Radiomics for the Prediction of Response to Antifibrotic Treatment in Patients with Idiopathic Pulmonary Fibrosis: A Pilot Study

Succinate promotes pulmonary fibrosis through GPR91 and predicts death in idiopathic pulmonary fibrosis

Idiopathic pulmonary fibrosis (IPF) is believed to be associated with a notable disruption of cellular energy metabolism. By detecting the changes of energy metabolites in the serum of patients with pulmonary fibrosis, we aimed to investigate the diagnostic and prognostic value of energy metabolites in IPF, and further elucidated the mechanism of their involvement in pulmonary fibrosis. Through metabolomics research, it was discovered that the TCA cycle intermediates changed dramatically in IPF patients. In another validation cohort of 55 patients with IPF compared to 19 healthy controls, it was found that succinate, an intermediate product of TCA cycle, has diagnostic and prognostic value in IPF. The cut-off levels of serum succinate were 98.36 μM for distinguishing IPF from healthy controls (sensitivity, 83.64%; specificity, 63.16%; likelihood ratio, 2.27, respectively). Moreover, a high serum succinate level was independently associated with higher rates of disease progression (OR 13.087, 95%CI (2.819-60.761)) and mortality (HR 3.418, 95% CI (1.308-8.927)). In addition, accumulation of succinate and increased expression of the succinate receptor GPR91 were found in both IPF patients and BLM mouse models of pulmonary fibrosis. Reducing succinate accumulation in BLM mice alleviated pulmonary fibrosis and 21d mortality, while exogenous administration of succinate can aggravate pulmonary fibrosis in BLM mice. Furthermore, GPR91 deficiency protected against lung fibrosis caused by BLM. In vitro, succinate promoted the activation of lung fibroblasts by activating ERK pathway through GPR91. In summary, succinate is a promising biomarker for diagnosis and prognosis of IPF. The accumulation of succinate may promote fibroblast activation through GPR91 and pulmonary fibrosis.

Fusing Diverse Decision Rules in 3D-Radiomics for Assisting Diagnosis of Lung Adenocarcinoma

This study aimed to develop an interpretable diagnostic model for subtyping of pulmonary adenocarcinoma, including minimally invasive adenocarcinoma (MIA), adenocarcinoma in situ (AIS), and invasive adenocarcinoma (IAC), by integrating 3D-radiomic features and clinical data. Data from multiple hospitals were collected, and 10 key features were selected from 1600 3D radiomic signatures and 11 radiological features. Diverse decision rules were extracted using ensemble learning methods (gradient boosting, random forest, and AdaBoost), fused, ranked, and selected via RuleFit and SHAP to construct a rule-based diagnostic model. The model's performance was evaluated using AUC, precision, accuracy, recall, and F1-score and compared with other models. The rule-based diagnostic model exhibited excellent performance in the training, testing, and validation cohorts, with AUC values of 0.9621, 0.9529, and 0.8953, respectively. This model outperformed counterparts relying solely on selected features and previous research models. Specifically, the AUC values for the previous research models in the three cohorts were 0.851, 0.893, and 0.836. It is noteworthy that individual models employing GBDT, random forest, and AdaBoost demonstrated AUC values of 0.9391, 0.8681, and 0.9449 in the training cohort, 0.9093, 0.8722, and 0.9363 in the testing cohort, and 0.8440, 0.8640, and 0.8750 in the validation cohort, respectively. These results highlight the superiority of the rule-based diagnostic model in the assessment of lung adenocarcinoma subtypes, while also providing insights into the performance of individual models. Integrating diverse decision rules enhanced the accuracy and interpretability of the diagnostic model for lung adenocarcinoma subtypes. This approach bridges the gap between complex predictive models and clinical utility, offering valuable support to healthcare professionals and patients.

Multidisciplinary Approach to the Diagnosis of Idiopathic Interstitial Pneumonias: Focus on the Pathologist's Key Role

Idiopathic Interstitial Pneumonias (IIPs) are a heterogeneous group of the broader category of Interstitial Lung Diseases (ILDs), pathologically characterized by the distortion of lung parenchyma by interstitial inflammation and/or fibrosis. The American Thoracic Society (ATS)/European Respiratory Society (ERS) international multidisciplinary consensus classification of the IIPs was published in 2002 and then updated in 2013, with the authors emphasizing the need for a multidisciplinary approach to the diagnosis of IIPs. The histological evaluation of IIPs is challenging, and different types of IIPs are classically associated with specific histopathological patterns. However, morphological overlaps can be observed, and the same histopathological features can be seen in totally different clinical settings. Therefore, the pathologist's aim is to recognize the pathologic-morphologic pattern of disease in this clinical setting, and only after multi-disciplinary evaluation, if there is concordance between clinical and radiological findings, a definitive diagnosis of specific IIP can be established, allowing the optimal clinical-therapeutic management of the patient.

