J Pers Med. 2025 Mar 13;15(3):110. doi: 10.3390/jpm15030110.

ABSTRACT

Background/Objectives: Pharmacogenomics is an emerging field in precision medicine that aims to improve patient outcomes by tailoring drug selection and dosage to an individual's genetic makeup. However, patients in rural communities often cannot take advantage of specialized services such as pharmacogenomics due to various barriers that limit access to healthcare. This article aims to identify the barriers to implementing pharmacogenomic initiatives in rural communities and assess strategies for integrating pharmacogenomics into rural healthcare systems. Methods: This article describes the qualitative research that was conducted using semi-structured interviews with various stakeholders in addition to explaining how strategic frameworks were used to synthesize secondary research. Results: The findings of this article indicated mixed awareness of pharmacogenomics as an option amongst stakeholders, highlighting the need for targeted outreach and education intervention. Solutions such as mail-in testing and telemedicine were determined to be feasible solutions to address various geographical and logistical barriers that exist for rural patients. This article determines that successful strategies will leverage existing infrastructure and prioritize patient care, workflow integration, and adoption. Conclusions: Making pharmacogenomics a viable option for rural patients will take a multi-faceted approach that combines outreach, education, and innovative delivery models to overcome the multiple barriers facing rural communities.

PMID:40137426 | DOI:10.3390/jpm15030110

J Pers Med. 2025 Mar 5;15(3):103. doi: 10.3390/jpm15030103.

ABSTRACT

Background: In the United States, approximately 1 in 5 children experience comorbidities with mental illness, including depression and anxiety, which lead to poor general health outcomes. Adolescents with substance use disorders exhibit high rates of co-occurring mental illness, with over 60% meeting diagnostic criteria for another psychiatric condition in community-based treatment programs. Comorbidities are influenced by both genetic (DNA antecedents) and environmental (epigenetic) factors. Given the significant impact of psychiatric comorbidities on individuals' lives, this study aims to uncover common mechanisms through a Genome-Wide Association Study (GWAS) meta-meta-analysis. Methods: GWAS datasets were obtained for each comorbid phenotype, followed by a GWAS meta-meta-analysis using a significance threshold of p < 5E-8 to validate the rationale behind combining all GWAS phenotypes. The combined and refined dataset was subjected to bioinformatic analyses, including Protein-Protein Interactions and Systems Biology. Pharmacogenomics (PGx) annotations for all potential genes with at least one PGx were tested, and the genes identified were combined with the Genetic Addiction Risk Severity (GARS) test, which included 10 genes and eleven Single Nucleotide Polymorphisms (SNPs). The STRING-MODEL was employed to discover novel networks and Protein-Drug interactions. Results: Autism Spectrum Disorder (ASD) was identified as the top manifestation derived from the known comorbid interaction of anxiety, depression, and attention deficit hyperactivity disorder (ADHD). The STRING-MODEL and Protein-Drug interaction analysis revealed a novel network associated with these psychiatric comorbidities. The findings suggest that these interactions are linked to the need to induce "dopamine homeostasis" as a therapeutic outcome. Conclusions: This study provides a reliable genetic and epigenetic map that could assist healthcare professionals in the therapeutic care of patients presenting with multiple psychiatric manifestations, including anxiety, depression, and ADHD. The results highlight the importance of targeting dopamine homeostasis in managing ASD linked to these comorbidities. These insights may guide future pharmacogenomic interventions to improve clinical outcomes in affected individuals.

PMID:40137419 | DOI:10.3390/jpm15030103

J Pers Med. 2025 Mar 4;15(3):101. doi: 10.3390/jpm15030101.

