet segmentation model which enables rapid labeling of organs in CT images with the model obtained by transfer learning using a small number of manually labeled images, called SEED images. We established a process for generating SEED images and conducting transfer learning a model. We evaluate the performance of various segmentation models such as vanilla UNet, UNETR, Swin-UNETR and VNet. Furthermore, assuming a scenario that a model is repeatedly trained with CT images collected from multiple devices, in which is catastrophic forgetting often occurs, we examine if the performance of our model degrades. Results: We show that transfer learning can train a model that does a good job of segmenting muscles with a small number of images. In addition, it was confirmed that VNet shows better performance when comparing the performance of existing semi-automated segmentation tools and other deep learning networks to muscle and liver segmentation tasks. Additionally, we confirmed that VNet is the most robust model to deal with catastrophic forgetting problems. Conclusion: In the 2D CT image segmentation task, we confirmed that the CNN-based network shows better performance than the existing semi-automatic segmentation tool or latest transformer-based networks. [ABSTRACT FROM AUTHOR]

Background: Maxillofacial complex automated segmentation could alternative traditional segmentation methods to increase the effectiveness of virtual workloads. The use of DL systems in the detection of maxillary sinus and pathologies will both facilitate the work of physicians and be a support mechanism before the planned surgeries. Objective: The aim was to use a modified You Only Look Oncev5x (YOLOv5x) architecture with transfer learning capabilities to segment both maxillary sinuses and maxillary sinus diseases on Cone-Beam Computed Tomographic (CBCT) images. Methods: Data set consists of 307 anonymised CBCT images of patients (173 women and 134 males) obtained from the radiology archive of the Department of Oral and Maxillofacial Radiology. Bilateral maxillary sinuses CBCT scans were used to identify mucous retention cysts (MRC), mucosal thickenings (MT), total and partial opacifications, and healthy maxillary sinuses without any radiological features. Results: Recall, precision and F1 score values for total maxillary sinus segmentation were 1, 0.985 and 0.992, respectively; 1, 0.931 and 0.964 for healthy maxillary sinus segmentation; 0.858, 0.923 and 0.889 for MT segmentation; 0.977, 0.877 and 0.924 for MRC segmentation; 1, 0.942 and 0.970 for sinusitis segmentation. Conclusion: This study demonstrates that maxillary sinuses can be segmented, and maxillary sinus diseases can be accurately detected using the AI model. [ABSTRACT FROM AUTHOR]

Although the rapid growth in the efficiency of medical imaging is undeniable, the expansion of health information technology (HIT) into medical imaging has not been as seamless or well-integrated as it was thought to be. The socio-technical complexities in medical imaging associated with HIT systems can cause risks to patient harm and inconvenience, both individually and collectively, often in new, unforeseen, and unexpected ways. This study reflects a retrospectively collected single incident report related to medical imaging HIT systems, aiming to develop a set of preventive and corrective strategies. A combination of multiple deductive approaches (existing frameworks), i.e., HIT Classification Systems and 18-step medical imaging process workflow and inductive method (content analysis), were used to analyze the incident. The incident was identified as a "system configuration"-related software issue, contributed by system upgrade-changes in hardware and software. The incident was determined to occur during steps 10-12, i.e., "study selection and retrieval," "calling up of patient's referral," and "image review and interpretation," causing severe disruptions in the clinical workflow for several weeks. We propose 16 preventive and corrective strategies grouped under four key areas based on the sociotechnical aspects associated with HIT systems. The key areas are (i) preparation and integration for upgraded systems, (ii) training for medical imaging specialists, (iii) contingency planning/immediate backup system, and (iv) system design and configuration. These strategies are expected to help healthcare staff, analysts, reporters, researchers, and relevant stakeholders improve care delivery and patient safety in medical imaging in the context of any system upgrades. [ABSTRACT FROM AUTHOR]

