The role of lung ultrasound and ultrasound elastography in diagnosis of interstitial lung diseases.

Introduction: Ultrasound elastography (US-E) is a novel, tissue stiffness-sensitive imaging method. We aimed to investigate whether lung ultrasound (US) and US-E can play a role in diagnosing interstitial lung diseases (ILDs) in which lung elasticity is affected due to fibrosis. Materials and Methods: A prospective cohort study. Patients with ILD were defined as ''ILD group'' and with other pulmonary diseases as ''control group". All subjects were examined and compared by lung US in B and elastography modes. Besides, the relationship between ultrasonography and high-resolution computerized tomography (HRCT) and chest X-ray findings was evaluated. Result: A total of 109 patients, 55 in ILD and 54 in the control group, with a mean age of 62 ± 14 years, were included. A positive correlation was found between the Warrick score (calculated from HRCT to determine the severity of ILD) and the number of B-lines (discrete vertical reverberation artifacts, indicating interstitial lung syndrome) in lung US (p= 0.001, r= 0.550) in the ILD group. In US-E, blue color (meaning more rigid tissue) dominated in the ILD group, and green color (indicating medium tissue stiffness) dominated in the control group (p= 0.001). Lung US diagnosed the ILD with 69% accuracy, 80% sensitivity, and 60% specificity compared to HRCT. Combined with chest X-ray, diagnostic accuracy was 74%, sensitivity 60%, and specificity 89%. Conclusions: Although lung US and US-E are not superior to gold standard HRCT in diagnosing ILDs, they can still be accepted as promising, novel, noninvasive tools, especially when combined with chest X-rays. Their role still needs to be clarified with further studies.

Can FDG-PET/CT imaging be used to predict decline in quality of life in interstitial lung disease? A prospective study of the relationship between FDG uptake and quality of life in a UK outpatient setting.

BACKGROUND: 18Fluorine-fluorodeoxyglucose (18F-FDG) positron emission tomography (PET) CT imaging has been used in many inflammatory and infectious conditions to differentiate areas of increased metabolic activity. FDG uptake differs between areas of normal lung parenchyma and interstitial lung disease (ILD). OBJECTIVES: In this study, we investigated whether FDG-PET/CT parameters were associated with a change in the quality of life (QoL) in patients with ILD over 4 years of follow-up. METHODS: Patients underwent PET-CT imaging at diagnosis and were followed up with annual QoL assessment using the St George's Respiratory Questionnaire (SGRQ) until death or 4 years of follow-up. Maximum standard uptake value (SUVmax) and Tissue-to-Background Ratio (TBR) were assessed against SGRQ overall and subscale scores. RESULTS: 193 patients (94 patients in the idiopathic pulmonary fibrosis (IPF) subgroup and 99 patients in the non-IPF subgroup) underwent baseline FDG-PET/CT imaging and QoL assessment. Weak-to-moderate correlation was observed between baseline SUVmax and SGRQ scores in both ILD subgroups. No relationship was observed between baseline SUVmax or TBR and change in SGRQ scores over 4 years of follow-up. In the IPF subgroup, surviving patients reported a decline in QoL at 4 years post diagnosis whereas an improvement in QoL was seen in surviving patients with non-IPF ILD. CONCLUSIONS: Weak-to-moderate positive correlation between baseline SUVmax and SGRQ scores was observed in both ILD subgroups (IPF:rs=0.187, p=0.047, non-IPF: rs=0.320, p=0.001). However, baseline SUVmax and TBR were not associated with change in QoL in patients with IPF and non-IPF ILD over 4 years of follow-up. At 4 years post diagnosis, surviving patients with IPF reported declining QoL whereas improvement was seen in patients with ILD who did not have IPF.

Screening value of lung ultrasound and pleural shear wave elastography in connective tissue disease-related interstitial lung disease: a preliminary study.

