Diagnostic Applications Research Articles

Quantitative and Computational Spinal Imaging in Neurodegenerative Conditions and Acquired Spinal Disorders: Academic Advances and Clinical Prospects

Introduction: Quantitative spinal cord imaging has facilitated the objective appraisal of spinal cord pathology in a range of neurological conditions both in the academic and clinical setting. Diverse methodological approaches have been implemented, encompassing a range of morphometric, diffusivity, susceptibility, magnetization transfer, and spectroscopy techniques. Advances have been fueled both by new MRI platforms and acquisition protocols as well as novel analysis pipelines. The quantitative evaluation of specific spinal tracts and grey matter indices has the potential to be used in diagnostic and monitoring applications. The comprehensive characterization of spinal disease burden in pre-symptomatic cohorts, in carriers of specific genetic mutations, and in conditions primarily associated with cerebral disease, has contributed important academic insights. Methods: A narrative review was conducted to examine the clinical and academic role of quantitative spinal cord imaging in a range of neurodegenerative and acquired spinal cord disorders, including hereditary spastic paraparesis, hereditary ataxias, motor neuron diseases, Huntington’s disease, and post-infectious or vascular disorders. Results: The clinical utility of specific methods, sample size considerations, academic role of spinal imaging, key radiological findings, and relevant clinical correlates are presented in each disease group. Conclusions: Quantitative spinal cord imaging studies have demonstrated the feasibility to reliably appraise structural, microstructural, diffusivity, and metabolic spinal cord alterations. Despite the notable academic advances, novel acquisition protocols and analysis pipelines are yet to be implemented in the clinical setting.

Surface-modified CMOS biosensors

Biosensors translate biological events into electronic signals that quantify biological processes. They are increasingly used in in vitro diagnostics applications that leverage their ability to process small sample volumes. One recent trend has been to integrate biosensors with complementary metal-oxide-semiconductor (CMOS) chips to provide enhanced miniaturization, parallel sensing, and low power consumption at a low cost. CMOS-enabled biosensors are used in monitoring DNA hybridization, enzymatic reactions, and cell proliferation, to name a few applications. This paper explores the materials and processes used in emerging CMOS biosensors. We discuss subtractive and additive processes for creating electrodes for electrochemical sensing applications. We discuss functionalization techniques for creating bioelectronic interfaces that allow molecular events to be transduced into the electrical domain using a plurality of modalities that are readily provided by CMOS chips. Example modalities featured are optical sensing, electrochemical detection, electrical detection, magnetic sensing, and mechanical sensing.

LINE-1 cfDNA Methylation as an Emerging Biomarker in Solid Cancers

Epigenetic dysregulation is a hallmark of many human malignancies, with DNA methylation being a primary mechanism influencing gene expression and maintaining genomic stability. Genome-wide hypomethylation, characteristic of many cancers, is partly attributed to the demethylation of repetitive elements, including LINE-1, a prevalent non-LTR retrotransposon. The methylation status of LINE-1 is closely associated with overall genomic methylation levels in tumors. cfDNA comprises extracellular DNA fragments found in bodily fluids such as plasma, serum, and urine, offering a dynamic snapshot of the genetic and epigenetic landscape of tumors. This real-time sampling provides a minimally invasive avenue for cancer diagnostics, prognostics, and monitoring. The methylation status of LINE-1 in cfDNA has emerged as a promising biomarker, with several studies highlighting its potential in diagnosing and predicting outcomes in cancer patients. Recent research also suggests that cfDNA-based LINE-1 methylation analysis could serve as a valuable tool in evaluating the efficacy of cancer therapies, including immunotherapy. The growing clinical significance of cfDNA calls for a closer examination of its components, particularly repetitive elements like LINE-1. Despite their importance, the role of LINE-1 elements in cfDNA has not been thoroughly gauged. We aim to address this gap by reviewing the current literature on LINE-1 cfDNA assays, focusing on their potential applications in diagnostics and disease monitoring.

A Cross-talk between Nanomedicines and Cardiac Complications: Comprehensive View.