ABSTRACT

Background/Objectives: While there is mounting scientific evidence supporting the effectiveness of PGx (pharmacogenomics)-guided medical treatment, its implementation into clinical care is still lagging. Stakeholder buy-in, in particular from prescribers, will be key in the implementation efforts. Previous implementation studies have primarily focused on prescriber attitudes or have used hypothetical scenario methodology in a variety of healthcare settings. Real-world studies provide better insight into prescriber experience and needs. In this prospective observational qualitative research study, we report the perspectives of prescribers working in military medical care after a one-year PGx implementation trial. Methods: At the end of the PGx implementation period, thirteen prescribers participated in a semi-structured interview. The interview was designed based on the Technology Acceptance Model and queried their perceptions of effectiveness and ease of use of the PGx innovation. Results: Three main themes emerged from the qualitative data: (1) the knowledge required for PGx testing, (2) the integration of the testing into the existing workflow and (3) the perceived clinical utility of the PGx results. Prescribers had educational and training opportunities prior to the study but still encountered difficulty with the interpretation of the test results. They generally managed well the workflow changes occasioned by the testing. They reported that the clinical value came primarily from an increased confidence in prescribing safe medications and improving the therapeutic alliance with their patients. There was uncertainty about which patient population would most benefit from the testing. Conclusions: Our results lend support to the general ongoing challenges identified in PGx implementation studies conducted in other clinical settings and using other methodologies. They also revealed specific factors that the prescribers found of value and areas that needed improvement to support future implementation efforts.

PMID:40137417 | DOI:10.3390/jpm15030101

J Pers Med. 2025 Feb 28;15(3):93. doi: 10.3390/jpm15030093.

ABSTRACT

Background: Certain CYP2D6 genotypes are linked to a lower efficacy of tamoxifen therapy. This study aimed to observe CYP2D6 polymorphisms and examine the impact of CYP2D6 genotyping among tamoxifen-treated breast cancer patients in Indonesia. Methods: 150 breast cancer participants were recruited. Buccal swab samples were collected; gDNA was extracted and genotyped using the qPCR method. Blood samples were collected, and measurement of tamoxifen metabolite levels was performed using UPLC-MS/MS. Results: 43.3% (n = 65) of participants were IMs. *10 was the most common haplotype (n = 89, 29.7%), followed by *36 (n = 73, 29.7%), making *10/*36 the most common diplotype (n = 34, 22.7%) in this study. The difference in endoxifen levels between the NM and IM-PM groups at baseline was statistically significant (p ≤ 0.001). A dose increase in tamoxifen to 40 mg daily successfully increased endoxifen levels in IMs to a similar level with NMs at baseline (p > 0.05) without exposing IMs to serious side effects. No statistically significant differences were observed between the 20mg group and the 40 mg group on the adjusted OS (p > 0.05) and the adjusted PFS (p > 0.05). Conclusions: Our study observed a considerably high proportion of CYP2D6 IMs. The dose adjustment of tamoxifen was proven to significantly and safely improve the level of endoxifen and survival.

PMID:40137409 | DOI:10.3390/jpm15030093

J Fungi (Basel). 2025 Mar 9;11(3):210. doi: 10.3390/jof11030210.

ABSTRACT

Cystic fibrosis is a severe, inherited, life-limiting disorder, and over half of those living with CF are children. Persistent airway infection and inflammation, resulting in progressive lung function decline, is the hallmark of this disorder. Aspergillus colonization and infection is a well-known complication in people with CF and can evolve in a range of Aspergillus disease phenotypes, including Aspergillus bronchitis, fungal sensitization, and allergic bronchopulmonary aspergillosis (ABPA). Management strategies for children with CF are primarily aimed at preventing lung damage and lung function decline caused by bacterial infections. The role of Aspergillus infections is less understood, especially during childhood, and therefore evidence-based diagnostic and treatment guidelines are lacking. This narrative review summarizes our current understanding of the impact of Aspergillus on the airways of children and young people with CF.

PMID:40137248 | DOI:10.3390/jof11030210

Cells. 2025 Mar 19;14(6):459. doi: 10.3390/cells14060459.

