Medical Applications Research Articles

Controlling the lifetime of biodegradable electronics: from dissolution kinetics to trigger acceleration

Biodegradable electronics have revolutionized the field of medical devices by offering inherent advantages such as natural disintegration after a useful functional period, thereby eliminating the need for removal surgery. This paradigm shift addresses challenges with long-term implantation, the risks of secondary surgeries, and potential complications, offering a safer and more patient-friendly approach to temporary implantable devices. This review delves into the dissolution kinetics of materials and strategies for lifetime control providing a comprehensive overview of recent advancements in biodegradable electronics. Understanding the kinetics is crucial for meeting the required functional lifetime for implantable medical applications, which varies based on application scope and target diseases. The dissolution kinetics of silicon and biodegradable metals form the core of the discussion, focusing on recent studies aimed at controlling the dissolution rate and enhancing properties. The exploration extends to ideas for accelerating material degradation or initiating on-demand degradation in biodegradable electronics after stable function. Additionally, the compilation of encapsulation layer materials and strategies enhances understanding of how to improve the stable operation time of devices. Emphasis is placed on efforts to adjust the lifetime of biodegradable electronics, particularly in medical applications.

American Society of Nephrology Kidney TREKS Program.

In response to decreasing numbers of individuals entering into nephrology fellowships, the American Society of Nephrology launched Kidney TREKS (Tutored Research and Education for Kidney Scholars) to stimulate interest in nephrology among medical students, graduate students, and postdoctoral fellows. The program combines a one-week intensive exposure to kidney physiology with a longitudinal mentorship program at the participants' home institutions. Ten years in, an analysis was conducted to assess its effectiveness. We surveyed participants to assess their opinions regarding nephrology before and after the course and followed them longitudinally to determine their career choices. TREKS applicants who were not selected to participate were used as a comparison group. 381 people participated in the program and 242 completed the survey. After TREKS, both medical students and graduate students showed increased interest in nephrology, with rank scores of 5.6±0.2 pre- to 7.5±0.1 post-course for medical students (mean ± standard deviation, n=189, p=0.001) and 7.3±0.3 to 8.7±0.3 (n=53, p=0.001) for graduate students. In long term follow-up, TREKS medical students chose a nephrology pipeline residency at a higher rate than medical students overall (57% vs. 31%, p=0.01) and TREKS applicants who did not participate (47% vs. 31%, p=0.04). Nephrology fellowship rates for these groups exceeded the general population but did not significantly differ between TREKS participants and applicants. PhD students and postdoctoral TREKS participants had a higher rate of participating in nephrology research compared to TREKS applicants (66% vs. 30%, p=0.01). In summary, the ASN Kidney TREKS program has demonstrated that it can improve interest in nephrology in the short term and increase the number of individuals going into nephrology careers. This long-term effect is most evident in PhD students and postdoctoral participants. Further study is needed to assess the impact of TREKS on enrollment in nephrology fellowship programs.

From genome to proteome: Comprehensive identification of venom toxins from the Chinese funnel-web spider (Macrothelidae: Macrothele yani)

Macrothelidae is a family of mygalomorph spiders containing the extant genera Macrothele and Vacrothele. China is an important center of diversity for Macrothele with 65 % of the known species occurring there. Previous work on Macrothele was able to uncover several important toxin compounds including Raventoxin which may have applications in biomedicine and agricultural chemistry. Despite the importance of Macrothele spiders, high-quality reference genomes are still lacking, which hinders our understanding and application of the toxin compounds. In this study, we assembled the genome of the Macrothele yani to help fill gaps in our understanding of toxin biology in this lineage of spiders to encourage the future study and applications of these compounds. The final assembled genome was 6.79 Gb in total length, had a contig N50 of 21.44 Mb, and scaffold N50 of 156.16 Mb. Hi-C scaffolding assigned 98.19 % of the genome to 46 pseudo-chromosomes with a BUSCO score of 95.7 % for the core eukaryotic gene set. The assembled genome was found to contain 75.62 % repetitive DNA and a total of 39,687 protein-coding genes were annotated making it the spider genome with highest number of genes. Through integrated analysis of venom gland transcriptomics and venom proteomics, a total of 194 venom toxins were identified, including 38 disulfide-rich peptide neurotoxins, among which 12 were ICK knottin peptides. In summary, we present the first high-quality genome assembly at the chromosomal level for any Macrothelidae spider, filling an important gap in our knowledge of these spiders. Such high-quality genomic data will be invaluable as a reference in resolving Araneae spider phylogenies and in screening different spider species for novel compounds applicable to numerous medical and agricultural applications.