Cardiovascular Diseases (CVDs) are the leading cause of global morbidity and mortality, necessitating innovative approaches for both therapeutics and diagnostics. Nanoscience has emerged as a promising frontier in addressing the complexities of CVDs. This study aims to explorethe interaction of CVDs and Nanomedicine (NMs), focusing on applications in therapeutics and diagnostics. In the realm of therapeutics, nanosized drug delivery systems exhibit unique advantages, such as enhanced drug bioavailability, targeted delivery, and controlled release. NMs platform, including liposomes, nanoparticles, and carriers, allows the precise drug targeting to the affected cardiovascular tissues with minimum adverse effects and maximum therapeutic efficacy. Moreover, nanomaterial (NM) enables the integration of multifunctional components, such as therapeutic agents and target ligands, into a single system for comprehensive CVD management. Diagnostic fronts of NMs offer innovative solutions for early detection and monitoring of CVDs. Nanoparticles and nanosensors enable highly sensitive and specific detection of Cardiac biomarkers, providing valuable insights into a disease state, its progression, therapeutic outputs, etc. Further, nano-based technology via imaging modalities offers high high-resolution imaging, aiding in the vascularization of cardiovascular structures and abnormalities. Nanotechnology-based imaging modalities offer high-resolution imaging and aid in the visualization of cardiovascular structures and abnormalities. The cross-talk of CVDs and NMs holds tremendous potential for revolutionizing cardiovascular healthcare by providing targeted and efficient therapeutic interventions, as well as sensitive and early detection for the improvement of patient health if integrated with Artificial Intelligence (AI).

The quality and detection limits of mitochondrial heteroplasmy by long read nanopore sequencing

This study evaluates long-read and short-read sequencing for mitochondrial DNA (mtDNA) heteroplasmy detection. 592,315 bootstrapped datasets generated from two single-nucleotide mismatched ultra-deep sequenced mtDNA samples were used to assess basecalling error and accuracy, limit of heteroplasmy detection, and heteroplasmy detection across various coverage depths. Results showed high Phred scores of data with GC-rich sequence bias for long reads. Limit of detection of 12% heteroplasmy was identified, showing strong correlation (R2 ≥ 0.955) with expected heteroplasmy but underreporting tendency of high-level variants. Nanopore sequencing shows potential for direct applicability in mitochondrial diseases diagnostics, but stringent validation processes to ensure diagnostic result quality are required.

Bridging organ transcriptomics for advancing multiple organ toxicity assessment with a generative AI approach

Translational research in toxicology has significantly benefited from transcriptomic profiling, particularly in drug safety. However, its application has predominantly focused on limited organs, notably the liver, due to resource constraints. This paper presents TransTox, an innovative AI model using a generative adversarial network (GAN) method to facilitate the bidirectional translation of transcriptomic profiles between the liver and kidney under drug treatment. TransTox demonstrates robust performance, validated across independent datasets and laboratories. First, the concordance between real experimental data and synthetic data generated by TransTox was demonstrated in characterizing toxicity mechanisms compared to real experimental settings. Second, TransTox proved valuable in gene expression predictive models, where synthetic data could be used to develop gene expression predictive models or serve as “digital twins” for diagnostic applications. The TransTox approach holds the potential for multi-organ toxicity assessment with AI and advancing the field of precision toxicology.

Multiplatform metabolomic interlaboratory study of a whole human stool candidate reference material from omnivore and vegan donors.

Human metabolomics has made significant strides in understanding metabolic changes and their implications for human health, with promising applications in diagnostics and treatment, particularly regarding the gut microbiome. However, progress is hampered by issues with data comparability and reproducibility across studies, limiting the translation of these discoveries into practical applications. This study aims to evaluate the fit-for-purpose of a suite of human stool samples as potential candidate reference materials (RMs) and assess the state of the field regarding harmonizing gut metabolomics measurements. An interlaboratory study was conducted with 18 participating institutions. The study allowed for the use of preferred analytical techniques, including liquid chromatography-mass spectrometry (LC-MS), gas chromatography-mass spectrometry (GC-MS), and nuclear magnetic resonance (NMR). Different laboratories used various methods and analytical platforms to identify the metabolites present in human stool RM samples. The study found a 40% to 70% recurrence in the reported top 20 most abundant metabolites across the four materials. In the full annotation list, the percentage of metabolites reported multiple times after nomenclature standardization was 36% (LC-MS), 58% (GC-MS) and 76% (NMR). Out of 9,300 unique metabolites, only 37 were reported across all three measurement techniques. This collaborative exercise emphasized the broad chemical survey possible with multi-technique approaches. Community engagement is essential for the evaluation and characterization of common materials designed to facilitate comparability and ensure data quality underscoring the value of determining current practices, challenges, and progress of a field through interlaboratory studies.