ABSTRACT

Polycystic kidney disease (PKD) is the most common hereditary disorder that disrupts renal function and frequently progresses to end-stage renal disease. Recent advances have elucidated the critical role of primary cilia and ciliary ion channels, including transient receptor potential (TRP) channels, cystic fibrosis transmembrane conductance regulator (CFTR), and polycystin channels, in the pathogenesis of PKD. While some channels primarily function as chloride conductance channels (e.g., CFTR), others primarily regulate calcium (Ca+2) homeostasis. These ion channels are essential for cellular signaling and maintaining the normal kidney architecture. Dysregulation of these pathways due to genetic mutations in PKD1 and PKD2 leads to disrupted Ca+2 and cAMP signaling, aberrant fluid secretion, and uncontrolled cellular proliferation, resulting in tubular cystogenesis. Understanding the molecular mechanisms underlying these dysfunctions has opened the door for innovative therapeutic strategies, including TRPV4 activators, CFTR inhibitors, and calcimimetics, to mitigate cyst growth and preserve renal function. This review summarizes the current knowledge on the roles of ciliary ion channels in PKD pathophysiology, highlights therapeutic interventions targeting these channels, and identifies future research directions for improving patient outcomes.

PMID:40136708 | DOI:10.3390/cells14060459

Cells. 2025 Mar 9;14(6):400. doi: 10.3390/cells14060400.

ABSTRACT

Pulmonary homeostasis can be agitated either by external environmental insults or endogenous factors produced during respiratory/pulmonary diseases. The lungs counter these insults by initiating mechanisms of inflammation as a localized, non-specific first-line defense response. Cytokines are small signaling glycoprotein molecules that control the immune response. They are formed by numerous categories of cell types and induce the movement, growth, differentiation, and death of cells. During respiratory diseases, multiple proinflammatory cytokines play a crucial role in orchestrating chronic inflammation and structural changes in the respiratory tract by recruiting inflammatory cells and maintaining the release of growth factors to maintain inflammation. The issue aggravates when the inflammatory response is exaggerated and/or cytokine production becomes dysregulated. In such instances, unresolving and chronic inflammatory reactions and cytokine production accelerate airway remodeling and maladaptive outcomes. Pro-inflammatory cytokines generate these deleterious consequences through interactions with receptors, which in turn initiate a signal in the cell, triggering a response. The cytokine profile and inflammatory cascade seen in different pulmonary diseases vary and have become fundamental targets for advancement in new therapeutic strategies for lung diseases. There are considerable therapeutic approaches that target cytokine-mediated inflammation in pulmonary diseases; however, blocking specific cytokines may not contribute to clinical benefit. Alternatively, broad-spectrum anti-inflammatory approaches are more likely to be clinically effective. Herein, this comprehensive review of the literature identifies various cytokines (e.g., interleukins, chemokines, and growth factors) involved in pulmonary inflammation and the pathogenesis of respiratory diseases (e.g., asthma, chronic obstructive pulmonary, lung cancer, pneumonia, and pulmonary fibrosis) and investigates targeted therapeutic treatment approaches.

PMID:40136649 | DOI:10.3390/cells14060400

Rev Paul Pediatr. 2025 Mar 24;43:e2024155. doi: 10.1590/1984-0462/2025/43/2024155. eCollection 2025.

ABSTRACT

OBJECTIVE: Determine the immediate effect of forced expiration technique (FET) on the respiratory mechanics of children and adolescents with cystic fibrosis (CF). As a secondary objective, the effect of cough induced by FET was evaluated by comparing respiratory mechanics and lung function between those who coughed and those who did not during the FET.

METHODS: A before-after clinical trial was conducted with children and adolescents with CF aged six to 15 years. Respiratory mechanics parameters were assessed using the impulse oscillometry system (IOS) in three stages: basal IOS, post-huff IOS, and final post-diaphragmatic breathing exercises (DBE) IOS. For the intervention, FET was requested with five low-volume followed by three high-volume huffs, and finally ten DBE repetitions. Coughing occurred randomly and was not previously requested. To investigate whether FET-induced coughing alters oscillometric parameters, the participants were divided into two groups: those who presented with cough (CG) during the protocol and those who did not (NCG).