Multi-channel prismatic localized surface plasmon resonance biosensor for real-time competitive assay multiple COVID-19 characteristic miRNAs

A multi-channel prismatic localized surface plasmon resonance (LSPR) biosensor was developed for quantitative and real-time detection of multiple COVID-19 characteristic miRNAs. The well-dispersed and dense gold nanoparticles (AuNPs) arrays for LSPR biosensing were fabricated through a nano-thickness diblock copolymer template (BCPT). Both theoretical and experimental analyses were conducted to investigate the effects of particle size, interparticle spacing, and surface coverage on LSPR sensing spectrum and intensity sensitivity of varied AuNPs sizes. A competitive assay strategy was proposed and used for non-amplification miRNA detection with a low limit detection of 3.41 nM, while a four-channel prismatic LSPR system enables parallel detection of multiple miRNAs. Furthermore, this sensing strategy can effectively and specifically identify target miRNA, distinguish mismatched miRNA and interfering miRNA, and exhibit low non-specific adsorption. This BCPT-based LSPR biosensor demonstrates the practicality and potential of a multi-channel, adaptable, and integrated prismatic sensor in medical testing and diagnostic applications.

Cancerous and Non-Cancerous MRI Classification Using Dual DCNN Approach.

Brain cancer is a life-threatening disease requiring close attention. Early and accurate diagnosis using non-invasive medical imaging is critical for successful treatment and patient survival. However, manual diagnosis by radiologist experts is time-consuming and has limitations in processing large datasets efficiently. Therefore, efficient systems capable of analyzing vast amounts of medical data for early tumor detection are urgently needed. Deep learning (DL) with deep convolutional neural networks (DCNNs) emerges as a promising tool for understanding diseases like brain cancer through medical imaging modalities, especially MRI, which provides detailed soft tissue contrast for visualizing tumors and organs. DL techniques have become more and more popular in current research on brain tumor detection. Unlike traditional machine learning methods requiring manual feature extraction, DL models are adept at handling complex data like MRIs and excel in classification tasks, making them well-suited for medical image analysis applications. This study presents a novel Dual DCNN model that can accurately classify cancerous and non-cancerous MRI samples. Our Dual DCNN model uses two well-performed DL models, i.e., inceptionV3 and denseNet121. Features are extracted from these models by appending a global max pooling layer. The extracted features are then utilized to train the model with the addition of five fully connected layers and finally accurately classify MRI samples as cancerous or non-cancerous. The fully connected layers are retrained to learn the extracted features for better accuracy. The technique achieves 99%, 99%, 98%, and 99% of accuracy, precision, recall, and f1-scores, respectively. Furthermore, this study compares the Dual DCNN's performance against various well-known DL models, including DenseNet121, InceptionV3, ResNet architectures, EfficientNetB2, SqueezeNet, VGG16, AlexNet, and LeNet-5, with different learning rates. This study indicates that our proposed approach outperforms these established models in terms of performance.

