Hydrogels in biosensing and medical diagnostics

Medical X-ray sources now and for the future

Knee osteoarthritis (KOA) is a leading cause of disability, particularly affecting older adults due to the deterioration of articular cartilage within the knee joint. This condition is characterized by pain, stiffness, and impaired movement, posing a significant challenge in medical diagnostics and treatment planning, especially due to the current inability for early and accurate detection or monitoring of disease progression. This research introduces a multifaceted approach employing feature extraction and machine learning (ML) to improve the accuracy of diagnosing and classifying KOA stages from radiographic images. Utilizing a dataset of 3154 knee X-ray images, this study implemented feature extraction methods such as Histogram of Oriented Gradients (HOG) with Linear Discriminant Analysis (LDA) and Min–Max scaling to prepare the data for classification. The study evaluates six ML classifiers—K Nearest Neighbors classifier, Support Vector Machine (SVM), Gaussian Naive Bayes, Decision Tree, Random Forest, and XGBoost—optimized via GridSearchCV for hyperparameter tuning within a 10-fold Stratified K-Fold cross-validation framework. An ensemble model has also been made for the already high-accuracy models to explore the possibility of enhancing the accuracy and reducing the risk of overfitting. The XGBoost classifier and the ensemble model emerged as the most efficient for multiclass classification, with an accuracy of 98.90%, distinguishing between healthy and unhealthy knees. These results underscore the potential of integrating advanced ML methodologies for the nuanced and accurate diagnosis and classification of KOA, offering new avenues for clinical application and future research in medical imaging diagnostics.

A review of non-invasive blood glucose monitoring through breath acetone and body surface

The healthcare sector is one of the sectors in the EU economy with significant employment potential, which is driven by the growing demand for healthcare. Job satisfaction contributes to the improvement of the quality of service, in every profession, and it is especially important in the field of work of professionals whose service should contribute to better and more efficient health care, and therefore it has been an important research topic for a number of years. The health care sector is facing many challenges. Therefore, providing a sufficient number of well-qualified and motivated staff will be a particular problem in countries facing a marked increase in the demand for health care. It is estimated that the shortage of health workforce will reach 12.9 million health workers by 2035. The health sector of the Republic of Serbia is facing similar challenges, but it also has its own specificities, conditioned by the historical, socio-economic, cultural, and political development of the country. This paper analyses the available literature on job satisfaction, factors influencing satisfaction, and challenges for improving job satisfaction in the health sector.

The ability to manufacture sensors rapidly and cost-effectively is essential for creating devices that would benefit healthcare, medical diagnostics, environmental monitoring and process industries. Because of the inherent complexity of sensing devices, new fabrication methods are needed to create sensors that could be fabricated in large scale and with high efficiency. Additive manufacturing techniques like 3D printing and 3D bioprinting enable printing of different biomaterials into intricate 3D architectures which could serve as sensing platforms. This presentation will describe the development and optimization of 3D printed hydrogel-based (bio)sensors with incorporated receptor molecules and transduction interfaces. In order to 3D print these sensors, an extrusion-based 3D bioprinter, and a computer-aided designing software were used along with a novel bio-ink formulation, which contains enzymes, catalytic and photoactive properties. The optimum hydrogel composite provided excellent mechanical properties to the printed biosensors enabling them to be used in a wide range of wearable applications. Fabrication and performance characteristics of two types of sensors will be presented: a lactate sensor for measuring physiological activity in sweat and a UV sensor for measuring harmful UV radiation exposure. The proposed sensors are reagent-less and highly portable making them ideal for use as wearable devices and printed biocatalytic platforms for portable bioelectronics. Examples of applications of these additively manufactured sensors for monitoring different analytes will be discussed. Figure 1

High-speed image analysis is a cornerstone in modern applications such as autonomous navigation, real-time surveillance, and medical imaging diagnostics. Traditional serial signal processing frameworks are often insufficient to meet the stringent latency and throughput requirements of these domains. This article explores the evolution and effectiveness of parallel signal processing techniques tailored for high-speed image analysis. We present an overview of data-level, task-level, and pipeline-level parallelism, highlighting their implementation across GPUs, FPGAs, and multi-core CPUs. Performance comparisons and architectural trade-offs are examined to identify suitable use cases for each method. Through benchmark analyses, we demonstrate that parallel signal processing enables significant improvements in computation speed, real-time responsiveness, and resource utilization, paving the way for scalable image analytics.

