Medical Imaging Research Articles

Multistage transfer learning for medical images

Deep learning is revolutionizing various domains and significantly impacting medical image analysis. Despite notable progress, numerous challenges remain, necessitating the refinement of deep learning algorithms for optimal performance in medical image analysis. This paper explores the growing demand for precise and robust medical image analysis by focusing on an advanced deep learning technique, multistage transfer learning. Over the past decade, multistage transfer learning has emerged as a pivotal strategy, particularly in overcoming challenges associated with limited medical data and model generalization. However, the absence of well-compiled literature capturing this development remains a notable gap in the field. This exhaustive investigation endeavors to address this gap by providing a foundational understanding of how multistage transfer learning approaches confront the unique challenges posed by insufficient medical image datasets. The paper offers a detailed analysis of various multistage transfer learning types, architectures, methodologies, and strategies deployed in medical image analysis. Additionally, it delves into intrinsic challenges within this framework, providing a comprehensive overview of the current state while outlining potential directions for advancing methodologies in future research. This paper underscores the transformative potential of multistage transfer learning in medical image analysis, providing valuable guidance to researchers and healthcare professionals.

Toward Smart Ultrasound Image Augmentation to Advance Tumor Treatment Monitoring: Exploring the Potential of Diffusion Generative Model

Abstract Medical imaging is a crucial tool in clinics to monitor tumor treatment progress. In practice, many imaging tools (such as magnetic resonance imaging (MRI) and computed tomography (CT) scans) are in general costly and may also expose patients to radiation, leading to potential side effects. Recent studies have demonstrated that ultrasound imaging, which is safe, low-cost, and easy to access, can monitor the drug delivery progress in solid tumors. However, the noisy nature of ultrasound images and the high-level uncertainty of cancer disease progression are still challenging in ultrasound-based tumor treatment monitoring. To overcome these barriers, this work presents a comparative study to explore the potential advantages of the emerging diffusion generative models against the commonly applied state-of-the-art generative models. Namely, the denoising diffusion models (DDMs), against the generative adversarial networks (GAN), and variational auto-encoders (VAE), are used for analyzing the ultrasound images through image augmentation. These models are evaluated based on their capacity to augment ultrasound images for exploring the potential variations of tumor treatment monitoring. The results across different cases indicate that the denoising diffusion implicit models (DDIM)/kernel inception distance (KID)-inception score (IS) model leveraged in this work outperforms the other models in the study in terms of similarity, diversity, and predictive accuracy. Therefore, further investigation of such diffusion generative models could be considered as they can potentially serve as a great predictive tool for ultrasound image-enabled tumor treatment monitoring in the future.

Attention-driven visual emphasis for medical volumetric image visualization

AbstractDirect volume rendering (DVR) is a commonly utilized technique for three-dimensional visualization of volumetric medical images. A key goal of DVR is to enable users to visually emphasize regions of interest (ROIs) which may be occluded by other structures. Conventional methods for ROIs visual emphasis require extensive user involvement for the adjustment of rendering parameters to reduce the occlusion, dependent on the user’s viewing direction. Several works have been proposed to automatically preserve the view of the ROIs by eliminating the occluding structures of lower importance in a view-dependent manner. However, they require pre-segmentation labeling and manual importance assignment on the images. An alternative to ROIs segmentation is to use ‘saliency’ to identify important regions. This however lacks semantic information and thus leads to the inclusion of false positive regions. In this study, we propose an attention-driven visual emphasis method for volumetric medical image visualization. We developed a deep learning attention model, termed as focused-class attention map (F-CAM), trained with only image-wise labels for automated ROIs localization and importance estimation. Our F-CAM transfers the semantic information from the classification task for use in the localization of ROIs, with a focus on small ROIs that characterize medical images. Additionally, we propose an attention compositing module that integrates the generated attention map with transfer function within the DVR pipeline to automate the view-dependent visual emphasis of the ROIs. We demonstrate the superiority of our method compared to existing methods on a multi-modality PET-CT dataset and an MRI dataset.

Rhodamine 6G in Oxidized Sodium Alginate Polymeric Hydrogel for Photodynamically Inactivating Cancer Cells.