On the Actuation Technologies of Biomedical Microrobot: A Summarized Review

In recent years, wireless microrobots have gotten more attention due to their huge potential in the biomedical field, especially drug delivery. Microrobots have several benefits, including small size, low weight, sensitivity, and flexibility. These characteristics have led to microscale improvements in control systems and power delivery with the development of submillimeter-sized robots. Wireless control of individual mobile microrobots has been achieved using a variety of propulsion systems, and improving the actuation and navigation of microrobots will have a significant impact. On the other hand, actuation tools must be integrated and compatible with the human body to drive these untethered microrobots along predefined paths inside biological environments. This study investigated key microrobot components, including medical applications, actuation systems, control systems, and design schemes. The efficiency of a microrobot is impacted by many factors, including the material, structure, and environment of the microrobot. Furthermore, integrating a hybrid actuation system and multimodal imaging can increase the microrobot's navigation effect, imaging algorithms, and working environment. In addition, taking into account the human body's moving distance, autonomous actuating technology could be used to deliver microrobots precisely and quickly to a specific position using a combination of quick approaches.

Synthesis and anti-breast cancer evaluation of fused imidazole-imidazo[1,2-c][1,2,3]triazoles: PEG-400 mediated one-pot reaction under ultrasonic irradiation

Synthetic chemists have organized easy and effective ways for perfect synthesis in response to the requirement for scaffolds that are crucial for medical applications. To investigate the synthesis of fused 1,2,3-triazoles in the presence of ultrasound, the [3 + 2] cycloaddition followed by CN bond formation reaction between 2-chloro-N-(1-methyl-1H-imidazol-2-yl)acetamide and substituted aryl iodoalkynes was carried out ourpreviously reported condition. The cancer activity of the synthesized compounds were then tested in vitro against two breast cancer cell lines MCF-7 and MDA-MB231 and some of the compounds 5g and 5j showed more potent activity against MCF-7 compared to standard erlotinib (IC50 = 4.70 ± 0.11 μM) with IC50 values 4.02 ± 0.74 and 4.23 ± 0.65 μM. Later, in vitro EGFR results revealed that compound 5g (IC50 = 0.38 ± 0.02 μM) have shown more effective than the conventional medicine erlotinib (IC50 = 0.42 ± 0.04 μM). Compounds 5j (IC50 = 0.42 ± 0.05 μM) have shown equipotent, 5i (IC50 = 0.53 ± 0.02 μM) and 5k (IC50 = 0.58 ± 0.02 μM) were shown good activity compared to the standard erlotinib. In silico investigations of more potent compounds (5g, 5i, 5j, 5k, and 5q) and erlotinib on the EGFR protein indicated that all five compounds had comparable binding energies and inhibition constants than the erlotinib.

Isolation, structural characterization and biological activities of polysaccharides from Chondrus crispus

Biomedical applications often exploit the biodegradability and biocompatibility of natural polysaccharides. In this study, polysaccharide extracts from Chondrus crispus (CC) were characterized using HPLC-SEC, FTIR and 1H NMR spectroscopy. The proliferative and migrative capacities of various extracts on RAW264.7 cells, along with their impact on nitric oxide (NO) production and phagocytosis, were assessed. Polysaccharide bioactivity was also explored using HDF and HaCaT cell lines. This study additionally investigated the anti-inflammatory effects of CC fractions on Caco-2 cells, examining their impact on cytokines (TNF-α, TGF-β1) and chemokine (MCP-1) expression. CC-derived polysaccharides effectively modulated inflammatory activity and improved scratch or exclusion wound healing. Hot (95 °C) extracted polysaccharides fraction CC-2B (λ and, κ-carrageenan) stimulated RAW264.7 cell proliferation, while cold (25 °C) extracted polysaccharides fraction CC-1A (0.5 μg/μL) inhibited cell proliferation at 24 h (69.7 ± 4.8%, p<0.001), while exhibiting increased phagocytosis activity. Furthermore, RAW264.7 cells treated with hot-extracted fraction CC-2B showed a significant reduction in LPS-stimulated NO production (p<0.0001) over 24 h. HaCaT and HDF cell proliferation were stimulated when treated with fraction CC-2B (p<0.0001). Moreover, Caco-2 cells treated with all fractions displayed enhanced TGF-β1 expression and reduced TNF-α expression at 24 h, indicating the potential of these fractions for future medical applications. Polysaccharide cold-extracted fraction (CC-1A) exhibited highly effective anticoagulant properties. All polysaccharide fractions (0.13 μg/μL) appeared to be non-cytotoxic for Caco-2 cells during 24 h. CC-derived polysaccharides have potential applications in the medical (wound healing, gastrointestinal distress), food, cosmetic, and pharmaceutical industries.