Diabetic Retinopathy (DR) is a leading cause of vision im-pairment and blindness among diabetic patients worldwide.Early detec-tion and timely intervention are crucial to preventing severe visual loss.Convolutional Neural Networks (CNNs) have emerged as a powerful tool for image analysis and have shown great potential in medical image di-agnostics.This study explores the application of CNNs for the early diagnosis of DR using retinal fundus images.By leveraging deep learn-ing techniques, our CNN model is trained to automatically detect and classify various stages of DR with high accuracy.The proposed method utilizes a large dataset of annotated retinal images to ensure robustness and generalizability.Experimental results demonstrate that the CNN-based approach significantly outperforms traditional methods, offering a reliable and efficient solution for early DR detection.This advancement in automated image analysis can facilitate timely and precise diagnosis, potentially reducing the burden of diabetic complications and improving patient outcomes.

Multimodal imaging is gaining in importance in the field of personalized medicine and can be described as a current trend in medical imaging diagnostics. The segmentation, classification and analysis of tissue structures is essential. The goal of this study is the evaluation of established segmentation methods on different medical image data sets acquired with different diagnostic procedures. Established segmentation methods were implemented using the latest state of the art and applied to medical image data sets. In order to benchmark the segmentation performance quantitatively, medical image data sets were superimposed with artificial Gaussian noise, and the mis-segmentation as a function of the image SNR or CNR was compared to a gold standard. The evaluation of the image segmentation showed that the best results of pixel-based segmentation (< 3%) can be achieved with methods of machine learning, multithreshold and advanced level-set method - even at high artificial noise (SNR< 18). Finally, the complexity of the object geometry and the contrast of the ROI to the surrounding tissue must also be considered to select the best segmentation algorithm.

The increasing deployment of deep learning (DL) models in high-stakes domains—such as medical imaging diagnostics, autonomous driving, and financial forecasting—has underscored a critical gap between model performance and interpretability. Conventional post hoc explainability techniques, including feature-attribution methods like LIME and SHAP, provide cursory insights into model decisions but often at the expense of fidelity, robustness, and scalability. This manuscript proposes a transformative framework of futuristic explainability models that seamlessly integrate interpretability mechanisms into the core of DL architectures. We introduce three novel paradigms—Predictive Concept Synthesis (PCS), Counterfactual Knowledge Distillation (CKD), and Adaptive Modular Exposition (AME)—each designed to meet the dual objectives of explanatory transparency and inference efficiency. PCS embeds disentangled, human-readable concepts during model training to generate real-time activation maps; CKD empowers “what-if” exploration by distilling counterfactual reasoning into lightweight student networks; and AME dynamically partitions networks into semantically coherent modules, offering modular rationales aligned with domain expertise. Through a mixed-methods evaluation involving 150 computer-vision specialists and a ResNet-50 model trained on the CIFAR-10 dataset, we assess explanation fidelity (Spearman’s ρ), expert trust (Likert scale), and computational overhead (milliseconds per inference). Statistical analyses—ANOVA followed by Tukey’s Honest Significant Difference tests—reveal that CKD significantly outperforms baseline methods in both fidelity (mean ρ = 0.81 vs. 0.68 for SHAP) and user trust (mean = 4.3/5 vs. 3.6/5), while PCS and AME also yield substantial gains over LIME and SHAP. Importantly, these paradigms maintain inference times within practical limits (< 200 ms), demonstrating that deep integration of causal, symbolic, and modular reasoning need not compromise operational viability. We discuss deployment strategies, potential for unsupervised concept discovery, domain adaptation challenges, and avenues for extending these paradigms to multi-modal and sequential data contexts. Our findings chart a path toward DL systems that are not only accurate but inherently transparent and accountable.

The integration of artificial intelligence (AI), particularly through machine learning (ML) and deep learning (DL) algorithms, marks a transformative progression in medical imaging diagnostics. This technical note elucidates a novel methodology for semantic segmentation of the vertebral column in CT scans, exemplified by a dataset of 250 patients from Riga East Clinical University Hospital. Our approach centers on the accurate identification and labeling of individual vertebrae, ranging from C1 to the sacrum-coccyx complex. Patient selection was meticulously conducted, ensuring demographic balance in age and sex, and excluding scans with significant vertebral abnormalities to reduce confounding variables. This strategic selection bolstered the representativeness of our sample, thereby enhancing the external validity of our findings. Our workflow streamlined the segmentation process by eliminating the need for volume stitching, aligning seamlessly with the methodology we present. By leveraging AI, we have introduced a semi-automated annotation system that enables initial data labeling even by individuals without medical expertise. This phase is complemented by thorough manual validation against established anatomical standards, significantly reducing the time traditionally required for segmentation. This dual approach not only conserves resources but also expedites project timelines. While this method significantly advances radiological data annotation, it is not devoid of challenges, such as the necessity for manual validation by anatomically skilled personnel and reliance on specialized GPU hardware. Nonetheless, our methodology represents a substantial leap forward in medical data semantic segmentation, highlighting the potential of AI-driven approaches to revolutionize clinical and research practices in radiology.