A nomogram based on clinical factors and CT radiomics for predicting anti-MDA5+ DM complicated by RP-ILD

OBJECTIVES

Anti-melanoma differentiation-associated gene 5 antibody-positive (anti-MDA5+) DM complicated by rapidly progressive interstitial lung disease (RP-ILD) has a high incidence and poor prognosis. The objective of this study was to establish a model for the prediction and early diagnosis of anti-MDA5+ DM-associated RP-ILD based on clinical manifestations and imaging features.

METHODS

A total of 103 patients with anti-MDA5+ DM were included. The patients were randomly split into training and testing sets of 72 and 31 patients, respectively. After image analysis, we collected clinical, imaging and radiomics features from each patient. Feature selection was performed first with the minimum redundancy and maximum relevance algorithm and then with the best subset selection method. The final remaining features comprised the radscore. A clinical model and imaging model were then constructed with the selected independent risk factors for the prediction of non-RP-ILD and RP-ILD. We also combined these models in different ways and compared their predictive abilities. A nomogram was also established. The predictive performances of the models were assessed based on receiver operating characteristics curves, calibration curves, discriminability and clinical utility.

RESULTS

The analyses showed that two clinical factors, dyspnoea (P = 0.000) and duration of illness in months (P = 0.001), and three radiomics features (P = 0.001, 0.044 and 0.008, separately) were independent predictors of non-RP-ILD and RP-ILD. However, no imaging features were significantly different between the two groups. The radiomics model built with the three radiomics features performed worse than the clinical model and showed areas under the curve (AUCs) of 0.805 and 0.754 in the training and test sets, respectively. The clinical model demonstrated a good predictive ability for RP-ILD in MDA5+ DM patients, with an AUC, sensitivity, specificity and accuracy of 0.954, 0.931, 0.837 and 0.847 in the training set and 0.890, 0.875, 0.800 and 0.774 in the testing set, respectively. The combination model built with clinical and radiomics features performed slightly better than the clinical model, with an AUC, sensitivity, specificity and accuracy of 0.994, 0.966, 0.977 and 0.931 in the training set and 0.890, 0.812, 1.000 and 0.839 in the testing set, respectively. The calibration curve and decision curve analyses showed satisfactory consistency and clinical utility of the nomogram.

CONCLUSION

Our results suggest that the combination model built with clinical and radiomics features could reliably predict the occurrence of RP-ILD in MDA5+ DM patients.

CCL18, CHI3L1, ANG2, IL-6 systemic levels are associated with the extent of lung damage and radiomic features in SARS-CoV-2 infection

OBJECTIVE AND DESIGN

We aimed to identify cytokines whose concentrations are related to lung damage, radiomic features, and clinical outcomes in COVID-19 patients.

MATERIAL OR SUBJECTS

Two hundred twenty-six patients with SARS-CoV-2 infection and chest computed tomography (CT) images were enrolled.

METHODS

CCL18, CHI3L1/YKL-40, GAL3, ANG2, IP-10, IL-10, TNFα, IL-6, soluble gp130, soluble IL-6R were quantified in plasma samples using Luminex assays. The Mann-Whitney U test, the Kruskal-Wallis test, correlation and regression analyses were performed. Mediation analyses were used to investigate the possible causal relationships between cytokines, lung damage, and outcomes. AVIEW lung cancer screening software, pyradiomics, and XGBoost classifier were used for radiomic feature analyses.

RESULTS

CCL18, CHI3L1, and ANG2 systemic levels mainly reflected the extent of lung injury. Increased levels of every cytokine, but particularly of IL-6, were associated with the three outcomes: hospitalization, mechanical ventilation, and death. Soluble IL-6R showed a slight protective effect on death. The effect of age on COVID-19 outcomes was partially mediated by cytokine levels, while CT scores considerably mediated the effect of cytokine levels on outcomes. Radiomic-feature-based models confirmed the association between lung imaging characteristics and CCL18 and CHI3L1.

CONCLUSION

Data suggest a causal link between cytokines (risk factor), lung damage (mediator), and COVID-19 outcomes.

Baseline whole-lung CT features deriving from deep learning and radiomics: prediction of benign and malignant pulmonary ground-glass nodules

Objective

To develop and validate the model for predicting benign and malignant ground-glass nodules (GGNs) based on the whole-lung baseline CT features deriving from deep learning and radiomics.