As cancer therapy progresses, challenges remain due to the inherent drawbacks of conventional treatments such as chemotherapy, gene therapy, radiation therapy, and surgical removal. Moreover, due to their associated side effects, conventional treatments affect both cancerous and normal cells, making photodynamic therapy (PDT) an attractive alternative. As a result of its minimal toxicity, exceptional specificity, and non-invasive characteristics, PDT represents an innovative and highly promising cancer treatment strategy using photosensitizers (PSs) and precise wavelength excitation light to introduce reactive oxygen species (ROS) in the vicinity of cancer cells. Poor aqueous solubility and decreased sensitivity of Rhodamine 6G (R6G) prevent its use as a photosensitizer in PDT, necessitating the development of oxidized sodium alginate (OSA) hydrogelated nanocarriers to enhance its bioavailability, targeted distribution, and ROS-quantum yield. The ROS quantum yield increased from 0.30 in an aqueous environment to 0.51 when using alginate-based formulations, and it was further enhanced to 0.81 in the case of OSA. Furthermore, the nanoformulations produced fluorescent signals suitable for use as cellular imaging agents, demonstrating contrast-enhancing capabilities in medical imaging and showing minimal toxicity.

Machine-Learning-Enabled Diagnostics with Improved Visualization of Disease Lesions in Chest X-ray Images.

The class activation map (CAM) represents the neural-network-derived region of interest, which can help clarify the mechanism of the convolutional neural network's determination of any class of interest. In medical imaging, it can help medical practitioners diagnose diseases like COVID-19 or pneumonia by highlighting the suspicious regions in Computational Tomography (CT) or chest X-ray (CXR) film. Many contemporary deep learning techniques only focus on COVID-19 classification tasks using CXRs, while few attempt to make it explainable with a saliency map. To fill this research gap, we first propose a VGG-16-architecture-based deep learning approach in combination with image enhancement, segmentation-based region of interest (ROI) cropping, and data augmentation steps to enhance classification accuracy. Later, a multi-layer Gradient CAM (ML-Grad-CAM) algorithm is integrated to generate a class-specific saliency map for improved visualization in CXR images. We also define and calculate a Severity Assessment Index (SAI) from the saliency map to quantitatively measure infection severity. The trained model achieved an accuracy score of 96.44% for the three-class CXR classification task, i.e., COVID-19, pneumonia, and normal (healthy patients), outperforming many existing techniques in the literature. The saliency maps generated from the proposed ML-GRAD-CAM algorithm are compared with the original Gran-CAM algorithm.

Enhancing cerebral arteriovenous malformation analysis: Development and application of patient-specific lumped parameter models based on 3D imaging data

Objectives:Cerebral arteriovenous malformations (AVMs) present complex neurovascular challenges, characterized by direct arteriovenous connections that disrupt normal brain blood flow dynamics. Traditional lumped parameter models (LPMs) offer a simplified angioarchitectural representation of AVMs, yet often fail to capture the intricate structure within the AVM nidus. This research aims at refining our understanding of AVM hemodynamics through the development of patient-specific LPMs utilizing three-dimensional (3D) medical imaging data for enhanced structural fidelity. Methods:This study commenced with the meticulous delineation of AVM vascular architecture using threshold segmentation and skeletonization techniques. The AVM nidus’s core structure was outlined, facilitating the extraction of vessel connections and the formation of a detailed fistulous vascular tree model. Sampling points, spatially distributed and derived from the pixel intensity in imaging data, guided the construction of a complex plexiform tree within the nidus by generating smaller Y-shaped vascular formations. This model was then integrated with an electrical analog model to enable precise numerical simulations of cerebral hemodynamics with AVMs. Results:The study successfully generated two distinct patient-specific AVM networks, mirroring the unique structural and morphological characteristics of the AVMs as captured in medical imaging. The models effectively represented the intricate fistulous and plexiform vessel structures within the nidus. Numerical analysis of these models revealed that AVMs induce a blood shunt effect, thereby diminishing blood perfusion to adjacent brain tissues. Conclusion:This investigation enhances the theoretical framework for AVM research by constructing patient-specific LPMs that accurately reflect the true vascular structures of AVMs. These models offer profound insights into the hemodynamic behaviors of AVMs, including their impact on cerebral circulation and the blood steal phenomenon. Further incorporation of clinical data into these models holds the promise of deepening the theoretical comprehension of AVMs and fostering advancements in the diagnosis and treatment of AVMs.