Improving deep learning U-Net++ by discrete wavelet and attention gate mechanisms for effective pathological lung segmentation in chest X-ray imaging.

Since its introduction in 2015, the U-Net architecture used in Deep Learning has played a crucial role in medical imaging. Recognized for its ability to accurately discriminate small structures, the U-Net has received more than 2600 citations in academic literature, which motivated continuous enhancements to its architecture. In hospitals, chest radiography is the primary diagnostic method for pulmonary disorders, however, accurate lung segmentation in chest X-ray images remains a challenging task, primarily due to the significant variations in lung shapes and the presence of intense opacities caused by various diseases. This article introduces a new approach for the segmentation of lung X-ray images. Traditional max-pooling operations, commonly employed in conventional U-Net++ models, were replaced with the discrete wavelet transform (DWT), offering a more accurate down-sampling technique that potentially captures detailed features of lung structures. Additionally, we used attention gate (AG) mechanisms that enable the model to focus on specific regions in the input image, which improves the accuracy of the segmentation process. When compared with current techniques like Atrous Convolutions, Improved FCN, Improved SegNet, U-Net, and U-Net++, our method (U-Net++-DWT) showed remarkable efficacy, particularly on the Japanese Society of Radiological Technology dataset, achieving an accuracy of 99.1%, specificity of 98.9%, sensitivity of 97.8%, Dice Coefficient of 97.2%, and Jaccard Index of 96.3%. Its performance on the Montgomery County dataset further demonstrated its consistent effectiveness. Moreover, when applied to additional datasets of Chest X-ray Masks and Labels and COVID-19, our method maintained high performance levels, achieving up to 99.3% accuracy, thereby underscoring its adaptability and potential for broad applications in medical imaging diagnostics.

TIA-UNet: Transformer-Enhanced Deep Learning for Adolescent Idiopathic Scoliosis Spinal X-Ray Image Segmentation

Abstract Adolescent Idiopathic Scoliosis (AIS) is a common spinal deformity where precise diagnosis is crucial for developing effective treatment strategies. Traditional manual X-ray image analysis is time-consuming and highly dependent on the operator’s expertise, thus constraining diagnostic efficiency and accuracy.This study aimed to develop an automated thoracolumbar spine segmentation method utilizing deep learning to enhance the efficiency and accuracy of AIS diagnosis. We introduced TIA-UNet, an innovative network architecture that combines Convolutional Neural Networks (CNN) with Transformer models.By integrating the IR Block and DA Block, TIA-UNet was optimized for feature extraction and multi-scale information fusion. The model underwent training and validation using our established Adolescent Scoliosis Medical Dataset (ASMD).TIA-UNet attained a Dice similarity coefficient of 90.02%, a Mean Intersection over Union (MIoU) of 81.96%, and a Hausdorff Distance (HD) of 4.09, significantly surpassing current state-of-the-art methods such as UNet and TransUNet. Moreover, TIA-UNet exhibited superior computational efficiency regarding parameter count, inference time, and floatingpoint operations (FLOPs).As an automated medical image segmentation algorithm, TIA-UNet enhanced segmentation accuracy while maintaining high computational efficiency, demonstrating significant potential for clinical diagnostic applications. This study provides compelling evidence supporting the utilization of deep learning
techniques in medical image analysis.

IGFBP-2 in Critical Illness: A Prognostic Marker in the Growth Hormone/Insulin-like Growth Factor Axis

Critical illness (CI) triggers complex disruptions in the growth hormone (GH)/insulin-like growth factor (IGF) axis, significantly affecting the dynamics of insulin-like growth-factor-binding proteins (IGFBPs). Among these, IGFBP-2 shows a sustained elevation during CI, which inversely correlates with serum levels of IGF-1, IGFBP-3, and the acid-labile subunit (ALS). Although IGFBP-2 does not directly interact with ALS, it may influence the availability of IGFs by competing with other IGFBPs for binding to IGF-1 and IGF-2. Research suggests that this persistent elevation of IGFBP-2 is largely driven by cytokine activity during CI, reflecting an adaptive response rather than a direct result of GH/IGF axis dysregulation. The clinical importance of IGFBP-2 is emphasized by its correlation with disease severity in conditions like sepsis and coronavirus disease 2019 (COVID-19), where its levels are markedly elevated compared to healthy controls and are similar to those observed in sepsis from various causes. Beyond its role in endocrine regulation, IGFBP-2 appears to play a part in metabolic and inflammatory pathways. Elevated IGFBP-2 levels have been linked to increased mortality and longer hospital stays, indicating its potential utility as a prognostic marker. Furthermore, measuring plasma IGFBP-2 may have other diagnostic applications, aiding in the assessment of CI when traditional biomarkers are inconclusive.