OBJECTIVE: To explore the diagnostic value of lung ultrasound (LUS) and pleural shear wave elastography (SWE) for connective tissue disease-interstitial lung disease (CTD-ILD). METHODS: We selected 104 patients diagnosed with connective tissue disease (CTD) at our hospital. All patients underwent LUS, SWE, and high-resolution computed tomography (HRCT). With HRCT as the imaging gold standard for diagnosis, patients were categorized into CTD-ILD and CTD-non-ILD groups. We employed paired chi-square tests to compare the diagnostic differences between HRCT and LUS for ILD. Receiver operating characteristic (ROC) curves were used to assess the diagnostic value of pleural SWE for ILD. Correlation analysis was performed between pleural elasticity values and lung ultrasound scores. RESULTS: The sensitivity, specificity, positive likelihood ratio, and negative likelihood ratio of LUS for diagnosing CTD-ILD were 93.3%, 86.2%, 6.761, and 0.078, respectively. There was no statistically significant difference in the results between HRCT and LUS (P = 1.000), with a kappa value of 0.720 (P < 0.001). There was a statistically significant difference in the pleural elasticity in the bilateral lower back region between the case and control groups (P < 0.001). The area under the receiver operating characteristic (ROC) curve (AUC) for pleural SWE in diagnosing CTD-ILD was 0.685. In CTD-ILD patients, there was no significant correlation between pleural elasticity values and LUS scores (P > 0.05). CONCLUSION: The LUS can serve as an important imaging method for screening for CTD-ILD and assessing the severity of the disease. However, pleural SWE has been shown to demonstrate lower diagnostic efficacy for CTD-ILD, and its ability to assess disease severity is limited.

Prognostic and predictive value of interstitial lung abnormalities and EGFR mutation status in patients with non-small cell lung cancer.

BACKGROUND: To determine the predictive value of interstitial lung abnormalities (ILA) for epidermal growth factor receptor (EGFR) mutation status and assess the prognostic significance of EGFR and ILA in patients with non-small cell lung cancer (NSCLC). METHODS: We reviewed 797 consecutive patients with a histologically proven diagnosis of primary NSCLC from January 2013 to October 2018. Of these, 109 patients with NSCLC were found to have concomitant ILA. Multivariate logistic regression analysis was used to identify the significant clinical and computed tomography (CT) findings in predicting EGFR mutations. Cox proportional hazard models were used to identify significant prognostic factors. RESULTS: EGFR mutations were identified in 22 of 109 tumors (20.2%). Multivariate analysis showed that the models incorporating clinical, tumor CT and ILA CT features yielded areas under the receiver operating characteristic curve (AUC) values of 0.749, 0.838, and 0.849, respectively. When combining the three models, the independent predictive factors for EGFR mutations were non-fibrotic ILA, female sex, and small tumor size, with an AUC value of 0.920 (95% confidence interval[CI]: 0.861-0.978, p < 0.001). In the multivariate Cox model, EGFR mutations (hazard ratio = 0.169, 95% CI = 0.042-0.675, p = 0.012; 692 days vs. 301 days) were independently associated with extended overall survival compared to the wild-type. CONCLUSION: Non-fibrotic ILA independently predicts the presence of EGFR mutations, and the presence of EGFR mutations rather than non-fibrotic ILA serves as an independent good prognostic factor for patients with NSCLC.

Interstitial Lung Abnormalities: Current Understanding.

Interstitial lung abnormalities (ILAs) are incidental findings on computed tomography scans, characterized by nondependent abnormalities affecting more than 5% of any lung zone. They are associated with factors such as age, smoking, genetic variants, worsened clinical outcomes, and increased mortality. Risk stratification based on clinical and radiological features of ILAs is crucial in clinical practice, particularly for identifying cases at high risk of progression to pulmonary fibrosis. Traction bronchiectasis/bronchiolectasis index has emerged as a promising imaging biomarker for prognostic risk stratification in ILAs. These findings suggest a spectrum of fibrosing interstitial lung diseases, encompassing from ILAs to pulmonary fibrosis.

Rheumatoid arthritis disease activity significantly impacts on the severity of interstitial lung disease.