A rare case report: Brucellosis as an extreme cause of shoulder subluxation and rotator cuff tendinopathy

Rationale: Shoulder subluxation and rotator cuff tendinopathy is commonly associated with degeneration or trauma, and brucellosis has been reported to be correlated with shoulder subacromial bursitis, arthritis. However, brucellosis as a cause of shoulder subluxation and rotator cuff tendinopathy has never been reported in the English medical literature. Patient concerns: A 46-year-old man complained of left shoulder weakness and limited movement ongoing for 1 month. He was primarily diagnosed with shoulder subluxation and rotator cuff tear based on magnetic resonance imaging, and was advised to undergo arthroscopic rotator cuff surgical repair. Diagnoses: Although rotator cuff tear was diagnosed in another hospital, some doubts remained. First, a tendon profile with a relative high signal in T2-weighted magnetic resonance images in the supraspinatus tear area still existed. The findings were clearly different from the usual full-thickness rotator cuff tear, and no tendon profile could be found in the tear area. Second, degenerative rotator cuff tear mainly occurs in the aged population and this patient was relatively young, without a history of trauma. His left shoulder was painless but the weakness and the limited movement were extreme and all of these factors did not support the diagnosis of rotator cuff tear. Some infectious disease was suspected to be the cause and relevant laboratory tests were conducted. This patient was eventually diagnosed as having brucellosis-associated shoulder subluxation and rotator cuff tendinopathy based on positive serum standard tube agglutination test and buffer plate agglutination test for brucellosis. Interventions: Oral drug therapy with rifampicin and doxycycline was used for 8 weeks. Outcomes: A complete response with normal shoulder strength and movement was achieved at 8 weeks. At the one-year follow-up no signs of recurrence were detected. Lessons: Brucellosis-associated shoulder subluxation and rotator cuff tendinopathy may be misdiagnosed as a rotator cuff tear in magnetic resonance imaging and be further managed with indeterminate arthroscopic repair. A comprehensive understanding of the patient’s previous medical history, imaging and laboratory data and a careful physical examination are of great clinic value for the accurate diagnosis of the condition and the subsequent prognosis of the patient.

CINNAMON-GUI: Revolutionizing Pap Smear Analysis with CNN-Based Digital Pathology Image Classification

Background Medical imaging has seen significant advancements through machine learning, particularly convolutional neural networks (CNNs). These technologies have transformed the analysis of pathological images, enhancing the accuracy of diagnosing and classifying cellular anomalies. Digital pathology methodologies, including image analysis, have improved cervical cancer diagnostics. However, existing commercial platforms are often costly and restrictive, limiting customization and scalability. Methods CINNAMON-GUI is an open-source digital pathology tool based on CNNs for classifying Pap smear images. Transitioning to a Shiny app in Python, it offers enhanced user interface and interactivity. The application supports dynamic web interactions, advanced features for image analysis, and state-of-the-art CNN models tailored for digital pathology. Key features include intuitive UI components, real-time image and plot generation, memory-efficient data handling, and robust training capabilities with customizable CNN architectures. The tool also integrates with Labelme for defining regions of interest and allows testing on external biospecimens. Results Model A (seed 42, 100 epochs) and Model B (same architecture with adjusted augmentation parameters) were compared. Model A stabilized with training accuracy around 0.88 and validation accuracy around 0.913. Model B showed improved performance with training accuracy around 0.91 and validation accuracy around 0.95. Feature mapping highlighted critical morphological aspects, improving classification accuracy. Model B reduced misclassification errors significantly compared to Model A. Conclusions CINNAMON-GUI demonstrates the potential of an open-source platform in digital pathology, providing transparency and collaborative opportunities. The tool enhances diagnostic accuracy through feature map analysis and optimized CNN training. Future development aims to extend its application to other cancer types, leveraging its dynamic and user-friendly interface for broader use in diagnostics.

G-Net: Implementing an enhanced brain tumor segmentation framework using semantic segmentation design.