Methods

This retrospective study included 385 GGNs from 3 hospitals, confirmed by pathology. We used 239 GGNs from Hospital 1 as the training and internal validation set; 115 and 31 GGNs from Hospital 2 and Hospital 3 as the external test sets 1 and 2, respectively. An additional 32 stable GGNs from Hospital 3 with more than five years of follow-up were used as the external test set 3. We evaluated clinical and morphological features of GGNs at baseline chest CT and extracted the whole-lung radiomics features simultaneously. Besides, baseline whole-lung CT image features are further assisted and extracted using the convolutional neural network. We used the back-propagation neural network to construct five prediction models based on different collocations of the features used for training. The area under the receiver operator characteristic curve (AUC) was used to compare the prediction performance among the five models. The Delong test was used to compare the differences in AUC between models pairwise.

Results

The model integrated clinical-morphological features, whole-lung radiomic features, and whole-lung image features (CMRI) performed best among the five models, and achieved the highest AUC in the internal validation set, external test set 1, and external test set 2, which were 0.886 (95% CI: 0.841-0.921), 0.830 (95%CI: 0.749-0.893) and 0.879 (95%CI: 0.712-0.968), respectively. In the above three sets, the differences in AUC between the CMRI model and other models were significant (all P < 0.05). Moreover, the accuracy of the CMRI model in the external test set 3 was 96.88%.

Conclusion

The baseline whole-lung CT features were feasible to predict the benign and malignant of GGNs, which is helpful for more refined management of GGNs.

Development of a radiomics nomogram to predict the treatment resistance of Chinese MPO-AAV patients with lung involvement: a two-center study

Background

Previous studies from our group and other investigators have shown that lung involvement is one of the independent predictors for treatment resistance in patients with myeloperoxidase (MPO)-anti-neutrophil cytoplasmic antibody (ANCA)-associated vasculitis (MPO-AAV). However, it is unclear which image features of lung involvement can predict the therapeutic response in MPO-AAV patients, which is vital in decision-making for these patients. Our aim was to develop and validate a radiomics nomogram to predict treatment resistance of Chinese MPO-AAV patients based on low-dose multiple slices computed tomography (MSCT) of the involved lung with cohorts from two centers.

Methods

A total of 151 MPO-AAV patients with lung involvement (MPO-AAV-LI) from two centers were enrolled. Two different models (Model 1: radiomics signature; Model 2: radiomics nomogram) were built based on the clinical and MSCT data to predict the treatment resistance of MPO-AAV with lung involvement in training and test cohorts. The performance of the models was assessed using the area under the curve (AUC). The better model was further validated. A nomogram was constructed and evaluated by DCA and calibration curves, which further tested in all enrolled data and compared with the other model.

Results

Model 2 had a higher predicting ability than Model 1 both in training (AUC: 0.948 vs. 0.824; p = 0.039) and test cohorts (AUC: 0.913 vs. 0.898; p = 0.043). As a better model, Model 2 obtained an excellent predictive performance (AUC: 0.929; 95% CI: 0.827-1.000) in the validation cohort. The DCA curve demonstrated that Model 2 was clinically feasible. The calibration curves of Model 2 closely aligned with the true treatment resistance rate in the training (p = 0.28) and test sets (p = 0.70). In addition, the predictive performance of Model 2 (AUC: 0.929; 95% CI: 0.875-0.964) was superior to Model 1 (AUC: 0.862; 95% CI: 0.796-0.913) and serum creatinine (AUC: 0.867; 95% CI: 0.802-0.917) in all patients (all p< 0.05).

Conclusion

The radiomics nomogram (Model 2) is a useful, non-invasive tool for predicting the treatment resistance of MPO-AAV patients with lung involvement, which might aid in individualizing treatment decisions.

Artificial-intelligence-based computed tomography histogram analysis predicting tumor invasiveness of lung adenocarcinomas manifesting as radiological part-solid nodules

Background

Tumor invasiveness plays a key role in determining surgical strategy and patient prognosis in clinical practice. The study aimed to explore artificial-intelligence-based computed tomography (CT) histogram indicators significantly related to the invasion status of lung adenocarcinoma appearing as part-solid nodules (PSNs), and to construct radiomics models for prediction of tumor invasiveness.

Methods

We identified surgically resected lung adenocarcinomas manifesting as PSNs in Peking University People's Hospital from January 2014 to October 2019. Tumors were categorized as adenocarcinoma in situ (AIS), minimally invasive adenocarcinoma (MIA), and invasive adenocarcinoma (IAC) by comprehensive pathological assessment. The whole cohort was randomly assigned into a training (70%, n=832) and a validation cohort (30%, n=356) to establish and validate the prediction model. An artificial-intelligence-based algorithm (InferRead CT Lung) was applied to extract CT histogram parameters for each pulmonary nodule. For feature selection, multivariate regression models were built to identify factors associated with tumor invasiveness. Logistic regression classifier was used for radiomics model building. The predictive performance of the model was then evaluated by ROC and calibration curves.