RESULTS: Forty-three children and adolescents with CF participated in the study (51.2% female), with an average age of 10.44±2.64 years, where forced expiratory value - FEV1=78.51±23.28%, and body mass index - BMI=17.18±2.24 kg/m2. The huffing sequence increased all oscillometric parameters, while DBE repetitions led to an increase in these parameters, without a complete return to baseline values. In terms of coughing, there was no significant difference between the NCG and CG in any of the parameters studied.

CONCLUSIONS: It was observed that, during the FET, diaphragmatic breathing exercises can attenuate the effort exerted by the forced expiratory maneuver on the airways.

PMID:40136119 | DOI:10.1590/1984-0462/2025/43/2024155

J Bacteriol. 2025 Mar 26:e0002925. doi: 10.1128/jb.00029-25. Online ahead of print.

ABSTRACT

Pseudomonas aeruginosa is an opportunistic pathogen that uses several mechanisms to survive in the iron-limiting host environment. The innate immune protein calprotectin (CP) sequesters ferrous iron [Fe(II)], among other divalent transition metal ions, to limit its availability to pathogens. CP levels are increased in individuals with cystic fibrosis (CF), a hereditary disease that leads to chronic pulmonary infection by P. aeruginosa. We previously showed that aerobic CP treatment of P. aeruginosa induces a multi-metal starvation response that alters expression of several virulence properties. However, the CF lung is a hypoxic environment due to the growth of P. aeruginosa in dense biofilms. Here, we report that anaerobic CP treatment of P. aeruginosa induces many processes associated with an aerobic iron starvation response, including decreased phenazine production and increased expression of the PrrF small regulatory RNAs (sRNAs). However, the iron starvation response elicited by CP in anaerobic conditions shows characteristics that are distinct from responses observed in aerobic growth, including a lack of siderophore production and increased induction of genes for the FeoAB Fe(II) and Phu heme uptake systems. Also distinct from aerobic conditions, CP treatment induces expression of genes for predicted manganese transporters MntH1 and MntH2 during anaerobic growth while eliciting a less robust zinc starvation response compared to aerobic conditions. Induction of mntH2 is dependent on the PrrF sRNAs, suggesting a novel example of metal regulatory cross-talk. Thus, anaerobic CP treatment results in a multi-metal starvation response with key distinctions from aerobic conditions, revealing differences in P. aeruginosa metal homeostasis during anaerobic growth.IMPORTANCEIron is critical for most microbial pathogens, and the innate immune system sequesters this metal to limit microbial growth. Pathogens must overcome iron sequestration to survive during infection. For many pathogens, iron homeostasis has primarily been studied in aerobic conditions. Nevertheless, some host environments are hypoxic, including chronic lung infection sites in individuals with cystic fibrosis (CF). Here, we use the innate immune protein calprotectin, which sequesters divalent metal ions including Fe(II), to study the anaerobic iron starvation response of a common CF lung pathogen, Pseudomonas aeruginosa. We report several distinctions of this response during anaerobiosis, highlighting the importance of carefully considering the host environment when investigating the role of nutritional immunity in host-pathogen interactions.

PMID:40135923 | DOI:10.1128/jb.00029-25

J Bacteriol. 2025 Mar 26:e0011623. doi: 10.1128/jb.00116-23. Online ahead of print.

