Exploring bronchial anatomical structures, disease progression and mechanism of action

As they play a central role in the functioning of the respiratory system, assessing airways is crucial in lung disease evaluation. Any obstruction, inflammation or narrowing of the airways can impede airflow and cause respiratory problems. Abnormalities such as narrowing (as in asthma), mucus build up (as in chronic bronchitis), or structural changes (as in COPD) can be indicative of certain lung conditions and their corresponding changes over time.

The ability of AI-based algorithms to analyze vast amounts of imaging data with unprecedented precision and accuracy, completely revolutionizes the way we can assess airways to better understand the underlying disease process, conduct personalized interventions and monitor treatment effect.

Fully automatic robust assessment of airway disease

Thirona's Bronchi suite of measurements contains a wide range of robust algorithms, extensively validated in a multitude of external studies, and trained to handle a variety of different patient cohorts and input scan characteristics. Following Thirona's proven 3-step methodology (Identification-Localization-Quantification), we are able to identify subtle structural changes with high precision, accuracy and sensitivity, allowing to monitor disease progression.

The AI-enabled analyses are based on anatomical biomarkers such as bronchial count, wall thickness, lumen diameter, bronchus-artery ratios, mucus plug count and many more. They provide robust quantitative assessment for multiple diseases, as well as the corresponding abnormalities like bronchial dropout, bronchiectasis, mucus impaction, etc.

Identification


Accurate identification and segmentation of all airways visible on CT to identify the bronchi is a crucial first step in our AI-based analysis. Visualization of the bronchial tree and extracting metrics like branch count, tree length, mucus obstruction as well as locating the adjacent arteries, forms the basis for all following steps.

Localization


While segmentation makes it possible to visualize the bronchial tree, labeling of the exact location within the tree allows us to repeatedly quantify the same locations, both longitudinally and cross-sectionally. Well-conducted localization is pivotal for reproducibility and consistency of results.

Quantification


Our quantification algorithms allow for measuring bronchial structures up to the small bronchi, with high accuracy and precision, even below the scan resolution. Robust quantification of bronchial measurements can be provided on multiple levels, from full lungs to lobar and (sub)segmental analysis, as well as for individual branches and generations.

Enabling imaging-guided lung interventions and pre-/post-treatment assessment

Based on the precisely analyzed and quantified chest CT images, in the context of specific treatment or intervention, Thirona's AI platform can generate a multitude of very detailed visualizations of the bronchi tree and related pathologies.

3D visual representation of outcomes, along with the quantitative data report, delivers unprecedented evidence for development of emerging therapies in treatment of airway disease as well as novel insights into lung anatomical structures, enabling next generation surgical and bronchoscopic interventions.

From patient selection, identification of possible treatment pathways and pre-operative navigation planning to intra-operative image analysis, and all the way to objective quantification of treatment efficacy and post-treatment monitoring of disease progression - Thirona's artificial intelligence-based technology fuels innovations transforming clinical paths in pulmonary precision medicine.

Precision medicine applications

The Bronchi suite of analysis offers unique capabilities for determining treatment efficacy for muco-obstructive and other respiratory diseases such as COPD, cystic fibrosis, severe asthma and bronchiectasis. The extended, and robust analysis  allows for more accurate patient phenotyping and for sensitive monitoring of longitudinal changes over time.

Accurate segmentation and visualization of bronchial anatomical structures, along with objective quantification of lung abnormalities, can also provide critical insights for bronchoscopic and surgical interventions for the treatment of obstructive diseases (COPD, severe asthma), diseases involving a great deal of mucus obstruction and bronchiectasis (CF, IPF) or lung cancer.

Thirona's major proprietary bronchial measurements

Bronchus-Artery Analysis

Bronchus-Artery (BA) analysis is one of the fundamental measurements of Thirona’s AI-based lung quantification platform. BA conducts an automatic quantification of the entire bronchi tree, while normalizing bronchus dimensions against the adjacent artery as the reference structure within the scan. Providing robust measurements, based on ratios between bronchial wall thickness, inner and outer bronchi diameters, and the artery diameters, LungQ™ BA consistently shows highly reliable performance in cystic fibrosis and a variety of diseases.