Results

In total, 299 AIS/MIAs and 889 IACs were included. In the training cohort, multivariate logistic regression analysis demonstrated that age [odds ratio (OR), 1.020; 95% CI, 1.004-1.037; p=0.017], smoking history (OR, 1.846; 95% CI, 1.058-3.221; p=0.031), solid mean density (OR, 1.014; 95% CI, 1.004-1.024; p=0.008], solid volume (OR, 5.858; 95% CI, 1.259-27.247; p = 0.037), pleural retraction sign (OR, 3.179; 95% CI, 1.057-9.559; p = 0.039), variance (OR, 0.570; 95% CI, 0.399-0.813; p=0.002), and entropy (OR, 4.606; 95% CI, 2.750-7.717; p<0.001) were independent predictors for IAC. The areas under the curve (AUCs) in the training and validation cohorts indicated a better discriminative ability of the histogram model (AUC=0.892) compared with the clinical model (AUC=0.852) and integrated model (AUC=0.886).

Conclusion

We developed an AI-based histogram model, which could reliably predict tumor invasiveness in lung adenocarcinoma manifesting as PSNs. This finding would provide promising value in guiding the precision management of PSNs in the daily practice.

Effects of Automatic Deep-Learning-Based Lung Analysis on Quantification of Interstitial Lung Disease: Correlation with Pulmonary Function Test Results and Prognosis

We investigated the feasibility of a new deep-learning (DL)-based lung analysis method for the evaluation of interstitial lung disease (ILD) by comparing it with evaluation using the traditional computer-aided diagnosis (CAD) system and patients’ clinical outcomes. We prospectively included 104 patients (84 with and 20 without ILD). An expert radiologist defined regions of interest in the typical areas of normal, ground-glass opacity, consolidation, consolidation with fibrosis (traction bronchiectasis), honeycombing, reticulation, traction bronchiectasis, and emphysema, and compared them with the CAD and DL-based analysis results. Next, we measured the extent of ILD lesions with the CAD and DL-based analysis and compared them. Finally, we compared the lesion extent on computed tomography (CT) images, as measured with the DL-based analysis, with pulmonary function tests results and patients’ overall survival. Pearson’s correlation analysis revealed a significant correlation between DL-based analysis and CAD results. Forced vital capacity was significantly correlated with DL-based analysis (r = 0.789, p < 0.001 for normal lung volume and r = −0.316, p = 0.001 for consolidation with fibrosis volume). Consolidation with fibrosis measured using DL-based analysis was independently associated with poor survival. The lesion extent measured using DL-based analysis showed a negative correlation with the pulmonary function test results and prognosis.

Surgomics: personalized prediction of morbidity, mortality and long-term outcome in surgery using machine learning on multimodal data

BACKGROUND

Personalized medicine requires the integration and analysis of vast amounts of patient data to realize individualized care. With Surgomics, we aim to facilitate personalized therapy recommendations in surgery by integration of intraoperative surgical data and their analysis with machine learning methods to leverage the potential of this data in analogy to Radiomics and Genomics.

METHODS

We defined Surgomics as the entirety of surgomic features that are process characteristics of a surgical procedure automatically derived from multimodal intraoperative data to quantify processes in the operating room. In a multidisciplinary team we discussed potential data sources like endoscopic videos, vital sign monitoring, medical devices and instruments and respective surgomic features. Subsequently, an online questionnaire was sent to experts from surgery and (computer) science at multiple centers for rating the features' clinical relevance and technical feasibility.

RESULTS

In total, 52 surgomic features were identified and assigned to eight feature categories. Based on the expert survey (n = 66 participants) the feature category with the highest clinical relevance as rated by surgeons was "surgical skill and quality of performance" for morbidity and mortality (9.0 ± 1.3 on a numerical rating scale from 1 to 10) as well as for long-term (oncological) outcome (8.2 ± 1.8). The feature category with the highest feasibility to be automatically extracted as rated by (computer) scientists was "Instrument" (8.5 ± 1.7). Among the surgomic features ranked as most relevant in their respective category were "intraoperative adverse events", "action performed with instruments", "vital sign monitoring", and "difficulty of surgery".

CONCLUSION

Surgomics is a promising concept for the analysis of intraoperative data. Surgomics may be used together with preoperative features from clinical data and Radiomics to predict postoperative morbidity, mortality and long-term outcome, as well as to provide tailored feedback for surgeons.