Co-ordinate-based positional embedding that captures resolution to enhance transformer’s performance in medical image analysis

Vision transformers (ViTs) have revolutionized computer vision by employing self-attention instead of convolutional neural networks and demonstrated success due to their ability to capture global dependencies and remove spatial biases of locality. In medical imaging, where input data may differ in size and resolution, existing architectures require resampling or resizing during pre-processing, leading to potential spatial resolution loss and information degradation. This study proposes a co-ordinate-based embedding that encodes the geometry of medical images, capturing physical co-ordinate and resolution information without the need for resampling or resizing. The effectiveness of the proposed embedding is demonstrated through experiments with UNETR and SwinUNETR models for infarct segmentation on MRI dataset with AxTrace and AxADC contrasts. The dataset consists of 1142 training, 133 validation and 143 test subjects. Both models with the addition of co-ordinate based positional embedding achieved substantial improvements in mean Dice score by 6.5% and 7.6%. The proposed embedding showcased a statistically significant advantage p-value< 0.0001 over alternative approaches. In conclusion, the proposed co-ordinate-based pixel-wise positional embedding method offers a promising solution for Transformer-based models in medical image analysis. It effectively leverages physical co-ordinate information to enhance performance without compromising spatial resolution and provides a foundation for future advancements in positional embedding techniques for medical applications.

Q-SupCon: Quantum-Enhanced Supervised Contrastive Learning Architecture within the Representation Learning Framework

In the evolving landscape of data privacy regulations, the challenge of providing extensive data for robust deep classification models arises. The accuracy of these models relies on the amount of training data, due to the multitude of parameters that require tuning. Unfortunately, obtaining such ample data proves challenging, particularly in domains like medical applications, where there is a pressing need for robust models for early disease detection but a shortage of labeled data. Nevertheless, the classical supervised contrastive learning models, have shown the potential to address this challenge up to a certain limit, by utilizing deep encoder models. However, recent advancements in quantum machine learning enable the extraction of meaningful representations from extremely limited and simple data. Thus, replacing classical counterparts in classical or hybrid quantum-classical supervised contrastive models enhances feature learning capability with minimal data. Therefore, this work proposes the Q-SupCon model, a fully quantum-powered supervised contrastive learning model comprising a quantum data augmentation circuit, quantum encoder, quantum projection head, and quantum variational classifier, enabling efficient image classification with minimal labeled data. Furthermore, the novel model attains 80%, 60%, and 80% test accuracy on MNIST, KMNIST, and FMNIST datasets, marking a significant advancement in addressing the data scarcity challenge.

Application of Artificial Intelligence in Tissue Engineering.

Tissue engineering, a crucial approach in medical research and clinical applications, aims to regenerate damaged organs. By combining stem cells, biochemical factors, and biomaterials, it encounters challenges in designing complex 3D structures. Artificial intelligence (AI) enhances tissue engineering through computational modeling, biomaterial design, cell culture optimization, and personalized medicine. This review explores AI applications in organ tissue engineering (bone, heart, nerve, skin, cartilage), employing various machine learning (ML) algorithms for data analysis, prediction, and optimization. Each section discusses common ML algorithms and specific applications, emphasizing the potential and challenges in advancing regenerative therapies.