OBJECTIVES: Rheumatoid arthritis (RA) related interstitial lung disease (ILD) impacts on the treatment strategy and its prognosis in patients with RA. However, the relationship between RA disease activity and the severity of comorbid ILD has not been fully investigated. This study aimed to investigate the impact of RA disease activity on the severity of comorbid ILD in detail based on currently established visual scoring method along with physiological severity. METHODS: Consecutive patients with RA visiting to our Rheumatology Centre between December 2020 and December 2023 were analysed. The radiological severity of ILD was evaluated by averaging the extent of the combined lesion of ground glass opacity, reticulation and honeycombing in 5% increments in six representative high-resolution computed tomography slices ranging from 0% (no involvement) to 100% (all lung fields affected) according to Goh and Walsh's method. Associations between the radiological and physiological severity of ILD and patients' features were investigated using linear regression analysis. RESULTS: Among 124 patients (32 men, 92 women), the median age was 70 years, and the median disease duration was 2.92 years. Radiological severity of ILD was 0% (without ILD) in 107 (86.2%), ILD with extent < 10% in nine (7.2%), ILD with extent ≥10% and < 20% in three (2.4%), ILD with extent ≥20% in five (4.0%). Both disease activity score (DAS)28-erythrocyte sedimentation rate (ESR) (standardized coefficient = 0.199, P = 0.03) and rheumatoid factor titre (standardized coefficient = 0.247, P = 0.01) were significantly associated with the radiological quantitative severity of ILD in multivariate analysis adjusted for age, sex, disease duration, smoking status and anti-citrullinated peptide antibody titre. DAS28-ESR was significantly associated with forced vital capacity% predicted (standardized coefficient = -0.230, P = 0.047). CONCLUSIONS: Disease activity of RA was significantly associated with the severity of RA-ILD both radiologically and physiologically.

Longitudinal assessment of interstitial lung abnormalities on CT in patients with COPD using artificial intelligence-based segmentation: a prospective observational study.

BACKGROUND: Interstitial lung abnormalities (ILAs) on CT may affect the clinical outcomes in patients with chronic obstructive pulmonary disease (COPD), but their quantification remains unestablished. This study examined whether artificial intelligence (AI)-based segmentation could be applied to identify ILAs using two COPD cohorts. METHODS: ILAs were diagnosed visually based on the Fleischner Society definition. Using an AI-based method, ground-glass opacities, reticulations, and honeycombing were segmented, and their volumes were summed to obtain the percentage ratio of interstitial lung disease-associated volume to total lung volume (ILDvol%). The optimal ILDvol% threshold for ILA detection was determined in cross-sectional data of the discovery and validation cohorts. The 5-year longitudinal changes in ILDvol% were calculated in discovery cohort patients who underwent baseline and follow-up CT scans. RESULTS: ILAs were found in 32 (14%) and 15 (10%) patients with COPD in the discovery (n = 234) and validation (n = 153) cohorts, respectively. ILDvol% was higher in patients with ILAs than in those without ILA in both cohorts. The optimal ILDvol% threshold in the discovery cohort was 1.203%, and good sensitivity and specificity (93.3% and 76.3%) were confirmed in the validation cohort. 124 patients took follow-up CT scan during 5 ± 1 years. 8 out of 124 patients (7%) developed ILAs. In a multivariable model, an increase in ILDvol% was associated with ILA development after adjusting for age, sex, BMI, and smoking exposure. CONCLUSION: AI-based CT quantification of ILDvol% may be a reproducible method for identifying and monitoring ILAs in patients with COPD.

Prevalence and prognostic meaning of interstitial lung abnormalities in remote CT scans of patients with interstitial lung disease treated with antifibrotic therapy.

OBJECTIVES: To describe the prevalence and characteristics of interstitial lung abnormalities (ILA) in CT scans performed prior to the initiation of antifibrotics in a series of patients with interstitial lung disease (ILD), and to identify characteristics apparent on early CT scans that could help to predict outcomes. METHODS: We conducted a retrospective observational study. The original cohort consisted of 101 patients diagnosed with ILD and treated with antifibrotics in a tertiary hospital. Patients were included if they had a thoracic CT scan performed at least one year before initiation of therapy. They were classified radiologically in three groups: without ILA, with radiological ILA and extensive abnormalities. ILA were classified as subpleural fibrotic, subpleural non-fibrotic and non-subpleural. The initial scan and the latest CT scan performed before treatment were read for assessing progression. The relationship between CT findings of fibrosis and the radiological progression rate and mortality were analyzed. RESULTS: We included 50 patients. Only 1 (2%) had a normal CT scan, 25 (50%) had extensive alterations and 24 (48%) had radiological criteria for ILA, a median of 98.2 months before initiation of antifibrotics, of them 18 (75%) had a subpleural fibrotic pattern. Significant bronchiectasis and obvious honeycombing in the lower zones were associated with shorter survival (p = 0.04). Obvious honeycombing in the lower zones was also significantly (p < 0.05) associated with a faster progression rate. CONCLUSIONS: Fibrotic ILAs are frequent in remote scans of patients with clinically relevant ILD, long before they require antifibrotics. Findings of traction bronchiectasis and honeycombing in the earliest scans, even in asymptomatic patients, are related to mortality and progression later on.