Mucus Plugs Quantification

This algorithm can conduct fully automatic quantification of airway-occluding mucus plugs from CT, throughout the entire bronchi tree. It is the most sensitive way to analyse non-occlusive mucus accumulation: in combination with the Bronchus-Artery analysis, as well as with VERA, which allows for the assessment of small bronchi not visible on CT. We can provide a comprehensive assessment of mucus impaction and distribution, monitoring changes over time with great sensitivity and precision.

VERA - Ventilation Estimation Small Airway Analysis

VERA algorithm combines inspiratory and expiratory scans to automatically detect areas of hypo-ventilated and/or hypo-perfused lung volume, allowing for the indirect assessment of obstructive changes within the small non-visible airways. The power of this algorithm involves the ability to estimate gas exchange on the alveolar level, a resolution that no CT scan is capable of capturing.

Validation studies and publications


Ongoing external validation studies play a vital role in assessing the ability of our algorithms to perform consistently on diverse patient populations. See below a selection of research studies and publications that speak toward the robustness and clinical applicability of LungQ™ Bronchi analysis.

Cystic Fibrosis

    • Fully Automatic Assessment of Bronchus-Artery Dimensions and Ratios, Mucus Plugs, and Low Attenuation Regions in Cystic Fibrosis Patients Before and After Elexacaftor-Tezacaftor-Ivacaftor | May 2024 | American Journal of Respiratory and Critical Care Medicine
    • Azithromycin reduces bronchial wall thickening in infants with cystic fibrosis | Apr 2024 | Journal of Cystic Fibrosis
    • The clinical impact of Lumacaftor-Ivacaftor on structural lung disease and lung function in children aged 6–11 with cystic fibrosis in a real-world setting | Aug 2023 | Respiratory Research
    • Automatic Bronchus and Artery Analysis on Chest Computed Tomography to Evaluate the Effect of Inhaled Hypertonic Saline in Children Aged 3-6 Years With Cystic Fibrosis in a Randomized Clinical Trial | May 2023 | Journal of Cystic Fibrosis
    • Sensitive automated airway-artery method to monitor progression of CF airway disease | Nov 2021 | European Respiratory Journal

Asthma

    • Children with severe asthma have substantial structural airway changes on computed tomography | Aug 2023 | ERJ Open Research
    • Automatic and Manual Quantification of Small Airways Disease on Chest CT of Children With Severe Asthma | May 2023 | American Thoracic Society
    • Small airways targeted treatment with smart nebulizer technology could improve severe asthma in children: a retrospective analysis | Nov 2021 | Journal of Asthma

COPD

    • Cross-sectional Analysis of Automated Mucus Plug Quantification and Airway Wall Thickness in COPD Using AI-based Detection | May 2024 | American Journal of Respiratory and Critical Care Medicine
    • Fully Automated Mucus Plug Quantification on Chest CTs and Its Correlation With All-cause Mortality | May 2024 | American Journal of Respiratory and Critical Care Medicine
    • Automated Quantification of Mucus Plugs in Smokers Using the LungQ Algorithm: A COPDGene Cohort Study | JAMA
    • Dysanapsis is differentially related to lung function trajectories with distinct structural and functional patterns in COPD and variable risk for adverse outcomes | Jan 2024 | The Lancet
    • Baseline characteristics from a 3-year longitudinal study to phenotype subjects with COPD: the FOOTPRINTS study | Nov 2023 | Respiratory Research

Normal Subjects

    • Fully Automatic Quantitative Analysis of Airway-artery Dimensions and Ratios of Normal Chest CT Scans From Infancy Into Adulthood | May 2023 | American Thoracic Society

Interventions

    • Bronchial Rheoplasty Reduces Mucus Plugging in Patients With Chronic Bronchitis | May 2024 | American Journal of Respiratory and Critical Care Medicine
    • Development of a quantitative CT-based diaphragm tool in severe COPD patients treated with one-way endobronchial valves | May 2023 | American Thoracic Society
    • High-Resolution Computed Tomography-approximated Perfusion Is Comparable to Nuclear Perfusion Imaging in Severe Chronic Obstructive Pulmonary Disease | May 2023 | American Thoracic Society
    • Virtual Reality and Artificial Intelligence for 3-dimensional Planning of Lung Segmentectomies | Jun 2021 | JTCVS Techniques
    • Endobronchial valves for severe emphysema | Jun 2019 | European Respiratory Review

Related blogs


Ready to apply AI image analysis to get better outcomes faster?


Talk to our representatives to explore how we can jointly define an image analysis program precisely tailored to your project requirements.

Let's explore the partnership opportunities!