ABSTRACT

Staphylococcus aureus asymptomatically colonizes the nasal cavity and pharynx of up to 60% of the human population and, as an opportunistic pathogen, can breach its normal habitat, resulting in life-threatening infections. S. aureus infections are of additional concern for populations with impaired immune function such as those with cystic fibrosis (CF) or chronic granulomatous disease. Multi-drug resistance is increasingly common in S. aureus infections, creating an urgent need for new antimicrobials or compounds that improve efficacy of currently available antibiotics. S. aureus biofilms, such as those found in the lungs of people with CF and in soft tissue infections, are notoriously recalcitrant to antimicrobial treatment due to the characteristic metabolic differences associated with a sessile mode of growth. In this work, we show that another CF pathogen, Burkholderia cenocepacia, produces one or more secreted compounds that can prevent S. aureus biofilm formation and inhibit existing S. aureus biofilms. The B. cenocepacia-mediated antagonistic activity is restricted to S. aureus species and perhaps some other staphylococci; however, this inhibition does not necessarily extend to other Gram-positive species. This inhibitory activity is due to death of S. aureus through a contact-independent mechanism, potentially mediated through the siderophore pyochelin and perhaps additional compounds. This works paves the way to better understanding of interactions between these two bacterial pathogens.IMPORTANCEStaphylococcus aureus is a major nosocomial pathogen responsible for infecting thousands of people each year. Some strains are becoming increasingly resistant to antimicrobials, and consequently new treatments must be sought. This paper describes the characterization of one or more compounds capable of inhibiting S. aureus biofilm formation and may potentially lead to development of a new therapeutic.

PMID:40135855 | DOI:10.1128/jb.00116-23

J Pers Med. 2025 Mar 20;15(3):121. doi: 10.3390/jpm15030121.

ABSTRACT

(1) Background: Helminth infections are a widespread global health concern, with Ascaris and taeniasis representing two of the most prevalent infestations. Traditional diagnostic methods, such as egg-based microscopy, are fraught with challenges, including subjectivity and low throughput, often leading to misdiagnosis. This study evaluates the efficacy of advanced deep learning models in accurately classifying Ascaris lumbricoides and Taenia saginata eggs from microscopic images, proposing a technologically enhanced approach for diagnostics in clinical settings. (2) Methods: Three state-of-the-art deep learning models, ConvNeXt Tiny, EfficientNet V2 S, and MobileNet V3 S, are considered. A diverse dataset comprising images of Ascaris, Taenia, and uninfected eggs was utilized for training and validating these models by performing multiclass experiments. (3) Results: All models demonstrated high classificatory accuracy, with ConvNeXt Tiny achieving an F1-score of 98.6%, followed by EfficientNet V2 S at 97.5% and MobileNet V3 S at 98.2% in the experiments. These results prove the potential of deep learning in streamlining and improving the diagnostic process for helminthic infections. The application of deep learning models such as ConvNeXt Tiny, EfficientNet V2 S, and MobileNet V3 S shows promise for efficient and accurate helminth egg classification, potentially significantly enhancing the diagnostic workflow. (4) Conclusion: The study demonstrates the feasibility of leveraging advanced computational techniques in parasitology and points towards a future where rapid, objective, and reliable diagnostics are standard.

PMID:40137437 | DOI:10.3390/jpm15030121

J Funct Morphol Kinesiol. 2025 Feb 22;10(1):73. doi: 10.3390/jfmk10010073.

ABSTRACT

BACKGROUND/OBJECTIVES: Falls among the older adult population represent a significant public health concern, often leading to diminished quality of life and serious injuries that escalate healthcare costs, and they may even prove fatal. Accurate fall risk prediction is therefore crucial for implementing timely preventive measures. However, to date, there is no definitive metric to identify individuals with high risk of experiencing a fall. To address this, the present study proposes a novel approach that transforms biomechanical time-series data, derived from gait analysis, into visual representations to facilitate the application of deep learning (DL) methods for fall risk assessment.

METHODS: By leveraging convolutional neural networks (CNNs) and Siamese neural networks (SNNs), the proposed framework effectively addresses the challenges of limited datasets and delivers robust predictive capabilities.

RESULTS: Through the extraction of distinctive gait-related features and the generation of class-discriminative activation maps using Grad-CAM, the random forest (RF) machine learning (ML) model not only achieves commendable accuracy (83.29%) but also enhances explainability.

CONCLUSIONS: Ultimately, this study underscores the potential of advanced computational tools and machine learning algorithms to improve fall risk prediction, reduce healthcare burdens, and promote greater independence and well-being among the older adults.