In the domain of medical imaging, the advent of deep learning has marked a significant progression, particularly in the nuanced area of periodontal disease diagnosis. This study specifically targets the prevalent issue of scarce labeled data in medical imaging. We introduce a novel unsupervised few-shot learning algorithm, meticulously crafted for classifying periodontal diseases using a limited collection of dental panoramic radiographs. Our method leverages UNet architecture for generating regions of interest (RoI) from radiographs, which are then processed through a Convolutional Variational Autoencoder (CVAE). This approach is pivotal in extracting critical latent features, subsequently clustered using an advanced algorithm. This clustering is key in our methodology, enabling the assignment of labels to images indicative of periodontal diseases, thus circumventing the challenges posed by limited datasets. Our validation process, involving a comparative analysis with traditional supervised learning and standard autoencoder-based clustering, demonstrates a marked improvement in both diagnostic accuracy and efficiency. For three real-world validation datasets, our UNet-CVAE architecture achieved up to average 14% higher accuracy compared to state-of-the-art supervised models including the vision transformer model when trained with 100 labeled images. This study not only highlights the capability of unsupervised learning in overcoming data limitations but also sets a new benchmark for diagnostic methodologies in medical AI, potentially transforming practices in data-constrained scenarios. [ABSTRACT FROM AUTHOR]

Introduction: Brain tumors are characterized by abnormal cell growth within or around the brain, posing severe health risks often associated with high mortality rates. Various imaging techniques, including magnetic resonance imaging (MRI), are commonly employed to visualize the brain and identify malignant growths. Computer-aided diagnosis tools (CAD) utilizing Convolutional Neural Networks (CNNs) have proven effective in feature extraction and predictive analysis across diverse medical imaging modalities. Methods: This study explores a CNN trained and evaluated with nine activation functions, encompassing eight established ones from the literature and a modified version of the soft sign activation function. Results: The latter demonstrates notable efficacy in discriminating between four types of brain tumors in MR images, achieving an accuracy of 97.6%. The sensitivity for glioma is 93.7%; for meningioma, it is 97.4%; for cases with no tumor, it is 98.8%; and for pituitary tumors, it reaches 100%. Discussion: In this manuscript, we propose an advanced CNN architecture that integrates a newly developed activation function. Our extensive experimentation and analysis showcase the model's remarkable ability to precisely distinguish between different types of brain tumors within a substantial and diverse dataset. The findings from our study suggest that this model could serve as an invaluable supplementary tool for healthcare practitioners, including specialized medical professionals and resident physicians, in the accurate diagnosis of brain tumors. [ABSTRACT FROM AUTHOR]

As one of the hottest topics in medical imaging, artificial intelligence (AI) plays a major role in large-scale medical image processing. In many studies, it has been applied for automatic recognition of complex patterns in image data and different stages of treatment. As the world's population is aging, countries around the world face a rapid increase in the demand for elderly care. The use of AI in medical image recognition supports clinical staff in their image analysis and facilitates faster diagnosis when there is a shortage of caregivers. It is particularly useful for the treatment of chronic wounds. Given the complexity and variety of issues associated with acute and chronic wounds, they are typically documented in diagnostic and nursing records with text descriptions and photographs. However, wounds may remain inflamed for a long time owing to the nursing staff's poor understanding of wound care procedures, inadequate wound assessment skills, and insufficient knowledge of the use of dressings. Not only does this cause physical and psychological suffering to patients, but also results in high medical expenses. To this end, this study proposes an AIoT-based prognostic assessment system that supports the treatment of pressure injuries in residential facilities. The system uses an intelligent mobile IoT device to quickly measure wound size and changes in tissue ratios with a deep recognition module, which allows caregivers to quickly classify wound stages and estimate the healing condition based on the system's results. Clinically, it can help physicians and nursing staff to automatically generate a pressure injury report. This study conforms to the government's guidelines on wound assessment, which cover causes of wounds, treatment goals, and care procedures. The system is intended to support physicians in clinical decision-making and nursing staff in efficient reporting. In addition, the system can be used for tracking patients' health to reduce the risk of wound deterioration and to provide more targeted medical care for the elderly community. [ABSTRACT FROM AUTHOR]