A mouse model of progressive lung fibrosis with cutaneous involvement induced by a combination of oropharyngeal and osmotic minipump bleomycin delivery.

Systemic sclerosis (SSc) with interstitial lung disease (SSc-ILD) lacks curative pharmacological treatments, thus necessitating effective animal models for candidate drug discovery. Existing bleomycin (BLM)-induced SSc-ILD mouse models feature spatially limited pulmonary fibrosis, spontaneously resolving after 28 days. Here, we present an alternative BLM administration approach in female C57BL/6 mice, combining oropharyngeal aspiration (OA) and subcutaneous mini-pump delivery (pump) of BLM to induce a sustained and more persistent fibrosis, while retaining stable skin fibrosis. A dose-finding study was performed with BLM administered as 10 µg (OA) +80 mg/kg (pump) (10 + 80), 10 + 100, and 15 + 100. Forty-two days after OA, micro-computed tomography (micro-CT) imaging and histomorphometric analyses showed that the 10 + 100 and 15 + 100 treatments induced significant alterations in lung micro-CT-derived readouts, Ashcroft score, and more severe fibrosis grades compared with saline controls. In addition, a marked reduction in hypodermal thickness was observed in the 15 + 100 group. A time-course characterization of the BLM 15 + 100 treatment at days 28, 35, and 42, including longitudinal micro-CT imaging, revealed progressing alterations in lung parameters. Lung histology highlighted a sustained fibrosis accompanied by a reduction in hypodermis thickness throughout the explored time-window, with a time-dependent increase in fibrotic biomarkers detected by immunofluorescence analysis. BLM-induced alterations were partly mitigated by Nintedanib treatment. Our optimized BLM delivery approach leads to extensive and persistent lung fibrotic lesions coupled with cutaneous fibrotic alterations: it thus represents a significant advance compared with current preclinical models of BLM-induced SSc-ILD.NEW & NOTEWORTHY This study introduces an innovative approach to enhance the overall performance of the mouse bleomycin (BLM)-induced model for systemic sclerosis with interstitial lung disease (SSc-ILD). By combining oropharyngeal aspiration and subcutaneous mini-pump delivery of BLM, our improved model leads to sustained lung fibrosis and stable skin fibrosis in female C57BL/6 mice. The optimized 15 + 100 treatment results in extensive and persistent lung fibrotic lesions and thus represents a significant improvement over existing preclinical models of BLM-induced SSc-ILD.

Automated CT quantification of interstitial lung abnormality in patients with resectable stage I non-small cell lung cancer: Prognostic significance.

BACKGROUND: In patients with non-small cell lung cancer (NSCLC), interstitial lung abnormalities (ILA) have been linked to mortality and can be identified on computed tomography (CT) scans. In the present study we aimed to evaluate the predictive value of automatically quantified ILA based on the Fleischner Society definition in patients with stage I NSCLC. METHODS: We retrospectively reviewed 948 patients with pathological stage I NSCLC who underwent pulmonary resection between April 2009 and October 2022. A commercially available deep learning-based automated quantification program for ILA was used to evaluate the preoperative CT data. The Fleischner Society definition, quantitative results, and interdisciplinary discussion led to the division of patients into normal and ILA groups. The sum of the fibrotic and nonfibrotic ILA components constituted the total ILA component and more than 5%. RESULTS: Of the 948 patients with stage I NSCLC, 99 (10.4%) patients had ILA. Shorter overall survival and recurrence-free survival was associated with the presence of ILA. After controlling for confounding variables, the presence of ILA remained significant for increased risk of death (hazard ratio [HR] = 3.09; 95% confidence interval [CI]: 1.91-5.00; p < 0.001) and the presence of ILA remained significant for increased recurrence (HR = 1.96; 95% CI: 1.16-3.30; p = 0.012). CONCLUSIONS: The automated CT quantification of ILA, based on the Fleischner Society definition, was significantly linked to poorer survival and recurrence in patients with stage I NSCLC.

The Value of Ultrasound for Detecting and Following Subclinical Interstitial Lung Disease in Systemic Sclerosis.