A fundamental computer vision task called semantic segmentation has significant uses in the understanding of medical pictures, including the segmentation of tumors in the brain. The G-Shaped Net architecture appears in this context as an innovative and promising design that combines components from many models to attain improved accuracy and efficiency. In order to improve efficiency, the G-Shaped Net architecture synergistically incorporates four fundamental components: the Self-Attention, Squeeze Excitation, Fusion, and Spatial Pyramid Pooling block structures. These factors work together to improve the precision and effectiveness of brain tumor segmentation. Self-Attention, a crucial component of G-Shaped architecture, gives the model the ability to concentrate on the image's most informative areas, enabling accurate localization of tumor boundaries. By adjusting channel-wise feature maps, Squeeze Excitation completes this by improving the model's capacity to capture fine-grained information in the medical pictures. Since the G-Shaped model's Spatial Pyramid Pooling component provides multi-scale contextual information, the model is capable of handling tumors of various sizes and complexity levels. Additionally, the Fusion block architectures combine characteristics from many sources, enabling a thorough comprehension of the image and improving the segmentation outcomes. The G-Shaped Net architecture is an asset for medical imaging and diagnostics and represents a substantial development in semantic segmentation, which is needed more and more for accurate brain tumor segmentation.

Interactive Cascaded Network for Prostate Cancer Segmentation from Multimodality MRI with Automated Quality Assessment.

The accurate segmentation of prostate cancer (PCa) from multiparametric MRI is crucial in clinical practice for guiding biopsy and treatment planning. Existing automated methods often lack the necessary accuracy and robustness in localizing PCa, whereas interactive segmentation methods, although more accurate, require user intervention on each input image, thereby limiting the cost-effectiveness of the segmentation workflow. Our innovative framework addresses the limitations of current methods by combining a coarse segmentation network, a rejection network, and an interactive deep network known as Segment Anything Model (SAM). The coarse segmentation network automatically generates initial segmentation results, which are evaluated by the rejection network to estimate their quality. Low-quality results are flagged for user interaction, with the user providing a region of interest (ROI) enclosing the lesions, whereas for high-quality results, ROIs were cropped from the automatic segmentation. Both manually and automatically defined ROIs are fed into SAM to produce the final fine segmentation. This approach significantly reduces the annotation burden and achieves substantial improvements by flagging approximately 20% of the images with the lowest quality scores for manual annotation. With only half of the images manually annotated, the final segmentation accuracy is statistically indistinguishable from that achieved using full manual annotation. Although this paper focuses on prostate lesion segmentation from multimodality MRI, the framework can be adapted to other medical image segmentation applications to improve segmentation efficiency while maintaining high accuracy standards.

IEA-Net: Internal and External Dual-Attention Medical Segmentation Network with High-Performance Convolutional Blocks.

Currently, deep learning is developing rapidly in the field of image segmentation, and medical image segmentation is one of the key applications in this field. Conventional CNN has achieved great success in general medical image segmentation tasks, but it has feature loss in the feature extraction part and lacks the ability to explicitly model remote dependencies, which makes it difficult to adapt to the task of human organ segmentation. Although methods containing attention mechanisms have made good progress in the field of semantic segmentation, most of the current attention mechanisms are limited to a single sample, while the number of samples of human organ images is large, ignoring the correlation between the samples is not conducive to image segmentation. In order to solve these problems, an internal and external dual-attention segmentation network (IEA-Net) is proposed in this paper, and the ICSwR (interleaved convolutional system with residual) module and the IEAM module are designed in this network. The ICSwR contains interleaved convolution and hopping connection, which are used for the initial extraction of the features in the encoder part. The IEAM module (internal and external dual-attention module) consists of the LGGW-SA (local-global Gaussian-weighted self-attention) module and the EA module, which are in a tandem structure. The LGGW-SA module focuses on learning local-global feature correlations within individual samples for efficient feature extraction. Meanwhile, the EA module is designed to capture inter-sample connections, addressing multi-sample complexities. Additionally, skip connections will be incorporated into each IEAM module within both the encoder and decoder to reduce feature loss. We tested our method on the Synapse multi-organ segmentation dataset and the ACDC cardiac segmentation dataset, and the experimental results show that the proposed method achieves better performance than other state-of-the-art methods.

Efficient MRI image enhancement by improved denoising techniques for better skull stripping using attention module-based convolution neural network

ABSTRACT Anatomical structure preservation throughout the denoising process is a challenge in the domain of medical imaging. The Rician noise introduced through the acquisition procedure by the Magnetic Resonance Imaging (MRI) scanner distorts the images. In this study, denoising using Wavelet-based Non-Local Median Filter (WBNLMF) and a novel contrast-enhancement method termed Improved Minimum Intensity Error Intuitionistic Fuzzy Contrast Enhancement (IMIEIFCET) is suggested. This methodology gives superior results while maintaining the edges and the brightness of the original image. An Attention Module-based Convolution Neural Network (AM-CNN) is suggested in the research as a methodology for skull stripping from MRI data. With a mean Dice coefficient of 0.998, a Sensitivity of 0.9975, and a Specificity of 0.9985, the proposed network exhibits result that are comparable to those of the specified Deep Learning (DL)-based technique.