PMID:40137325 | DOI:10.3390/jfmk10010073

J Imaging. 2025 Mar 20;11(3):91. doi: 10.3390/jimaging11030091.

ABSTRACT

Human activity recognition (HAR) has emerged as a transformative field with widespread applications, leveraging diverse sensor modalities to accurately identify and classify human activities. This paper provides a comprehensive review of HAR techniques, focusing on the integration of sensor-based, vision-based, and hybrid methodologies. It explores the strengths and limitations of commonly used modalities, such as RGB images/videos, depth sensors, motion capture systems, wearable devices, and emerging technologies like radar and Wi-Fi channel state information. The review also discusses traditional machine learning approaches, including supervised and unsupervised learning, alongside cutting-edge advancements in deep learning, such as convolutional and recurrent neural networks, attention mechanisms, and reinforcement learning frameworks. Despite significant progress, HAR still faces critical challenges, including handling environmental variability, ensuring model interpretability, and achieving high recognition accuracy in complex, real-world scenarios. Future research directions emphasise the need for improved multimodal sensor fusion, adaptive and personalised models, and the integration of edge computing for real-time analysis. Additionally, addressing ethical considerations, such as privacy and algorithmic fairness, remains a priority as HAR systems become more pervasive. This study highlights the evolving landscape of HAR and outlines strategies for future advancements that can enhance the reliability and applicability of HAR technologies in diverse domains.

PMID:40137203 | DOI:10.3390/jimaging11030091

J Imaging. 2025 Mar 19;11(3):88. doi: 10.3390/jimaging11030088.

ABSTRACT

The objective of this study is to propose an advanced image enhancement strategy to address the challenge of reducing radiation doses in pediatric renal scintigraphy. Data from a public dynamic renal scintigraphy database were used. Based on noisier images, four denoising neural networks (DnCNN, UDnCNN, DUDnCNN, and AttnGAN) were evaluated. To evaluate the quality of the noise reduction, with minimal detail loss, the kidney signal-to-noise ratio (SNR) and multiscale structural similarity (MS-SSIM) were used. Although all the networks reduced noise, UDnCNN achieved the best balance between SNR and MS-SSIM, leading to the most notable improvements in image quality. In clinical practice, 100% of the acquired data are summed to produce the final image. To simulate the dose reduction, we summed only 50%, simulating a proportional decrease in radiation. The proposed deep-learning approach for image enhancement ensured that half of all the frames acquired may yield results that are comparable to those of the complete dataset, suggesting that it is feasible to reduce patients' exposure to radiation. This study demonstrates that the neural networks evaluated can markedly improve the renal scintigraphic image quality, facilitating high-quality imaging with lower radiation doses, which will benefit the pediatric population considerably.

PMID:40137200 | DOI:10.3390/jimaging11030088

J Imaging. 2025 Mar 18;11(3):85. doi: 10.3390/jimaging11030085.

ABSTRACT

Semantic segmentation in deep learning is a crucial area of research within computer vision, aimed at assigning specific labels to each pixel in an image. The segmentation of crops, plants, and weeds has significantly advanced the application of deep learning in precision agriculture, leading to the development of sophisticated architectures based on convolutional neural networks (CNNs). This study proposes a segmentation algorithm for identifying plants and weeds using broadband multispectral images. In the first part of this algorithm, we utilize the PIF-Net model for feature extraction and fusion. The resulting feature map is then employed to enhance an optimized U-Net model for semantic segmentation within a broadband system. Our investigation focuses specifically on scenes from the CAVIAR dataset of multispectral images. The proposed algorithm has enabled us to effectively capture complex details while regulating the learning process, achieving an impressive overall accuracy of 98.2%. The results demonstrate that our approach to semantic segmentation and the differentiation between plants and weeds yields accurate and compelling outcomes.

PMID:40137197 | DOI:10.3390/jimaging11030085

J Imaging. 2025 Mar 18;11(3):84. doi: 10.3390/jimaging11030084.