BACKGROUND: Interstitial lung disease (ILD) is a complication in patients with systemic sclerosis (SSc). Accurate strategies to identify its presence in early phases are essential. We conducted the study aiming to determine the validity of ultrasound (US) in detecting subclinical ILD in SSc, and to ascertain its potential in determining the disease progression. METHODS: 133 patients without respiratory symptoms and 133 healthy controls were included. Borg scale, Rodnan skin score (RSS), auscultation, chest radiographs, and respiratory function tests (RFT) were performed. A rheumatologist performed the lung US. High-resolution CT (HRCT) was also performed. The patients were followed every 12 weeks for 48 weeks. RESULTS: A total of 79 of 133 patients (59.4%) showed US signs of ILD in contrast to healthy controls (4.8%) (p = 0.0001). Anti-centromere antibodies (p = 0.005) and RSS (p = 0.004) showed an association with ILD. A positive correlation was demonstrated between the US and HRCT findings (p = 0.001). The sensitivity and specificity of US in detecting ILD were 91.2% and 88.6%, respectively. In the follow-up, a total of 30 patients out of 79 (37.9%) who demonstrated US signs of ILD at baseline, showed changes in the ILD score by US. CONCLUSIONS: US showed a high prevalence of subclinical ILD in SSc patients. It proved to be a valid, reliable, and feasible tool to detect ILD in SSc and to monitor disease progression.

Clinical utility, reproducibility, and radiologist acceptance of ILD-RADS.

PURPOSE: To investigate the clinical utility, reproducibility, and radiologists' acceptance of the Interstitial Lung Disease Imaging-Reporting and Data System (ILD-RADS). METHOD: In this single-institutional retrospective study, three radiologists independently reviewed the chest high-resolution CT (HRCT) scans of 111 consecutive patients diagnosed with ILDs. They assessed the HRCT pulmonary features using the ILD-RADS template and assigned an ILD-RADS category (1-4) to each scan based on the identified imaging pattern. Patients were classified into idiopathic pulmonary fibrosis (IPF) (n = 14) and non-IPF ILD (n = 97) groups based on clinical diagnoses determined by multidisciplinary discussion. Association between ILD-RADS categories and clinical diagnoses was assessed using the Chi-square test for trend. Reproducibility was evaluated using kappa (k) scores, and radiologists' acceptance of the ILD-RADS was evaluated with a questionnaire. RESULTS: We found a significant association between the ILD-RADS categories and patients' clinical diagnoses (P ≤ 0.0001) for the three readers, with a trend toward increased assignment of ILD-RADS-1 to IPF patients (50 %-57.1 %), and ILD-RADS-4 to non-IPF patients (46.4 %-49.5 %). The ILD-RADS categories showed excellent intra-reader agreement (k = 0.873) and moderate inter-reader agreement (k = 0.440). ILD-RADS-1 and -4 categories showed the highest inter-reader agreement (k = 0.681 and 0.481, respectively). Radiologists gave a positive response to using the ILD-RADS in daily practice. CONCLUSIONS: The clinical utility of the ILD-RADS was demonstrated by the significant association between the ILD-RADS categories and patients' clinical diagnoses, particularly the ILD-RADS-1 and -4 categories. Excellent intra-reader and moderate inter-reader reproducibility was observed. ILD-RADS has the potential to be widely accepted for standardized HRCT reporting among radiologists.

[Progressive pulmonary Fibrosis]. / Progressive Pulmonale Fibrose PPF.

INTRODUCTION: Progressive pulmonary Fibrosis Abstract: Cough and dyspnea on excertion are common and early symptoms of interstitial lung diseases (ILD). Thoracic imaging (particularly computed tomography) detects such lung structural alterations early in the disease course. Knowledge of these diseases and their management is necessary in the daily business. The term "progressive pulmonary fibrosis" subsumes a heterogene group of interstitial lung diseases with a similar course of progressive fibrosis. The management of these diseases should be discussed interdisciplinary, similar to the management of the Idiopathic pulmonary fibrosis (IPF). Antifibrotic drugs are new therapeutic options.

Association of Acute Respiratory Disease Events with Quantitative Interstitial Abnormality Progression at CT in Individuals with a History of Smoking.