Breast cancer is the most common malignant tumor in females around the world, representing 25.2% of all cancers in women. About 1.7 million women were diagnosed with breast cancer worldwide in 2012 with a death rate of about 522,000 1,2 . The most frequently used methods in breast cancer screening are imaging methods, i.e. ultrasonography and mammography. A common feature of these methods is that they inherently involve the use of expensive and advanced equipment. The development of advanced computer systems allowed for the continuation of research started already in the 1980s 3 and the use of contact thermography in breast cancer screening. The physiological basis for the application of thermography in medical imaging diagnostics is the so-called dermothermal effect related to higher metabolism rate around focal neoplastic lesion. This phenomenon can occur on breast surface as localized temperature anomalies 4 . The device developed by Braster is composed of a detector that works on the basis of thermotropic liquid crystals, image acquisition device and a computer system for image data processing and analysis. Production of the liquid crystal detector was based on a proprietary CLCF technology (Continuous Liquid Crystal Film). In 2014 Braster started feasibility study to prove that there is a potential for artificial intelligence in early breast cancer detection using Braster’s proprietary technology. The aim of this study was to develop a computer system, using a client-server architecture, to an automatic interpretation of thermographic pictures created by the Braster devices.

The role of polysaccharide-based biodegradable soft polymers in the healthcare sector

Objective Nanomedicine is the application of nanotechnology to medicine and, therefore, a potentially transformative approach to healthcare. This new and interdisciplinary field has three main applications in diagnostics, controlled/targeted drug delivery, and regenerative medicine. Nanomedicine has great potential to change human health by advancing diagnosis, prevention and treatment for a wide variety of diseases, including cancer, cardiovascular conditions, and neurological disorders. Despite the overwhelming potential, there are some issues impeding the complete integration of nanomedicines into healthcare systems. Significance This policy brief addresses such critical issues to inform and guide decision-making on effectively deploying nanomedicine to improve patient outcomes and advance important public health initiatives. In addition, some prospects were also presented for the future. It discusses the current barriers to their wide application, particularly regarding regulatory hurdles and the production of robust clinical evidence. Key findings These brand-new nanosystems have some serious drawbacks in regard to safety, efficacy, and regulatory compliance, not to mention public acceptance. Nanomaterials can be so complex that their manufacturing processes become complex, which may potentially bring into question long-term effects on human health and the environment. Conclusions This policy brief identifies key considerations for policymakers and stakeholders, highlighting the requirement for integration among researchers, clinicians, and regulatory bodies. To facilitate the safe and successful integration of nanomedicines into patient care, continued collaboration are imperative. Priority in the future should be given to developing comprehensive regulatory frameworks, raising public awareness, and promoting interdisciplinary research to resolve existing challenges and unlock the potential of nanomedicines in the healthcare sector.

The concern has been raised regarding errors in drugs prescription and medical diagnostics that need to be carefully thought through. Both patient diagnosis and medication prescription are the responsibilities of healthcare providers. As the number of people with health issues rises, the healthcare professionals’ burden is increased. Medical errors may occur in the healthcare sector as a result of healthcare professionals prescribing drugs medicines based on inadequate information related to patient history and drug side effects. Therefore, this study aims to propose a drug recommender system to assist healthcare providers in decision making when prescribing drugs for patients depending on their diagnoses. Drug reviews sentiments are analyzed to find the drug effectiveness among the users. Furthermore, the most suitable recommender algorithm for recommending drugs based on the data from healthcare professionals are selected for this study. Opinion mining is applied on drug reviews, and a hybrid method is implemented to overcome the limitations of content-based and collaborative filtering methods, such as the cold start problem and increasing client preference. The system is developed and tested successfully. The proposed system can assist healthcare professionals in drug decision making and sustain the whole digital care pathway for various diseases.

In recent years, there has been significant growth and burgeoning interest in utilizing nanoparticles for various biomedical applications, including medical diagnostics, targeted drug delivery, tissue engineering, regenerative medicine, and biomedical textiles. In particular, nanoparticles functionalized with biological molecules have unique properties and are very effective in medical diagnostics. Besides that, nanoparticles have a wide range of therapeutic applications, including the development of nanodrug delivery systems, the design of novel drugs, as well as their contribution to the design of therapeutic materials. This chapter provides an overview of recent advancements in the biomedical applications of nanoparticles. Finally, this chapter discusses the challenges of the toxicological evaluation of engineered nanoparticles and the importance of conducting detailed studies on the synthesis of future nanomaterials to develop cutting-edge technologies for addressing a wide range of biomedical issues.