ABSTRACT

Background: This study aims to objectively evaluate the overall quality of colonoscopies using a specially trained deep learning-based semantic segmentation neural network. This represents a modern and valuable approach for the analysis of colonoscopy frames. Methods: We collected thousands of colonoscopy frames extracted from a set of video colonoscopy files. A color-based image processing method was used to extract color features from specific regions of each colonoscopy frame, namely, the intestinal mucosa, residues, artifacts, and lumen. With these features, we automatically annotated all the colonoscopy frames and then selected the best of them to train a semantic segmentation network. This trained network was used to classify the four region types in a different set of test colonoscopy frames and extract pixel statistics that are relevant to quality evaluation. The test colonoscopies were also evaluated by colonoscopy experts using the Boston scale. Results: The deep learning semantic segmentation method obtained good results, in terms of classifying the four key regions in colonoscopy frames, and produced pixel statistics that are efficient in terms of objective quality assessment. The Spearman correlation results were as follows: BBPS vs. pixel scores: 0.69; BBPS vs. mucosa pixel percentage: 0.63; BBPS vs. residue pixel percentage: -0.47; BBPS vs. Artifact Pixel Percentage: -0.65. The agreement analysis using Cohen's Kappa yielded a value of 0.28. The colonoscopy evaluation based on the extracted pixel statistics showed a fair level of compatibility with the experts' evaluations. Conclusions: Our proposed deep learning semantic segmentation approach is shown to be a promising tool for evaluating the overall quality of colonoscopies and goes beyond the Boston Bowel Preparation Scale in terms of assessing colonoscopy quality. In particular, while the Boston scale focuses solely on the amount of residual content, our method can identify and quantify the percentage of colonic mucosa, residues, and artifacts, providing a more comprehensive and objective evaluation.

PMID:40137196 | DOI:10.3390/jimaging11030084

J Imaging. 2025 Mar 3;11(3):76. doi: 10.3390/jimaging11030076.

ABSTRACT

COVID-19 can cause acute infectious diseases of the respiratory system, and may probably lead to heart damage, which will seriously threaten human health. Electrocardiograms (ECGs) have the advantages of being low cost, non-invasive, and radiation free, and is widely used for evaluating heart health status. In this work, a lightweight deep learning network named GM-CBAM-ResNet is proposed for diagnosing COVID-19 based on ECG images. GM-CBAM-ResNet is constructed by replacing the convolution module with the Ghost module (GM) and adding the convolutional block attention module (CBAM) in the residual module of ResNet. To reveal the superiority of GM-CBAM-ResNet, the other three methods (ResNet, GM-ResNet, and CBAM-ResNet) are also analyzed from the following aspects: model performance, complexity, and interpretability. The model performance is evaluated by using the open 'ECG Images dataset of Cardiac and COVID-19 Patients'. The complexity is reflected by comparing the number of model parameters. The interpretability is analyzed by utilizing Gradient-weighted Class Activation Mapping (Grad-CAM). Parameter statistics indicate that, on the basis of ResNet19, the number of model parameters of GM-CBAM-ResNet19 is reduced by 45.4%. Experimental results show that, under less model complexity, GM-CBAM-ResNet19 improves the diagnostic accuracy by approximately 5% in comparison with ResNet19. Additionally, the interpretability analysis shows that CBAM can suppress the interference of grid backgrounds and ensure higher diagnostic accuracy under lower model complexity. This work provides a lightweight solution for the rapid and accurate diagnosing of COVD-19 based on ECG images, which holds significant practical deployment value.

PMID:40137188 | DOI:10.3390/jimaging11030076

J Imaging. 2025 Feb 26;11(3):72. doi: 10.3390/jimaging11030072.