Clinically used lipiodol as an effective radioenhancer

High atomic number (Z) element-based radiosensitizers have been widely reported for enhancing radiotherapeutic effect, while its limited tumor distribution and long-term safety are major limitations in clinics. In this study, we propose the clinically used lipiodol, a liquid radiological imaging agent with high iodine concentration, as a potential radiosensitizer. Compared to clinically approved high Z nanoradiosensitizer HfO2, lipiodol emulsion shows much better tumor spatial distribution and tumor inhibiting ability at in vivo level upon radiation treatment. In addition to primary tumor, lipiodol emulsion also exhibited efficient targetability and inhibition of lymph node metastasis with radiation treatment. Furthermore, the in vitro mechanism study indicates that lipiodol enhances radiotherapeutic effect by generating extra reactive oxygen species to cause DNA damage and cell death, irrespective of cancer type. Lastly, toxicity assessments demonstrate the safety of lipiodol for practical medical applications. This report provides scientific evidence supporting the century-old lipiodol as an effective liquid radiosensitizer ready for clinical implementation.

Abrasive waterjet machining of titanium alloy using an integrated approach of taguchi-based passing vehicle search algorithm

AbstractThe titanium alloy, Ti6Al4V, is a popularly used material in aerospace and medical applications due to its specific attributes, such as excellent strength-to-weight ratio and resistance to corrosion. Ti6Al4V is challenging to machine using conventional methods due to its poor thermal conductivity, which necessitates the use of unconventional machining methods like abrasive water jet machining (AWJM). In this work, AWJM was preferred for machining Ti6Al4V, considering three key process variables: nozzle traverse speed (Tv), abrasive flow rate (Af), and stand-off distance (Sd). The performance of the AWJM process was assessed using three main metrics: Material removal rate (MRR), Surface roughness (SR), and Kerf taper angle (θ), which were crucial for evaluating the effectiveness of the AWJM. Taguchi’s L9 array, a design of experiments method, was used to plan the experiments. The adequacy of the developed models was assessed by analysis of variance (ANOVA). ANOVA results have shown that Tv, Af, and Sd were found to have a significant effect on MRR, SR, θ with contributions of 73.15%, 49.72%, and 78.35% respectively. The Passing Vehicle Search algorithm was adopted to find the global optimal solution. Simultaneous optimization results using the PVS algorithm have shown the optimal MRR, SR, and θ values of 0.17 g/min, 3.28 μm, and 2.45, respectively, at Tv of 237 mm/min, Af of 450 g/min, and Sd of 2.0 mm. The optimization results with Pareto points will help to achieve desired outcomes by selecting appropriate input conditions.

Overtaking Effects on Spherical Shock Wave Propagation in Metallic Mediums

The propagation of spherical shock waves through metallic mediums have been considered in this paper. The effect of overtaking disturbances in the propagation of spherical shock waves through solids, especially metals such as Aluminium and Copper, has been analysed by employing Chisnell-Chester-Whitham method. Under the influence of spherical shock waves flow variables have been obtained analytically as well as estimated numerically. Expressions for the pressure and the particle velocity behind the shock wave have also been obtained. It is found that shock strength in both the metals decreases as shock propagates through them. Shock velocity and modified shock velocity depict a decreasing pattern with increasing propagation distance. These observations suggest many industrial and medical applications of shock waves.

Guest‐Induced Thermally Activated Delayed Fluorescence Organic Supramolcular Macrocycle Scintillators for High‐Resolution X‐Ray Imaging

AbstractOrganic scintillators, pivotal in security and medical applications, face challenges due to limited X‐ray absorption and exciton utilization. Herein, a novel class of organic scintillators is introduced, named guest‐induced thermally activated delayed fluorescence (TADF) within supramolecular macrocycles via host‐guest through‐space charge transfer (TSCT). Four co‐crystals are obtained through orthogonal crystallizations involving calix[3]acridan (C[3]A) and calix[3]phenothiazine (C[3]P) macrocycles as hosts, along with 1,2‐dicyanobenzene (DCB) and 4‐bromo‐1,2‐benzenedicarbonitrile (BrDCB) as guests. Interestingly, DCB@C[3]A and BrDCB@C[3]A co‐crystals exhibit strong host‐guest TSCT with reduced single‐triplet energy gap for efficient TADF emission, which leads to enhanced exciton utilization and X‐ray absorption, yielding radioluminescence intensities over 29 and 25 times higher than C[3]A. Similarly, substituting C[3]A with C[3]P, the obtained TADF co‐crystals also outperform C[3]P in scintillation performance. Additionally, the scintillation color of co‐crystals can be adjusted by varying the electron‐donating abilities of macrocycles and the electron‐accepting abilities of guests, offering a simpler color‐tuning mechanism than covalent‐bonded scintillators. Furthermore, the flexible film based on DCB@C[3]A exhibits promising application in X‐ray radiography, showcasing a high spatial resolution of 20 lp mm−1 @MTF = 0.77. The innovative strategy of fabricating organic scintillators via reversible non‐covalent interactions presents a novel solution for designing color‐tunable and high‐performance scintillators.