A Review on Multiple-Ocular Disease Detection Methodology using ML and DL Techniques

The rapid advancement in machine learning (ML) and deep learning (DL) techniques has significantly impacted the detection and diagnosis of ocular diseases, which are critical for preserving vision and overall eye health. This review aims to explore the various ML and DL methodologies applied to the detection of multiple ocular diseases, highlighting their effectiveness, limitations, and areas for improvement. The motivation behind this review stems from the increasing prevalence of ocular diseases and the need for efficient, accurate diagnostic tools. Despite the promising results of existing techniques, limitations such as data variability, the need for extensive training data, and computational resource requirements persist. The objective is to synthesize current methodologies and propose enhancements, particularly through the integration of attention mechanisms in convolutional neural networks (CNNs). This review identifies gaps in current research and suggests directions for future work to enhance diagnostic accuracy and clinical applicability.

Optimizing classification models for medical image diagnosis: a comparative analysis on multi-class datasets

The surge in machine learning (ML) and artificial intelligence has revolutionized medical diagnosis, utilizing data from chest ct-scans, COVID-19, lung cancer, brain tumor, and alzheimer parkinson diseases. However, the intricate nature of medical data necessitates robust classification models. This study compares support vector machine (SVM), naïve Bayes, k-nearest neighbors (K-NN), artificial neural networks (ANN), and stochastic gradient descent on multi-class medical datasets, employing data collection, Canny image segmentation, hu moment feature extraction, and oversampling/under-sampling for data balancing. Classification algorithms are assessed via 5-fold cross-validation for accuracy, precision, recall, and F-measure. Results indicate variable model performance depending on datasets and sampling strategies. SVM, K-NN, ANN, and SGD demonstrate superior performance on specific datasets, achieving accuracies between 0.49 to 0.57. Conversely, naïve Bayes exhibits limitations, achieving precision levels of 0.46 to 0.47 on certain datasets. The efficacy of oversampling and under-sampling techniques in improving classification accuracy varies inconsistently. These findings aid medical practitioners and researchers in selecting suitable models for diagnostic applications.

Enzyme Activity Inhibition of α-Amylase Using Molecularly Imprinted Polymer (MIP) Hydrogel Microparticles.

Molecularly imprinted polymers (MIPs) are a class of synthetic recognition materials that offer a cost-effective and robust alternative to antibodies. While MIPs have found predominant use in biosensing and diagnostic applications, their potential for alternative uses, such as enzyme inhibition, remains unexplored. In this work, we synthesized a range of acrylamide-based hydrogel MIP microparticles (35 μm) specific for the recognition of α-amylase. These MIPs also showed good selectivity toward the target protein with over 96% binding of the target protein, compared with the control nonimprinted polymer (NIP) counterparts. Specificity of the MIPs was determined with the binding of nontarget proteins, trypsin, human serum albumin (HSA), and bovine serum albumin (BSA). The MIPs were further evaluated for their ability to inhibit α-amylase enzymatic activity, showing a significant decrease in activity. These findings highlight the potential of MIPs as enzyme inhibitors, suggesting an innovative application beyond their conventional use.

Vibration Signal Analysis for Intelligent Rotating Machinery Diagnosis and Prognosis: A Comprehensive Systematic Literature Review