ABSTRACT

In an era where security concerns are ever-increasing, the need for advanced technology to detect visible and concealed weapons has become critical. This paper introduces a novel two-stage method for concealed handgun detection, leveraging thermal imaging and deep learning, offering a potential real-world solution for law enforcement and surveillance applications. The approach first detects potential firearms at the frame level and subsequently verifies their association with a detected person, significantly reducing false positives and false negatives. Alarms are triggered only under specific conditions to ensure accurate and reliable detection, with precautionary alerts raised if no person is detected but a firearm is identified. Key contributions include a lightweight algorithm optimized for low-end embedded devices, making it suitable for wearable and mobile applications, and the creation of a tailored thermal dataset for controlled concealment scenarios. The system is implemented on a chest-worn Android smartphone with a miniature thermal camera, enabling hands-free operation. Experimental results validate the method's effectiveness, achieving an mAP@50-95 of 64.52% on our dataset, improving state-of-the-art methods. By reducing false negatives and improving reliability, this study offers a scalable, practical solution for security applications.

PMID:40137184 | DOI:10.3390/jimaging11030072

Metabolites. 2025 Mar 3;15(3):174. doi: 10.3390/metabo15030174.

ABSTRACT

Background/Objectives: Metabolomics has recently emerged as a key tool in the biological sciences, offering insights into metabolic pathways and processes. Over the last decade, network-based machine learning approaches have gained significant popularity and application across various fields. While several studies have utilized metabolomics profiles for sample classification, many network-based machine learning approaches remain unexplored for metabolomic-based classification tasks. This study aims to compare the performance of various network-based machine learning approaches, including recently developed methods, in metabolomics-based classification. Methods: A standard data preprocessing procedure was applied to 17 metabolomic datasets, and Bayesian neural network (BNN), convolutional neural network (CNN), feedforward neural network (FNN), Kolmogorov-Arnold network (KAN), and spiking neural network (SNN) were evaluated on each dataset. The datasets varied widely in size, mass spectrometry method, and response variable. Results: With respect to AUC on test data, BNN, CNN, FNN, KAN, and SNN were the top-performing models in 4, 1, 5, 3, and 4 of the 17 datasets, respectively. Regarding F1-score, the top-performing models were BNN (3 datasets), CNN (3 datasets), FNN (4 datasets), KAN (4 datasets), and SNN (3 datasets). For accuracy, BNN, CNN, FNN, KAN, and SNN performed best in 4, 1, 4, 4, and 4 datasets, respectively. Conclusions: No network-based modeling approach consistently outperformed others across the metrics of AUC, F1-score, or accuracy. Our results indicate that while no single network-based modeling approach is superior for metabolomics-based classification tasks, BNN, KAN, and SNN may be underappreciated and underutilized relative to the more commonly used CNN and FNN.

PMID:40137139 | DOI:10.3390/metabo15030174

Biomimetics (Basel). 2025 Mar 20;10(3):191. doi: 10.3390/biomimetics10030191.

ABSTRACT

Chemical oxygen demand (COD) is a critical parameter employed to assess the level of organic pollution in water. Accurate COD detection is essential for effective environmental monitoring and water quality assessment. Ultraviolet-visible (UV-Vis) spectroscopy has become a widely applied method for COD detection due to its convenience and the absence of the need for chemical reagents. This non-destructive and reagent-free approach offers a rapid and reliable means of analyzing water. Recently, deep learning has emerged as a powerful tool for automating the process of spectral feature extraction and improving COD prediction accuracy. In this paper, we propose a novel multi-scale one-dimensional convolutional neural network (MS-1D-CNN) fusion model designed specifically for spectral feature extraction and COD prediction. The architecture of the proposed model involves inputting raw UV-Vis spectra into three parallel sub-1D-CNNs, which independently process the data. The outputs from the final convolution and pooling layers of each sub-CNN are then fused into a single layer, capturing a rich set of spectral features. This fused output is subsequently passed through a Flatten layer followed by fully connected layers to predict the COD value. Experimental results demonstrate the effectiveness of the proposed method, as it was compared with three traditional methods and three deep learning methods on the same dataset. The MS-1D-CNN model showed a significant improvement in the accuracy of COD prediction, highlighting its potential for more reliable and efficient water quality monitoring.

PMID:40136845 | DOI:10.3390/biomimetics10030191