Medical Image Classification with a Hybrid SSM Model Based on CNN and Transformer

Medical image classification, a pivotal task for diagnostic accuracy, poses unique challenges due to the intricate and variable nature of medical images compared to their natural counterparts. While Convolutional Neural Networks (CNNs) and Transformers are prevalent in this domain, each architecture has its drawbacks. CNNs, despite their strength in local feature extraction, fall short in capturing global context, whereas Transformers excel at global information but can overlook fine-grained details. The integration of CNNs and Transformers in a hybrid model aims to bridge this gap by enabling simultaneous local and global feature extraction. However, this approach remains constrained in its capacity to model long-range dependencies, thereby hindering the efficient extraction of distant features. To address these issues, we introduce the MambaConvT model, which employs a state-space approach. It begins by locally processing input features through multi-core convolution, enhancing the extraction of deep, discriminative local details. Next, depth-separable convolution with a 2D selective scanning module (SS2D) is employed to maintain a global receptive field and establish long-distance connections, capturing the fine-grained features. The model then combines hybrid features for comprehensive feature extraction, followed by global feature modeling to emphasize on global detail information and optimize feature representation. This paper conducts thorough performance experiments on different algorithms across four publicly available datasets and two private datasets. The results demonstrate that MambaConvT outperforms the latest classification algorithms in terms of accuracy, precision, recall, F1 score, and AUC value ratings, achieving superior performance in the precise classification of medical images.

IQAGPT: computed tomography image quality assessment with vision-language and ChatGPT models

Large language models (LLMs), such as ChatGPT, have demonstrated impressive capabilities in various tasks and attracted increasing interest as a natural language interface across many domains. Recently, large vision-language models (VLMs) that learn rich vision–language correlation from image–text pairs, like BLIP-2 and GPT-4, have been intensively investigated. However, despite these developments, the application of LLMs and VLMs in image quality assessment (IQA), particularly in medical imaging, remains unexplored. This is valuable for objective performance evaluation and potential supplement or even replacement of radiologists’ opinions. To this end, this study introduces IQAGPT, an innovative computed tomography (CT) IQA system that integrates image-quality captioning VLM with ChatGPT to generate quality scores and textual reports. First, a CT-IQA dataset comprising 1,000 CT slices with diverse quality levels is professionally annotated and compiled for training and evaluation. To better leverage the capabilities of LLMs, the annotated quality scores are converted into semantically rich text descriptions using a prompt template. Second, the image-quality captioning VLM is fine-tuned on the CT-IQA dataset to generate quality descriptions. The captioning model fuses image and text features through cross-modal attention. Third, based on the quality descriptions, users verbally request ChatGPT to rate image-quality scores or produce radiological quality reports. Results demonstrate the feasibility of assessing image quality using LLMs. The proposed IQAGPT outperformed GPT-4 and CLIP-IQA, as well as multitask classification and regression models that solely rely on images.

Optimizing Glioblastoma, IDH-wildtype Treatment Outcomes : A Radiomics and Support Vector Machine -Based Approach to Overall Survival Estimation.

Glioblastoma multiforme (GBM), particularly the IDH-wildtype type, represents a significant clinical challenge due to its aggressive nature and poor prognosis. Despite advancements in medical imaging and its modalities, survival rates have not improved significantly, demanding innovative treatment planning and outcome prediction approaches. This study utilizes a Support Vector Machine (SVM) classifier using radiomics features to predict the overall survival (OS) of GBM, IDH-wildtype patients to short (< 12 Months) and long (>=12 Months) survivors. A dataset comprising multi-parametric MRI (mpMRI) scans from 574 patients was analyzed. Radiomic features were extracted from T1, T2, FLAIR, and T1-Gd sequences. Low variance features were removed, and Recursive Feature Elimination (RFE) was used to select the most informative features. The SVM model was trained using a k-fold cross-validation approach. Furthermore, clinical parameters such as age, gender, and MGMT promoter methylation status were integrated to enhance prediction accuracy. The model showed reasonable results in terms of cross-validated AUC of 0.84 (95% CI: 0.80-0.90) with (p-value < 0.001) effectively categorizing patients into short and long survivors. Log-rank test (Chi-square statistics) analysis for the developed model was 0.00029 along with the 1.20 Cohen's d effect size. Most importantly, clinical data integration further refined the survival estimates, providing a more fitted prediction that considers individual patient characteristics by Kaplan-Meier curve with p-value<0.0001. The proposed method significantly enhances the predictive accuracy of OS outcomes in GBM, IDH-wildtype patients. By integrating detailed imaging features with key clinical indicators, this model offers a robust tool for personalized treatment planning, potentially improving OS.