Many industrial processes, from manufacturing to food processing, incorporate rotating elements as principal components in their production chain. Failure of these components often leads to costly downtime and potential safety risks, further emphasizing the importance of monitoring their health state. Vibration signal analysis is now a common approach for this purpose, as it provides useful information related to the dynamic behavior of machines. This research aimed to conduct a comprehensive examination of the current methodologies employed in the stages of vibration signal analysis, which encompass preprocessing, processing, and post-processing phases, ultimately leading to the application of Artificial Intelligence-based diagnostics and prognostics. An extensive search was conducted in various databases, including ScienceDirect, IEEE, MDPI, Springer, and Google Scholar, from 2020 to early 2024 following the PRISMA guidelines. Articles that aligned with at least one of the targeted topics cited above and provided unique methods and explicit results qualified for retention, while those that were redundant or did not meet the established inclusion criteria were excluded. Subsequently, 270 articles were selected from an initial pool of 338. The review results highlighted several deficiencies in the preprocessing step and the experimental validation, with implementation rates of 15.41% and 10.15%, respectively, in the selected prototype studies. Examination of the processing phase revealed that time scale decomposition methods have become essential for accurate analysis of vibration signals, as they facilitate the extraction of complex information that remains obscured in the original, undecomposed signals. Combining such methods with time–frequency analysis methods was shown to be an ideal combination for information extraction. In the context of fault detection, support vector machines (SVMs), convolutional neural networks (CNNs), Long Short-Term Memory (LSTM) networks, k-nearest neighbors (KNN), and random forests have been identified as the five most frequently employed algorithms. Meanwhile, transformer-based models are emerging as a promising venue for the prediction of RUL values, along with data transformation. Given the conclusions drawn, future researchers are urged to investigate the interpretability and integration of the diagnosis and prognosis models developed with the aim of applying them in real-time industrial contexts. Furthermore, there is a need for experimental studies to disclose the preprocessing details for datasets and the operational conditions of the machinery, thereby improving the data reproducibility. Another area that warrants further investigation is differentiation of the various types of fault information present in vibration signals obtained from bearings, as the defect information from the overall system is embedded within these signals.

A BDD-Au/SPCE-based electrochemical DNA aptasensor for selective and sensitive detection of theophylline

The proposed aptasensor using a DNA aptamer for the detection of theophylline (THEO) is being reported for the first time, as previous methods predominantly employed RNA aptamers. DNA aptamers have attracted significant attention due to their advantages over antibodies, including greater stability and cost-effectiveness, making them ideal for various diagnostic applications as biosensors. THEO, a drug commonly used in the treatment of respiratory diseases such as asthma, requires precise monitoring in the bloodstream due to its narrow therapeutic index. In this paper, we present the first optimization of THEO detection using a DNA aptamer-based sensor (aptasensor), utilizing a composite boron-doped diamond/Gold (BDD-Au) on screen-printed carbon electrodes. Thiolated DNA aptamers were immobilized on the BDD-Au surface, where they selectively targeted THEO in biological samples. Optimization of the aptasensor was achieved using a Box-Behnken design, determining optimal conditions of 3.0 μM for aptamer concentration, 90 min for aptamer immobilization time, and 1.0 mM for tris(2-carboxyethyl)phosphine concentration. The resulting DNA aptasensor demonstrated excellent selectivity for THEO, with a linear detection range from 1 to 1.0 mM and a detection limit of 52 nM. The sensor maintained stability for up to four weeks and exhibited high recovery rates when tested in blood serum, highlighting its potential for clinical monitoring of THEO levels.

A matrix-centered view of mass spectrometry platform innovation for volatilome research

Volatile organic compounds (VOCs) are carbon-containing molecules with high vapor pressure and low water solubility that are released from biotic and abiotic matrices. Because they are in the gaseous phase, these compounds tend to remain undetected when using conventional metabolomic profiling methods. Despite this omission, efforts to profile VOCs can provide useful information related to metabolic status and identify potential signaling pathways or toxicological impacts in natural or engineered environments. Over the past several decades mass spectrometry (MS) platform innovation has instigated new opportunities for VOC detection from previously intractable matrices. In parallel, volatilome research linking VOC profiles to other forms of multi-omic information (DNA, RNA, protein, and other metabolites) has gained prominence in resolving genotype/phenotype relationships at different levels of biological organization. This review explores both on-line and off-line methods used in VOC profiling with MS from different matrices. On-line methods involve direct sample injection into the MS platform without any prior compound separation, while off-line methods involve chromatographic separation prior to sample injection and analyte detection. Attention is given to the technical evolution of platforms needed for increasingly resolved VOC profiles, tracing technical progress over time with particular emphasis on emerging microbiome and diagnostic applications.

Photonic-Metamaterial-Based, Near-Field-Enhanced Biosensing Approach for Early Detection of Lung and Ovarian Cancer

Early detection of lung and ovarian cancers relies heavily on identifying tumor biomarkers, but current methods require large blood samples and complex genetic testing. This study presents a novel photonic-metamaterial-based biosensing approach that leverages near-field radiative enhancement to detect cancer biomarkers (CA 125, CEA, and CYFRA 21-1) with high sensitivity. By utilizing structured photonic metamaterials, we optimize specific wavelengths to identify these biomarkers in interstitial fluid, which can be easily collected via minimally invasive microneedle arrays. Integrating near-field interactions with wavelength-selective metamaterials amplifies the thermal response at the nanoscale, allowing for the detection of deficient concentrations of biomarkers. This photonic metamaterial technique provides a faster, more accessible, and affordable alternative to conventional blood-based methods, significantly improving early detection and monitoring of cancer. Ultimately, this approach offers a transformative tool for clinical and research applications in cancer diagnostics.