Background Acute respiratory disease (ARD) events are often thought to be airway-disease related, but some may be related to quantitative interstitial abnormalities (QIAs), which are subtle parenchymal abnormalities on CT scans associated with morbidity and mortality in individuals with a smoking history. Purpose To determine whether QIA progression at CT is associated with ARD and severe ARD events in individuals with a history of smoking. Materials and Methods This secondary analysis of a prospective study included individuals with a 10 pack-years or greater smoking history recruited from multiple centers between November 2007 and July 2017. QIA progression was assessed between baseline (visit 1) and 5-year follow-up (visit 2) chest CT scans. Episodes of ARD were defined as increased cough or dyspnea lasting 48 hours and requiring antibiotics or corticosteroids, whereas severe ARD episodes were those requiring an emergency room visit or hospitalization. Episodes were recorded via questionnaires completed every 3 to 6 months. Multivariable logistic regression and zero-inflated negative binomial regression models adjusted for comorbidities (eg, emphysema, small airway disease) were used to assess the association between QIA progression and episodes between visits 1 and 2 (intercurrent) and after visit 2 (subsequent). Results A total of 3972 participants (mean age at baseline, 60.7 years ± 8.6 [SD]; 2120 [53.4%] women) were included. Annual percentage QIA progression was associated with increased odds of one or more intercurrent (odds ratio [OR] = 1.29 [95% CI: 1.06, 1.56]; P = .01) and subsequent (OR = 1.26 [95% CI: 1.05, 1.52]; P = .02) severe ARD events. Participants in the highest quartile of QIA progression (≥1.2%) had more frequent intercurrent ARD (incidence rate ratio [IRR] = 1.46 [95% CI: 1.14, 1.86]; P = .003) and severe ARD (IRR = 1.79 [95% CI: 1.18, 2.73]; P = .006) events than those in the lowest quartile (≤-1.7%). Conclusion QIA progression was independently associated with higher odds of severe ARD events during and after radiographic progression, with higher frequency of intercurrent severe events in those with faster progression. Clinical trial registration no. NCT00608764 © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Little in this issue.

[Diffuse cystic lung disease]. / Cystische Lungenerkrankungen.

INTRODUCTION: Diffuse cystic lung disease (DCLD) represents a heterogeneous group of conditions, typically characterized by the presence of multiple thin-walled, predominantly round parenchymal lucencies. The increased accessibility of computed tomography (CT) underscores the growing relevance of a relatively rare group of diseases as more clinicians are confronted with the presence of multiple lung cysts on the chest CT scan. Although the etiology of these conditions is very diverse, the focus of the differential diagnosis revolves around four primary causative factors - Lymphangioleiomyomatosis (LAM), Pulmonary Langerhanscell histiocytosis (PLCH), Birt-Hogg-Dubé (BHD) and lymphoid interstitial pneumonia (LIP). Achieving an accurate diagnosis poses a challenge and typically necessitates lung biopsies; however, it is crucial for ensuring proper management.

Pattern classification of interstitial lung diseases from computed tomography images using a ResNet-based network with a split-transform-merge strategy and split attention.

In patients with interstitial lung disease (ILD), accurate pattern assessment from their computed tomography (CT) images could help track lung abnormalities and evaluate treatment efficacy. Based on excellent image classification performance, convolutional neural networks (CNNs) have been massively investigated for classifying and labeling pathological patterns in the CT images of ILD patients. However, previous studies rarely considered the three-dimensional (3D) structure of the pathological patterns of ILD and used two-dimensional network input. In addition, ResNet-based networks such as SE-ResNet and ResNeXt with high classification performance have not been used for pattern classification of ILD. This study proposed a SE-ResNeXt-SA-18 for classifying pathological patterns of ILD. The SE-ResNeXt-SA-18 integrated the multipath design of the ResNeXt and the feature weighting of the squeeze-and-excitation network with split attention. The classification performance of the SE-ResNeXt-SA-18 was compared with the ResNet-18 and SE-ResNeXt-18. The influence of the input patch size on classification performance was also evaluated. Results show that the classification accuracy was increased with the increase of the patch size. With a 32 × 32 × 16 input, the SE-ResNeXt-SA-18 presented the highest performance with average accuracy, sensitivity, and specificity of 0.991, 0.979, and 0.994. High-weight regions in the class activation maps of the SE-ResNeXt-SA-18 also matched the specific pattern features. In comparison, the performance of the SE-ResNeXt-SA-18 is superior to the previously reported CNNs in classifying the ILD patterns. We concluded that the SE-ResNeXt-SA-18 could help track or monitor the progress of ILD through accuracy pattern classification.