Synthetic Medical Imaging Generation with Generative Adversarial Networks for Plain Radiographs

In medical imaging, access to data is commonly limited due to patient privacy restrictions, and it can be difficult to acquire enough data in the case of rare diseases. The purpose of this investigation was to develop a reusable open-source synthetic image-generation pipeline, the GAN Image Synthesis Tool (GIST), that is easy to use as well as easy to deploy. The pipeline helps to improve and standardize AI algorithms in the digital health space by generating high quality synthetic image data that is not linked to specific patients. Its image generation capabilities include the ability to generate imaging of pathologies or injuries with low incidence rates. This improvement of digital health AI algorithms could improve diagnostic accuracy, aid in patient care, decrease medicolegal claims, and ultimately decrease the overall cost of healthcare. The pipeline builds on existing Generative Adversarial Networks (GANs) algorithms, and preprocessing and evaluation steps were included for completeness. For this work, we focused on ensuring the pipeline supports radiography, with a focus on synthetic knee and elbow X-ray images. In designing the pipeline, we evaluated the performance of current GAN architectures, studying the performance on available X-ray data. We show that the pipeline is capable of generating high-quality and clinically relevant images based on a lay person’s evaluation and the Fréchet Inception Distance (FID) metric.

Extrahepatic Portal Venous Gas Is the Strongest Predictor of Mortality in Patients with Portal Venous Gas and Pneumatosis Intestinalis.

Very few studies have examined the association between contrast-enhanced computed tomography (CT) findings observed in portal venous gas (PVG) and pneumatosis intestinalis (PI) and the underlying diseases in these conditions. In this study, we analyzed this association and report the findings for predicting mortality. Overall, 50 patients diagnosed with PVG or PI, observed on contrast-enhanced CT, underwent treatment at our hospital. Based on the underlying disease, we divided the patients into three groups, those with ischemic disease, infectious disease, or gastrointestinal dilatation. Furthermore, cases that underwent surgical treatment or needed surgery but were inoperable were assigned to the high risk group (n=16) and patients who received conservative treatment were assigned to the low risk group (n=34). We reviewed the patients' medical charts, laboratory data, and CT images retrospectively, and analyzed the relationship between CT findings, underlying disease, and association with the high risk or low risk group in each case. Poor enhancement of the intestinal wall, mesenteric fat stranding, extrahepatic PVG, advanced age, and renal disease were significantly associated with ischemic disease (p=0.02, p=0.02, p=0.005, p=0.008 and p=0.049, respectively). PI alone was strongly associated with gastrointestinal dilatation (p=0.009). Patients in the low risk group had more favorable outcomes with conservative treatment. In multivariate analysis, extrahepatic PVG was the only factor associated with the high risk group (p=0.002). Extrahepatic PVG associated with ischemic disease was the strongest predictive factor of mortality. Other CT findings, though useful in diagnosing the underlying disease, were not significant predictive factors.

A Robust Deep Learning Model for Brain Tumor Detection and Classification Using Efficient Net: A Brief Meta-Analysis

