RESUMEN

In therapeutic diagnostics, early diagnosis and monitoring of heart disease is dependent on fast time-series MRI data processing. Robust encryption techniques are necessary to guarantee patient confidentiality. While deep learning (DL) algorithm have improved medical imaging, privacy and performance are still hard to balance. In this study, a novel approach for analyzing homomorphivally-encrypted (HE) time-series MRI data is introduced: The Multi-Faceted Long Short-Term Memory (MF-LSTM). This method includes privacy protection. The MF-LSTM architecture protects patient's privacy while accurately categorizing and forecasting cardiac disease, with accuracy (97.5%), precision (96.5%), recall (98.3%), and F1-score (97.4%). While segmentation methods help to improve interpretability by identifying important region in encrypted MRI images, Generalized Histogram Equalization (GHE) improves image quality. Extensive testing on selected dataset if encrypted time-series MRI images proves the method's stability and efficacy, outperforming previous approaches. The finding shows that the suggested technique can decode medical image to expose visual representation as well as sequential movement while protecting privacy and providing accurate medical image evaluation.

Asunto(s)

Cardiopatías , Imagen por Resonancia Magnética , Privacidad , Humanos , Imagen por Resonancia Magnética/métodos , Cardiopatías/diagnóstico por imagen , Seguridad Computacional , Algoritmos , Procesamiento de Imagen Asistido por Computador/métodos , Femenino , Masculino , Aprendizaje Profundo , Memoria a Corto Plazo , Confidencialidad , Persona de Mediana Edad

RESUMEN

Today, doctors rely heavily on medical imaging to identify abnormalities. Proper classification of these abnormalities enables them to take informed actions, leading to early diagnosis and treatment. This paper introduces the "EfficientKNN" model, a novel hybrid deep learning approach that combines the advanced feature extraction capabilities of EfficientNetB3 with the simplicity and effectiveness of the k-Nearest Neighbors (k-NN) algorithm. Initially, EfficientNetB3, pre-trained on ImageNet, is repurposed to serve as a feature extractor. Subsequently, a GlobalAveragePooling2D layer is applied, followed by an optional Principal Component Analysis (PCA) to reduce dimensionality while preserving critical information. PCA is used selectively when deemed necessary. The extracted features are then classified using an optimized k-NN algorithm, fine-tuned through meticulous cross-validation.Our model underwent rigorous training using a curated dataset containing benign, malignant, and normal medical images. Data augmentation techniques, including rotations, shifts, flips, and zooms, were employed to help the model generalize and efficiently handle new, unseen data. To enhance the model's ability to identify the important features necessary for accurate predictions, the dataset was refined using segmentation and overlay techniques. The training utilized an ensemble of optimization algorithms-SGD, Adam, and RMSprop-with hyperparameters set at a learning rate of 0.00045, a batch size of 32, and up to 120 epochs, facilitated by early stopping to prevent overfitting.The results demonstrate that the EfficientKNN model outperforms traditional models such as VGG16, AlexNet, and VGG19 in terms of accuracy, precision, and F1-score. Additionally, the model showed better results compared to EfficientNetB3 alone. Achieving a 100 % accuracy rate on multiple tests, the EfficientKNN model has significant potential for real-world diagnostic applications. This study highlights the model's scalability, efficient use of cloud storage, and real-time prediction capabilities, all while minimizing computational demands.By integrating the strengths of EfficientNetB3's deep learning architecture with the interpretability of k-NN, EfficientKNN presents a significant advancement in medical image classification, promising improved diagnostic accuracy and clinical applicability.

Asunto(s)

Aprendizaje Profundo , Humanos , Femenino , Neoplasias de la Mama/diagnóstico por imagen , Ultrasonografía Mamaria/métodos , Algoritmos , Mama/diagnóstico por imagen , Interpretación de Imagen Asistida por Computador/métodos , Análisis de Componente Principal

RESUMEN

BACKGROUND: Melanoma is one of the most malignant forms of skin cancer, with a high mortality rate in the advanced stages. Therefore, early and accurate detection of melanoma plays an important role in improving patients' prognosis. Biopsy is the traditional method for melanoma diagnosis, but this method lacks reliability. Therefore, it is important to apply new methods to diagnose melanoma effectively. AIM: This study presents a new approach to classify melanoma using deep neural networks (DNNs) with combined multiple modal imaging and genomic data, which could potentially provide more reliable diagnosis than current medical methods for melanoma. METHOD: We built a dataset of dermoscopic images, histopathological slides and genomic profiles. We developed a custom framework composed of two widely established types of neural networks for analysing image data Convolutional Neural Networks (CNNs) and networks that can learn graph structure for analysing genomic data-Graph Neural Networks. We trained and evaluated the proposed framework on this dataset. RESULTS: The developed multi-modal DNN achieved higher accuracy than traditional medical approaches. The mean accuracy of the proposed model was 92.5% with an area under the receiver operating characteristic curve of 0.96, suggesting that the multi-modal DNN approach can detect critical morphologic and molecular features of melanoma beyond the limitations of traditional AI and traditional machine learning approaches. The combination of cutting-edge AI may allow access to a broader range of diagnostic data, which can allow dermatologists to make more accurate decisions and refine treatment strategies. However, the application of the framework will have to be validated at a larger scale and more clinical trials need to be conducted to establish whether this novel diagnostic approach will be more effective and feasible.