Experimental Determination of the Temperature‐Dependent Broadening Effect of Water Vapour on the S‐Branch Raman Linewidths of Nitrogen

ABSTRACTThis work reports the S‐branch Raman linewidths of N2 broadened by H2O vapour at temperatures 870–1900 K determined using the picosecond time‐resolved pure rotational coherent anti‐Stokes Raman scattering (RCARS) approach. The coherent dephasing time constant of nitrogen in a binary mixture was determined from the time trace fitting of the RCARS signal. Raman linewidths of N2 for four different binary mixture compositions up to 20% H2O were obtained for each temperature and the N2‐H2O broadening coefficient was determined. The decay of nitrogen broadened by water vapour is faster than that of self‐broadened nitrogen for all temperatures leading to a higher Raman linewidth. Respectively, the S‐branch Raman linewidths of N2‐H2O deviate significantly in magnitude and slope from the Q‐branch Raman linewidth calculated by the modified energy gap (MEG) law. Therefore, taking into account the broadening effect of water molecules in nitrogen has the potential to significantly improve thermometry and species concentration determination. The determined S‐branch Raman linewidths and broadening coefficient will therefore find application in combustion and reactive flow diagnostics.

Application of active piezoresistive cantilevers in high-eigenmode surface imaging

Abstract One of the most important limitations of the atomic force microscopy (AFM) is scanning speed, whose high values are required for contemporary high-resolution, long-range diagnostic applications. The measurement bandwidth of an AFM depends on several factors, but usually results from the time constant of the oscillating cantilever, which is correlated with its resonance frequency and quality factor. We propose a method to overcome this problem by performing the surface measurements when the cantilever is vibrating in higher eigenmodes. In this paper we demonstrate the application of active piezoresistive cantilevers operating in this mode. The active piezoresistive cantilever comprises a piezoresistive deflection sensor, a deflection actuator and a nanotip. It is a complete micro-electro-mechanical system (MEMS), ensuring the highest reliability of cantilever vibration control and detection. Higher eigenmode operations are usually difficult to implement as they usually result in lower deflection and lower sensitivity of the probe vibration deflection. Here we present an experimental modification of the structure of an active piezoresistive cantilever using focused ion beam (FIB) machining that mitigates both weaknesses. This has enabled the cantilever to scan the surface at a scanning rate of 10 lines/s with a maximum speed of 500 μm/s and a data acquisition rate of 10 kS/s, when the probe is vibrating at 380 kHz in the second eigenmode. We also describe a traceable calibration routine (based on analysis of the response of the piezoresistive detector, the output of the HeNe interferometer and precise control of the deflection actuator), together with the cantilever modification process and the development of the measurement setup. We show measurement results of dedicated calibration samples and silicon carbide crystal lattice references.

Porcine Nose Atrophy Assessed by Automatic Imaging and Detection of Bordetella bronchiseptica and Other Respiratory Pathogens in Lung and Nose

The nasal mucosa is a crucial filtering organ to prevent attachment and invasion of pathogens. To assess nasal health in relation to lung health, transverse cross sections of the nasal turbinates of 121 pigs suffering from respiratory disease and sent for diagnostic necropsy were scored visually and by an artificial intelligence (AI) medical diagnostic application (AI DIAGNOS), resulting in a high correlation of both scores (p < 0.001). Nasal samples of the diseased pigs were examined only for Bordetella (B.) bronchiseptica (PCR and bacteriological culture) and Pasteurella (P.) multocida (bacteriological culture). All pigs showed various degrees of inflammatory lung tissue alterations, and 35.5% of the pigs had atrophy of the nasal turbinates with no relation to detection rates of B. bronchiseptica (54.5%) and P. multocida (29.0%) in the nose. All P. multocida strains from nose samples were negative for the toxA gene so non-progressive atrophic rhinitis was diagnosed. Pigs positive for B. bronchiseptica in the nose were more often positive for B. bronchiseptica in the lung (p < 0.001) and for other bacterial species in the lower respiratory tract (p = 0.005). The new diagnostic application for scoring cross sections of nasal turbinates is a valuable tool for a fast and reproducible diagnostic.