Accurately detecting and classifying brain tumors’, is critical for timely diagnosis and effective treatment planning. The purpose of this paper is to provide a comprehensive examination utilizing the EfficientNet family of deep learning architectures to automatically identify and categorize three forms of brain tumors from magnetic resonance imaging (MRI) scans. The primary aim of the study is to assess the performance of different EfficientNet models (ranging from EfficientNet-B0 to EfficientNet-B7) and determine their capability to achieve high accuracy in brain tumor classification. In the implementation, a diverse dataset is compiled comprising approximately 18,500 MRI images, representing various types of brain tumors. EfficientNet models are trained, validated, and tested using a combination of Softmax and ReLU activation functions, along with the Adam optimizer by employing learning rates of 0.0005 and 0.00001 for model training and optimization. The experimental results underscore the significant potential of the EfficientNet family in brain tumor classification. The findings reveal a consistent improvement in tumor detection accuracy as model complexity increases. The attained accuracies for different EfficientNet models are; 96.07% (EfficientNet-B0), 97.86% (EfficientNet-B1), 98.21% (EfficientNet-B2), 97.86% (EfficientNet-B3), 98.93% (EfficientNet-B4), 99.64% (EfficientNet-B5), 98.57% (EfficientNet-B6), and 99.64% (EfficientNet-B7), respectively. The proposed research not only validates the efficiency of the EfficientNet architecture in classifying brain tumors but also offers valuable insights into how model complexity influences classification performance. The notably high accuracy rates emphasize the clinical promise of employing deep learning methods to aid radiologists and medical experts in precise and efficient brain tumor diagnosis. Additionally, the paper's scope adds to the growing body of knowledge regarding the application of deep learning techniques to enhance medical image analysis and diagnostic capabilities.

Advancing Accuracy in Multimodal Medical Tasks Through Bootstrapped Language-Image Pretraining (BioMedBLIP): Performance Evaluation Study.

Medical image analysis, particularly in the context of visual question answering (VQA) and image captioning, is crucial for accurate diagnosis and educational purposes. Our study aims to introduce BioMedBLIP models, fine-tuned for VQA tasks using specialized medical data sets such as Radiology Objects in Context and Medical Information Mart for Intensive Care-Chest X-ray, and evaluate their performance in comparison to the state of the art (SOTA) original Bootstrapping Language-Image Pretraining (BLIP) model. We present 9 versions of BioMedBLIP across 3 downstream tasks in various data sets. The models are trained on a varying number of epochs. The findings indicate the strong overall performance of our models. We proposed BioMedBLIP for the VQA generation model, VQA classification model, and BioMedBLIP image caption model. We conducted pretraining in BLIP using medical data sets, producing an adapted BLIP model tailored for medical applications. In VQA generation tasks, BioMedBLIP models outperformed the SOTA on the Semantically-Labeled Knowledge-Enhanced (SLAKE) data set, VQA in Radiology (VQA-RAD), and Image Cross-Language Evaluation Forum data sets. In VQA classification, our models consistently surpassed the SOTA on the SLAKE data set. Our models also showed competitive performance on the VQA-RAD and PathVQA data sets. Similarly, in image captioning tasks, our model beat the SOTA, suggesting the importance of pretraining with medical data sets. Overall, in 20 different data sets and task combinations, our BioMedBLIP excelled in 15 (75%) out of 20 tasks. BioMedBLIP represents a new SOTA in 15 (75%) out of 20 tasks, and our responses were rated higher in all 20 tasks (P<.005) in comparison to SOTA models. Our BioMedBLIP models show promising performance and suggest that incorporating medical knowledge through pretraining with domain-specific medical data sets helps models achieve higher performance. Our models thus demonstrate their potential to advance medical image analysis, impacting diagnosis, medical education, and research. However, data quality, task-specific variability, computational resources, and ethical considerations should be carefully addressed. In conclusion, our models represent a contribution toward the synergy of artificial intelligence and medicine. We have made BioMedBLIP freely available, which will help in further advancing research in multimodal medical tasks.

A dynamic uncertainty-aware ensemble model: Application to lung cancer segmentation in digital pathology

Ensemble models have emerged as a powerful technique for improving robustness in medical image segmentation. However, traditional ensembles suffer from limitations such as under-confidence and over-reliance on poor performing models. In this work, we introduce an Adaptive Uncertainty-based Ensemble (AUE) model for tumor segmentation in histopathological slides. Our approach leverages uncertainty estimates from Monte Carlo dropout during testing to dynamically select the optimal pair of models for each whole slide image. The AUE model combines predictions from the two most reliable models (K-Net, ResNeSt, Segformer, Twins), identified through uncertainty quantification, to enhance segmentation performance. We validate the AUE model on the ACDC@LungHP challenge dataset, systematically comparing it against state-of-the-art approaches. Results demonstrate that our uncertainty-guided ensemble achieves a mean Dice score of 0.8653 and outperforms traditional ensemble techniques and top-ranked methods from the challenge by over 3 %. Our adaptive ensemble approach provides accurate and reliable lung tumor delineation in histopathology images by managing model uncertainty.