A Review On Early Detection Of Chronic Kidney Disease

Early detection of Chronic Kidney Disease (CKD) is critical for timely intervention and effective treatment. Deep learning algorithms have demonstrated promise in medical applications, including disease detection. In this study, we propose a deep learning-based system for early CKD detection using the Chronic Kidney Disease dataset from Kaggle. Additionally, we incorporate the Grasshopper Optimization Algorithm (GOA) for feature selection to enhance the system's performance and interpretability. Our system employs a convolutional neural network (CNN) architecture to analyze clinical and laboratory attributes from the CKD dataset, obtained from Kaggle, consisting of 4,000 instances with 25 attributes. These attributes encompass patient demographics, blood tests, and medical history, providing a comprehensive representation of CKD-related factors. To improve the system's performance, we integrate the GOA for feature selection. The GOA is a nature-inspired metaheuristic optimization algorithm that mimics the foraging behavior of grasshoppers. It aims to identify the most relevant attributes associated with CKD from the dataset. By selecting a subset of informative features, we enhance the model's predictive accuracy and reduce overfitting. During the training phase, the CNN learns to automatically extract relevant features and patterns associated with CKD from the selected attributes. Additionally, data preprocessing techniques such as normalization and feature scaling are applied to further improve the model's performance and generalizability. To evaluate the system's performance, we conduct experiments using a separate test dataset comprising 1,000 instances from the CKD dataset. The incorporation of the GOA for feature selection in our deep learning system not only improves its performance but also enhances interpretability. By identifying the most relevant attributes associated with CKD, we focus on key biomarkers and risk factors, enhancing the system's accuracy and providing valuable insights into the disease. Our research showcases the potential of deep learning algorithms, coupled with GOA-based feature selection, for early CKD detection. By leveraging the Kaggle CKD dataset and incorporating the GOA, we contribute to improving the accuracy and applicability of the system in real-world clinical settings. To handle Big data we are proposing to implement this problem on Pyspark one of the Big data computational environments for effective learning. In this platform, we can dynamically scale the infrastructure as per the demand of the data. Ultimately, our work aims to advance the early detection and management of CKD, leading to improved patient outcomes and more effective healthcare interventions.

The Implementation of AFM-Based Nanoscale Diagnostic Methods in the Investigation of the Degradation Process of Bacteriostatic Acrylic Film with Silver Nanoparticles

In this paper, a custom-tailored investigation protocol aimed at the tests of the resistance of bacteriostatic acrylic-based film containing silver nanoparticles is presented. As hospital appliance applications were considered, it was necessary to provide a unique approach, enabling specific media exposure and utilizing high-sensitivity measurement methods to observe fine indications of material wear. Due to the presence of nanoparticles in the tested film, nanometer-resolution surface imaging is necessary. Therefore, the main source of information about its degradation process is atomic force microscopy (AFM) measurements. This particular tool is an appreciated source of information, providing quantitative data about both morphological and mechanical changes in the properties of the surface. Using such an approach, supported by standard diagnostic methods, such as colorimetry and wettability angle determination, it was possible to enable insights into the way the bacteriostatic film deteriorates and evaluate its usefulness in medical appliance applications. Further tests of various films developed by companies can be performed using the described protocol to determine the lifetime of certain products. This paper reveals the company’s practical utilization of both standardized and novel test techniques in the evaluation of new products.