Asunto(s)

Aprendizaje Profundo , Dermoscopía , Melanoma , Neoplasias Cutáneas , Humanos , Melanoma/genética , Melanoma/diagnóstico por imagen , Melanoma/diagnóstico , Melanoma/patología , Neoplasias Cutáneas/genética , Neoplasias Cutáneas/diagnóstico por imagen , Neoplasias Cutáneas/patología , Dermoscopía/métodos , Redes Neurales de la Computación , Reproducibilidad de los Resultados , Genómica/métodos , Femenino , Masculino , Persona de Mediana Edad , Adulto , Anciano

RESUMEN

The structure and function of dietary proteins, as well as their subcellular prediction, are critical for designing and developing new drug compositions and understanding the pathophysiology of certain diseases. As a remedy, we provide a subcellular localization method based on feature fusion and clustering for dietary proteins. Additionally, an enhanced PseAAC (Pseudo-amino acid composition) method is suggested, which builds upon the conventional PseAAC. The study initially builds a novel model of representing the food protein sequence by integrating autocorrelation, chi density, and improved PseAAC to better convey information about the food protein sequence. After that, the dimensionality of the fused feature vectors is reduced by using principal component analysis. With prediction accuracies of 99.24% in the Gram-positive dataset and 95.33% in the Gram-negative dataset, respectively, the experimental findings demonstrate the practicability and efficacy of the proposed approach. This paper is basically exploring pseudo-amino acid composition of not any clinical aspect but exploring a pharmaceutical aspect for drug repositioning.

Asunto(s)

Proteínas en la Dieta , Proteínas en la Dieta/química , Proteínas en la Dieta/análisis , Proteínas en la Dieta/metabolismo , Aminoácidos/química , Aminoácidos/análisis , Preparaciones Farmacéuticas/química , Preparaciones Farmacéuticas/análisis

RESUMEN

The exponential distribution optimizer (EDO) represents a heuristic approach, capitalizing on exponential distribution theory to identify global solutions for complex optimization challenges. This study extends the EDO's applicability by introducing its multi-objective version, the multi-objective EDO (MOEDO), enhanced with elite non-dominated sorting and crowding distance mechanisms. An information feedback mechanism (IFM) is integrated into MOEDO, aiming to balance exploration and exploitation, thus improving convergence and mitigating the stagnation in local optima, a notable limitation in traditional approaches. Our research demonstrates MOEDO's superiority over renowned algorithms such as MOMPA, NSGA-II, MOAOA, MOEA/D and MOGNDO. This is evident in 72.58% of test scenarios, utilizing performance metrics like GD, IGD, HV, SP, SD and RT across benchmark test collections (DTLZ, ZDT and various constraint problems) and five real-world engineering design challenges. The Wilcoxon Rank Sum Test (WRST) further confirms MOEDO as a competitive multi-objective optimization algorithm, particularly in scenarios where existing methods struggle with balancing diversity and convergence efficiency. MOEDO's robust performance, even in complex real-world applications, underscores its potential as an innovative solution in the optimization domain. The MOEDO source code is available at: https://github.com/kanak02/MOEDO .

RESUMEN

Wireless sensor networks (WSNs) enable communication among sensor nodes and require efficient energy management for optimal operation under various conditions. Key challenges include maximizing network lifetime, coverage area, and effective data aggregation and planning. A longer network lifetime contributes to improved data transfer durability, sensor conservation, and scalability. In this paper, an enhanced dual-selection krill herd (KH) optimization clustering scheme for resource-efficient WSNs with minimal overhead is introduced. The proposed approach increases overall energy utilization and reduces inter-node communication, addressing energy conservation challenges in node deployment and clustering for WSNs as optimization problems. A dynamic layering mechanism is employed to prevent repetitive selection of the same cluster head nodes, ensuring effective dual selection. Our algorithm is designed to identify the optimal solution through enhanced exploitation and exploration processes, leveraging a modified krill-based clustering method. Comparative analysis with benchmark approaches demonstrates that the proposed model enhances network lifetime by 23.21%, increases stable energy by 19.84%, and reduces network latency by 22.88%, offering a more efficient and reliable solution for WSN energy management.