Theoretical investigations on the laser isotope separation of 175Yb for medical applications

The possibility of laser isotope separation of 175Yb from irradiated natural Yb has been investigated. The optimum process parameters such as powers and bandwidths of the lasers, Doppler broadening and the number density of the atoms have been derived through density matrix calculations. It has been shown that it is possible to produce 175Yb (>42% enriched) at a production rate of 62 μg/hour (or 1.5 mg/day). This corresponds to the production rate of 1350 patient doses (of 7.4 GBq each) per day. The radionuclidic purity of the isotopic mixture is expected to be 99.9999%. The method is highly suitable for the countries having only low-flux nuclear reactors.

Noisecut: a python package for noise-tolerant classification of binary data using prior knowledge integration and max-cut solutions

BackgroundClassification of binary data arises naturally in many clinical applications, such as patient risk stratification through ICD codes. One of the key practical challenges in data classification using machine learning is to avoid overfitting. Overfitting in supervised learning primarily occurs when a model learns random variations from noisy labels in training data rather than the underlying patterns. While traditional methods such as regularization and early stopping have demonstrated effectiveness in interpolation tasks, addressing overfitting in the classification of binary data, in which predictions always amount to extrapolation, demands extrapolation-enhanced strategies. One such approach is hybrid mechanistic/data-driven modeling, which integrates prior knowledge on input features into the learning process, enhancing the model’s ability to extrapolate.ResultsWe present NoiseCut, a Python package for noise-tolerant classification of binary data by employing a hybrid modeling approach that leverages solutions of defined max-cut problems. In a comparative analysis conducted on synthetically generated binary datasets, NoiseCut exhibits better overfitting prevention compared to the early stopping technique employed by different supervised machine learning algorithms. The noise tolerance of NoiseCut stems from a dropout strategy that leverages prior knowledge of input features and is further enhanced by the integration of max-cut problems into the learning process.ConclusionsNoiseCut is a Python package for the implementation of hybrid modeling for the classification of binary data. It facilitates the integration of mechanistic knowledge on the input features into learning from data in a structured manner and proves to be a valuable classification tool when the available training data is noisy and/or limited in size. This advantage is especially prominent in medical and biomedical applications where data scarcity and noise are common challenges. The codebase, illustrations, and documentation for NoiseCut are accessible for download at https://pypi.org/project/noisecut/. The implementation detailed in this paper corresponds to the version 0.2.1 release of the software.

Multimodality Fusion Strategies in Eye Disease Diagnosis.

Multimodality fusion has gained significance in medical applications, particularly in diagnosing challenging diseases like eye diseases, notably diabetic eye diseases that pose risks of vision loss and blindness. Mono-modality eye disease diagnosis proves difficult, often missing crucial disease indicators. In response, researchers advocate multimodality-based approaches to enhance diagnostics. This study is a unique exploration, evaluating three multimodality fusion strategies-early, joint, and late-in conjunction with state-of-the-art convolutional neural network models for automated eye disease binary detection across three datasets: fundus fluorescein angiography, macula, and combination of digital retinal images for vessel extraction, structured analysis of the retina, and high-resolution fundus. Findings reveal the efficacy of each fusion strategy: type 0 early fusion with DenseNet121 achieves an impressive 99.45% average accuracy. InceptionResNetV2 emerges as the top-performing joint fusion architecture with an average accuracy of 99.58%. Late fusion ResNet50V2 achieves a perfect score of 100% across all metrics, surpassing both early and joint fusion. Comparative analysis demonstrates that late fusion ResNet50V2 matches the accuracy of state-of-the-art feature-level fusion model for multiview learning. In conclusion, this study substantiates late fusion as the optimal strategy for eye disease diagnosis compared to early and joint fusion, showcasing its superiority in leveraging multimodal information.